P4, Af %S APe A i'>-' V A N N~~ s A p ,~ '<' #,1 F ' ), , ? (F > ^$ 5 se;), ,- 9 8 f 'F fV'4" 's:f>)'ts'?tS %e F AiF'. F)o%5 h, 'i), 'F F'FC FFg C AA A.. A, A ..;:.',AAA-1 ..'[ .:.:''zAr k'* -.', WB-P AGW-M AJPA' AMNH -B-C INAI OI3CA FLL PMP ICAk k, A,r IJ AL''r' ' :!e ;' A.. t '- :,L + i i -, ' .. , ..A .,I -B t :.Y X A I E F - 1.j I ;'~~ ~ -a ''' ' ~ + r -M; -" 'o. B E'Jx'-" P1.W r . . I , .- P ~'V ' Y r- D; S AP' 4 l . 01 : ,, ,, , w , ,, -A1 0'-. GKNA 0:f:t UO-P/l:6 A AR5, UP, ' , AP : u-'' FL, v- :. ; 'F63 -P' 44 ! ,.' N AppRE tONS D JAnhr; p. '; VAthropol ogle. -mericn pologist< Ameca Anhthropologica As ;boh ation vIemfra;. 0_ Arohir, fi"' Antiit nlnni American tn gical S ations. Ant#hro,9poilogishei' es,,llschat in Wie ,it iw'ge;. A r :4ct n Joual, of P opolg Amen,, a. squm of 'NatualHistory- A, Anthrop thologic l-6Paprals. fe utWletin. iV:i- emoirs ; : ;-! :0- 0 lV[emoirs, Anthopo ogieal Sees ,- J :;: eM?Ow. Jesup ;Expedt . D; D-n ;-E-- : Bureau e-f American Ethnao1ogyg-, A :- ;tontributions Xto North American Zhnolo7 yt t 0- oIumbia U;TJersi, O iont Xuin to Athr6pology.0i0-; Field `Juseum of Atl istory- : : moix. - S; Pub t i,,,;,',Authr'opoloi l' ies. -:- ornationales Arhiv f;: :t grahie Pocd iVong.e", o d? IF 0 Om ) ,, . .Internatilonal; J~ouvK,n..of AinFican Llgist,ics.', 'i'v ;3, j.,; .f,,> Jurnal' of Ameri Fo- r Lire, '-,Jornal,of the Rl hopologial Virstitute. . iLMueum of t.e American India, 'Heye Foundation- Oontibutions. i Indian L M Notes an ographs. 'P ebo Museum (of ,,arvar4 Unverst,)- Papers., Public' uYluseu (Of theCity)' ofliw ileti '. Soc6tE:des A6-imcanite e Paris; J . &ni~hsonma Institution--, Miscelaneou Colletions 'Univbrity of co-Ifornmis Publications ulAm can' Areha'eology jU,ver.i en* yl. ia*(Un' rers'0 e's 'y)'K'sem tb>pt ;loga Puicaons* xi Uf ieed States N.til Muse ' '' ' 6r yL 0 s, O', ,p'- utf. Tub,,40<4 1;, th'' 'W ~'''0'' ';r ' ;' a U&bhversio Wasin Publii In Anh-p-o; ,:;Zeltschrift f-ur:thxkologie;.' 'f' s..>- }.'''-v' University of California Publications in AMERICAN ARCHAEOLOGY AND ETHNOLOGY VOLUME XXX 1930 EDITORS A. L. KROEBER ROBERT H. LOWIE KLAMATH ETHNOGRAPHY BY LESLIE SPIER UNIVERSITY OF CALIFORNIA PRESS BERKELEY, CALIFORNIA 1930 UNIVERSITY OF CALIFORNIA PUBLICATIONS IN AMERICAN ARCHAEOLOGY AND ETHNOLOGY Volume 30, pp. x + 1-338, 22 figures in text Issued December 6, 1930 UNIVERSiTY OF CALIFORNIA PRESS BERKELoEY, CALIFORNIA CAMBRIDGE UNIVERSITY PRESS LONDON, ENGLAND ABBREVIATIONS USED IN THE UNIVERSITY OF CALIFORNIA PUBLICATIONS IN AMERICAN ARCHAEOLOGY AND ETHNOLOGY A I'A AA AAA-M ArA AES-P AGW-M AJPA AMNH -AP -B -M -MA -MJ BAE -B -R CNAE CU-CA FL FMNH -M -PAS IAE ICA IJAL JAFL JRAI MAIHF -C -IN -INM PM -M Anthropos. L'Anthropologie. American Anthropologist. American Anthropological Association, Memoirs. Archiv fur Anthropologie. American Ethnological Society, Publications. Anthropologische Gesellschaft in Wien, Mitteilungen. American Journal of Physical Anthropology. American Museum of Natural History- Anthropological Papers. Bulletin. Memoirs. Memoirs, Anthropological Series. Memoirs, Jesup Expedition. Bureau of American Ethnology- Bulletins. (Annual) Reports. Contributions to North American Ethnology. Columbia University, Contributions to Anthropology. Folk-Lore. Field Museum of Natural History- Memoirs. Publications, Anthropological Series. Internationales Archiv fUir Ethnographie. International Congress of Americanists (Comptes Rendus, Proceedings). International Journal of American Linguistics. Journal of American Folk-Lore. Journal of the Royal Anthropological Institute. Museum of the American Indian, Heye Foundation- Contributions. Indian Notes. Indian Notes and Monographs. Peabody Museum (of Harvard University)- Memoirs. -P Papers. -R Reports. PMM-B Public Museum (of the City) of Milwaukee, Bulletin SAP-J Societ6 des Am6ricanistes de Paris, Journal. SI Smithsonian Institution- AR Annual Reports. CK Contributions to Knowledge. -MC Miscellaneous Collections. UC-PAAE University of California, Publications in American Archaeology and Ethnology. UPM-AP University of Pennsylvania (University) Museum, Anthropological Publications. USNM United States National Museum- -R Reports. -P Proceedings. UW-PA University of Washington, Publications in Anthropology. ZE Zeitschrift fur Ethnologie. CONTENTS PAGE Introduction ............................................................ vii I. Klamath social life .............................................................1 The Klamath and their territory .............................................................1 Tribal relations and population ............................................................ 1 Early contacts with the whites ............................................................ 5 Habitat .............................................................8 Settlements ............................................................ 10 Tribal subdivisions ............................................................ 21 W arfare and feuds ............................................................ 24 Chieftainship ............................................................ 35 Slaves, trade, and wealth ............................................................ 39 M arriage ............................................................ 43 Transvestites ............................................................ 51 House groups ............................................................ 53 Birth ............................................................ 55 Names ............................................................ 59 Status and kinship terms ............................................................ 61 Age classification terms ............................................................ 61 Kinship terms ............................................................ 62 Puberty and n ienstrual customs ............................................................ 68 Death and mourning ............................................................ 71 Games ............................................................ 76 Smoking ............................................................ 86 Musical instruments ............................................................ 88 Social dances ............................................................ 90 Etiquette ............................................................ 91 II. Klamath shamanism ............................................................ 93 Introduction ............................................................ 93 The power quest ............................................................. 94 Spirits, souls, and ghosts ............................................................ 100 Shamanism ............................................................ 107 The shamanistic performance ............................................................ 112 W eather control and clairvoyance ............................................................ 118 Curing ............................................................ 122 The songs ............................................................ 131 Miscellaneous beliefs ............................................................ 138 III. Economic life and material culture ........................................ .................... 144 Division of labor ............................................................ 144 Food and its acquisition ........................: 145 Fishing ....................... 147 Hunting ......................... 155 Seed and root gathering ....................... 160 Canoes ....................... 169 Various manufactures and tools ....................... 171 Basketry ....................... 177 Weapons ....................... 193 Houses and houselife ...................................... : 197 Sweat-lodges ..................................... 205 Clothing and adornment ..................................... 207 Time reckoning, directions, and numeration ..................................... 217 IV. The Klamath and western culture ..................................... 224 The place of Klamath culture ..................................... 224 PAGE The shamanistic complex .............................................. . 239 Spirits ............................................... 245 Inheritance of spirit power .............................................. . 247 T he pow er quest .............................................. 249 The period of the quest . ............................................. . 256 The conduct of the quest .. ............................................. 257 The novice's performance .............................................. . 259 The winter performance .............................................. . 266 The shaman's assistant .............................................. . 267 Accessory elements . . ...................................... 269 Clairvoyance . .............................................. 273 Various beliefs . .............................................. 274 Summary .............................................. .. 275 The marriage complex . . ...................................... 279 Purchase, rank, and cxogamy . . ............................... 283 The amount of the bride price . . ............................... 293 Marriage visits . ............................................. . 296 Family compacts . ............................................. . 299 Post-nuptial residence .............................................. . 303 Summary .............................................. .. 304 Social stratification . ............................................. . 307 Basis of chieftainship .............................................. . 308 Chief's functions . ............................................. . 311 Speaker . .............................................. 312 Slavery . .............................................. 313 Summary ................................................ 314 Girls' puberty rites and women's observances . ................................ 314 Appendix ............................................... 326 The M odoc novice dance .............................................. . 326 Bibliography ............................................... 329 Index ............................................... 335 FIGURES IN TEXT 1. Map of Klamath territory .............................................. Facing page 8 2. Settlements in northern Klamath territory .............................................. 12 3. Settlements in southern Klamath territory .............................................. 17 4. Plan of an earth-lodge .............................................. 54 5. Cradle boards .............................................. 58 6. Head-scratchers .............................................. 70 7. Mourners' sweat-lodge near Algoma .............................................. 74 8. String figures .............................................. 84 9. Pictograph on the eastern shore of Klamath lake .............................................. 142 10. Tail bone of a sucker used as a minnow hook ................................... .......... 154 11. Pestles and mortars .176 12. Mode of commencing twined basket .179 13. Base of a twined basket .180 14. Method of finishing the edge of a twined basket .182 15. Modified simple twine stitch of basket hats .183 16. Methods of adding warps in a conical carrying basket .185 17. Basket designs .192 18. Plan and section of earth-lodge showing construction .199 19. Named parts of the house .201 20. Winter sweat-lodge .205 21. Patterns of skin moccasins 209 22. Distribution of the quest and dream as sources of spirit power in western North America .250 INTRODUCTION The present study is based on information gathered among the Klamath for the University of California during two visits, each for a month, in 1925 and 1926. Klamath material culture was partly known from Bar.rett's description of his collection made twenty years ago. The immediate need appeared to be a study of their religion and social organization. Gatschet's account of The Klamath Indians published in 1890 contains many data on shamans' songs and allied topics, but it does not provide any systematic notion of their meaning. The Klamath still live in their old home in southern Oregon, but their ancient culture exists only in their memories, save as the older people keep up domestic pursuits. Yet even middle-aged people have a very good knowledge of it. My elderly informants had all par- ticipated in the old life in their youth. These were Nancy Phillips, Pat Kane, and Coley Ball, all three about seventy, and Thomas Lang, about fifty-five. With the exception of Mrs. Phillips they gave inform- ation in English: in her case (the most valuable material) Mr. Lang and his son Delford acted as interpreters. The social life of the Klamath is so much a thing of the past that only its formal aspects can now be recovered. An intimate picture is impossible. But to avoid a stilted expression I have written the account in the historical present. The description of material culture and economic life is not complete. It is still partly extant among the Klamath and well worth investigating. In a final section I have attempted to place the Klamath in a setting of western cultures as a whole. This has been difficult because of the lack of comparative data from much of the northwest; further, only a fraction of the published accounts is available to me at this writing (in Norman, Oklahoma). The phonetic scheme is the simpler system of the Phonetic Tran- scription of Indian Languages (Smithsonian Misc. Coll., 66, no. 6, 1916), modified as: a as in Allen, pronounced more like Ellen. -1 as in sing. E indefinite, obscure vowel; very short duration. k very far back. it cerebral; tongue tip probably turned back as Gatschet states (the sequen. e kr also occurs). g cerebral; may be the same sound. A number of observations were made by my Klamath informants on the cultural habits of MIolala, Northern Paiute, Upland Takelma, Modoc, and Achomawi-Atsugewi. I have included these as they demonstrate the sort of resemblances and differences of interest to my informants, and also because so few data are in print concerning these peoples. They will be found on the pages indicated: Molala, pp. 4, 9, 24, 41, 59, 107, 158, 194, 214, 215. Northern Paiute, pp. 58, 79, 81, 86, 90, 105, 158, 188, 190. Upland Takelma, pp. 4, 24, 28, 29, 32, 35, 58, 59, 71, 177, 217. Modoe, pp. 41, 42, 58, 81, 120, 158, 164, 175, 178, 190. Achomawi-Atsugewi, pp. 59, 79, 190, 194. KLAMATH ETHNOGRAPHY BY LESLIE SPIER I. KLAMATH SOCIAL LIFE THE KLAMATH AND THEIR TERRITORY Tribal Relations and Population The Klamath are the northern of the two major dialectic groups of the Lutuami linguistic stock, the other, the ifodoc. Dialectically there is little difference, but of the two the Klamath have the richer, more specialized culture. The Klamath recognize a number of local autonomous subdivisions but consider themselves a social unit. They call themselves ma'klaks, men;' all other tribes are known by directional or locality names. The origin of the designation Klamath is obscure; incidentally they pro- nounce this klTamet, much as the early travelers wrote it. It is not a native name and evidently differs so much from native forms that none of my informants even attempted an etymology. Curiously enough, all the Klamath reserve the name for the Klamath marsh- Williamson river subdivision (a'ukckni); others are "Klamath" only by courtesy. The reason may be that this is not only the numerically strongest subdivision, but the reservation on which they are now con- centrated is little more than the old 'aukckni territory. Gatschet suggested that the name is a Yurok word. Its earliest use is perhaps that of Ogden in 1826. He uses the name "Clammitte" before reaching the country, possibly learning it among the Columbia river tribes. Yet members of his party had visited the Klamath country the year previous and may have learned it there. Elliott suggests an origin in "French Clair-metis, meaning a light mist or cloud. '2 Incidentally there is now no good reason for ' Gatschet states that the form ma'klaks (ma'klaks in my orthography) means one of themselves, maklaks, any other Indian (Gatschet, 1, xxxiii). My informants denied the existenee of any such distinction, and indeed of the first form. 2 Elliott, Ogden Journals, 202. 2 University of California Publication, in, Am. Arch. and Ethn. [Vol. 30 using the name for others on the Klamath river, whatever the local usage in northern California. It will save the need for cumbersome titles like "the Klamath Lake Indians" if the name is reserved for the people described in this paper. The Klamath are called dak'-ts!aam-al-a'e or dak'-ts!aew-an-a'!, "those above the lakes," by the Takelma of middle Rogue river3 and wols by the Hanis or Upland Takelma of the Medford-Ashland dis- trict.4 The Kalapuya call them Athlameth,5 which Kroeber suggests as a possible origin of Klamath.6 Gatschet states that the Warm Spring and other Sahaptin Indians possess a generic name for all the Indians living upon this reservation and its vicinity: Aigspaluma, abbr. Aigspalo, Aikspalu, people of the chipmunks, from the innumerable rodents peopling that pine covered district. This term comprises Snake, Payute, and Modoe Indians, as well as the Klamath Lake people.7 This appeals to me much as a folk etymology from a hybrid "people of a'ukei' (Klamath marsh)." The same authority gives as Shasta designations makai'tserk and au'ksiwash. This Kroeber reproduces in the form auk-" siwash," as if compounded from the Chinook jargon word for Indian, but the first part also suggests a'ukei. To the Pit river Indians (Achomawi and Atsugewi) they are ala'mmimakt ish, from ala'mmig, upper Klamath lake.8 Curiously enough none of my informants knew what the Northern Paiute call them; Gatschet gives sa'yi, which is suspiciously like the name they give the Paiute, sat. The Klamath make little distinction between tribes at a distance, naming them indiscriminately by direction. Thus the Modoc are mo'ad6k ma'klaks, south people, or mo'adokkni, southerners, the Pit river tribes, moa'tuwas, all derived from mo'as, south. At least one group of Modoc, gu'mbatwas, is distinguished by dialect.9 No dis- tinction is made between the several tribes of the upper Pit river (Achomawi, Atsugewi, etc.). A southern people called k.ad'6'kiwos is known but probably only since the white advent. They lived thirty or forty miles west of Mount Shasta, near the Shasta, and 3 Sapir, Takelma Texts, 257; Notes on the Takelma, 255. 4 This is Klamath information. By this word (wAlx, meaning enemies, Upland Takelma, or Shasta in Takelma proper) I identify the Hanis or Wa!liimskni with Sapir 's Upland Takelma (Notes, 252). 5 Gatschet, 1:xxxiv. 6 Handbook, 319. 7 Op cit., 1:xxXiii. 8 See also Merriam, 48. 9 "The Kidmbatuash lived on southeastern end of Tule (or Rhett) Lake, California." (Gatschet, 1:14.) Spier: Kliamath. Ethnography hunted on the Klamath river above them. This seems to be the Shasta name for the Wintun of Trinity river, hatukwiwa or hatukeyu, which agrees with Gatschet's ascription, Sacramento Valley Indians.10 The Shasta themselves are known only as ca.'sti. This might be taken as the long sought origin of the name Shasta, was there not the suspicion that its use may be only recent, displacing another and older applica- tion of "southerners" to them. Their principal seat near Yreka is c'stiknia, Shasta place. Similarly all northern tribes are ya'mokkni, northerners (from ya'mot, north). By exclusion of the Indians at Warm Springs and the Dalles, however, this has come to mean the half Plains-like tribes farther up the Columbia. The Warm Springs people and those at Celilo near the Dalles are watA'nkkni, a form which indicates that they are named for their locality. Gatschet gives wiiitiing'sham and l6kuashtkni.11 The linked tribes at the Dalles, Wasco and Wishram, are the o'mpk-ankni, those of the locality 6'mpkan, which by descrip- tion must be in the neighborhood of Wasco, Oregon. This suggests that the word is from a'mbE, wa.ter (with a common locative suffix, cf. ca'stik'an, t6okan, service berry tract), hence is a translation into Klamath of the name Wasco. This, according to my own Wishra.m information, means "like a pan" (wask'o) and refers to a spring at or near the present town of that name. Since I completed field work with the Klamath, there has appeared Teit 's paper defining the original occupation and the extensive tribal movements in northern Oregon and eastern Washington.12 Had this been available, it would undoubtedly have been possible to render more exactly the meanings of ya'm6kkni and wata'nkkni. It appears from Teit 's conclusive evidence that at the opening of the eighteenth century all of eastern Washington was Salish territory as far as the Dalles. South of the Columbia was a band of Waiilatpuan territory, with the Molala on the middle Deschutes and in the mountains to the west, and with the Cayuse eastward of them nearly to the Snake river. South of these again were the Sahaptin peoples with the Cascade range and the Klamath on their west, the Nez Perce to the northeast, and the Snake groups on their other boundaries. A series of movements beginning in the eighteenth century resulted in the withdrawal of the Salish to north of Priest Rapids because of the occupation of their territory by Sahaptins and the Cayuse. The Molala also withdrew into 10 Kroeber, Handbook, 355; Gatschet, 1:393. 11 Op. cit., 1:lvi, 18. My informants ascribed to them painted tipis (so'kluiks) of buffalo skin, long dance lodges made on a foundation of tipis, but no earth- lodges. 12 The Middle Columbia Salish. 1930] 3 4 University of California Pubication in Am. Aroh. and Ethn. [Vol. 30 the mountains and westward even beyond the Willamette valley, the Sahaptin Tyighpam settling in part of their country and the Tenino joining the Wasco. The immediate cause of these movements seems to have been the pressure of Snake raids, at their height about 1800-30. This is consonant with the very early acquisition of the horse by the Snake and their known warlike relations with the Plains tribes, which Wissler has demonstrated.13 This would explain the wide expanse of vacant country between the Klamath and the Dalles or at least as far as Warm Springs. In addition, Teit writes: "For a number of years the south side of the Columbia between Wallawalla and The Dalles was practically devoid of inhabitants" for fear of Snake raids. My informant, Coley Ball, stated that about 1865-70 there were no settlements between Klamath marsh and the Wasco villages at the Dalles. There were no Molala east of the Cascades. How recently the mixed group of Wasco-Tyighpam- Tenino occupied the Warm Springs country, I do not know. This would also explain the anomalous position of a Molala group on the very headwaters of the Rogue river where presumably they represent the final southerly drift along the western crest of the Caseade range. West of the Cascades are Upland Takelma in the region from Ashland to Table Rock. They are wa'liumskni; Mount Pitt is wa'lims- yaina (yaina, mountain); the name then is regional, not directional. They are called ha'nis quite as frequently, but this name, which is properly that of the Coos group at Coos bay, has evidently been extended to them and others since the remnants of these tribes were collected at Siletz reservation.14 In the extreme upper Rogue valley are Molala, tco'ka'nkni, those of the service berry (tcok) patch.Y Gatschet gives mokai or moke as the Klamath name for the Kalapuya.16 Beyond these groups none of the western Indians are known. Those of the lower Rogue river are now called soljo'k.kni, those of the salt water, but this is from the Chinook jargon, saltehuck.17 All the Shoshoneans to the east are called sat; no distinctions a-re made. The actual neighbors are those groups known to us variously as Northern Paiute (Paviotso) and Snake. The Snake proper and Bannock may not have been in Oregon prior to the acquisition of the horse. 13 The Influence of the Horse, 13. 14 Frachtenberg, Coos and Siuslawan (Lower Umpqua), 306, 441. One inform- ant told me that the last of the waluimskni, three old women, are still to be found at Siletz. 16 My informants insisted this name was given because the Molala eat service berries. On the other hand one gave me as a Molala tale an incident in which the culture hero makes people of service berry bushes. One way or the other this looks like folk etymology. Their winter town was bu'kstubui'ks, a little below Prospect. 16 Op. cit., 2:217. 17 See my Tribal Distribution in Southwestern Oregon. Spier: Kl4math Ethnography The Klamath numbered 3000 in the historic period, the Modoc not so many. This is the testimony of Tom Lang, yet the number is much too large. Palmer'8 put their number at 415 in thirteen villages in 1854, but his list of villages indicates that he counted only the a'ukckni (Klamath marsh) division. This would be about eighty-three families, a reasonable average of six houses to a settlement. At the time, how- ever, the Klamath were said to be considerably reduced, and I was told of twenty-nine or more aukckni settlements dating from this period, and said to be all occupied at the time. Yet, as no great epidemic was reported, it is doubtful that the population had been diminished by half. Possibly the occupation of the sites was more seasonal than I was told: a movement to Klamath marsh in summer, returning to the warmer Williamson valley in winter. Assuming a partial reduc- tion and some fluctuating occupation, we may arbitrarily set the a ukckni at half as many again, say 600. The 'aukckni settlements numbered half of the total and my informants' impression was that in population they were as many as all the other divisions together. This would mean that the Klamath as a whole numbered only 1200. Palmer's figure would give a mere 800, and the evidence of the settle- ments, a possible 2000. Our twelve hundred is not far from that of Kroeber, who set the combined Klamath and Modoc population at 2000, two-thirds of them being Klamath."' Mooney's estimate of 800 seems low.20 Early Contacts with the Whites Contacts with the whites date back to 1825 or somewhat earlier. In 1826-27 Peter Ogden spent two months among the Klamath.21 His first meeting with them seems to have been on Klamath marsh. Wednesday, 30th Nov. [1826]. Course south to Clammitt River [Klamath marsh?] 25 miles from River of the Falls [Deschutes]. Mr. McKay proceeded ahead to an Indian village distant 3 miles. It was composed of 20 tents built on the water approachable only by canoes, the tents built of large logs shaped like block houses the foundation stone or gravel made solid by piles sunk 6 ft. deep. Their tents are constantly guarded. They regretted we had opened a communication from the mountains. They said "The Nez Perc6s have made different attempts to reach our village but could not succeed. Even last sum- mer we discovered a war party of Cayuse and Nez Perc6s in search of us; but they did not find us. Now they will have yr. road to follow. We have no fire arms. Still we fear them not. 18 Quoted by Gatschet, 1:lviii. 20 Mooney, Aboriginal Population, 18. 19 Kroeber, Handbook, 320. 21 Elliott, Ogden Journals, 210. 1930] 5 6 University of California Publications in, An. Arch. an-d Ethn. [Vol. 30 They are well provided with bows and arrows. They have only one horse. Snow is so deep, horses perish for want of food. In winter, they live on roots. In summer on antelope and fish. December Ist. 30 Clammite Indians paid us a visit: fine men in good condition, but wretchedly clad. They say the river to the ocean is far distant and beaver they do not know. They say the Indians become more numerous as we advance to the ocean. . . . Tuesday 6th. Reached Indian village of five huts, hut large size square made of earth flat on top the door at the top a defence against arrows but not balls. 200 of them collected about our eamp and traded 4 days. The 2 chiefs delivered traps lost by Mr. McDonald last year with 8 beaver. This is much in favor of their honesty. On our march this day, we passed the camps from wh. Mr. McDonald turned back last year and are consequently strangers to the coun- try in advance. Thursday 8th. About 300 Indians around our camp. We advanced 6 miles south following the river south. I estimate the Clammitte nation 250 men. Monday 12th. Reached the lake [Klamath lake?] 1Y2 X 15 miles well wooded with maple and hazel; course S. The second visit on record is that of Fremont, who reached the northern shore of Klamath marsh on December 10, 1843. He saw smoke at what appeared to be the middle of the marsh and along the opposite shore. An unoccupied northern shore agrees with my own information. The following day he visited the village "in the middle of the lake (or savannah)," which appears to be one of those at the outlet of the marsh. The huts were grouped together on the bank of the [Klamath] river, which, from being spread out in a shallow marsh at the upper end of the lake, was collected here into a single stream. They were large round huts, perhaps 20 feet in diameter, with rounded tops, on which was the door by which they descended into the interior. Within, they were supported by posts and beams. Almost like plants, these people seem to have adapted themselves to the soil, and to growing on what the immediate locality afforded. Their only sub- sistence at this time appeared to be a small fish, great quantities of which, that had been smoked and dried, were suspended on strings about the lodge. Heaps of straw were lying around; and their residence in the midst of grass and rushes had taught them a peculiar skill in converting this material to useful purposes. Their shoes were made of straw or grass, which seemed well adapted for a snowy country; and the women wore on their head a closely woven basket, which made a very good cap. Among other things, were parti-colored mats about four feet square, which we purchased to lay on the snow under our blankets, and to use for table cloths. Numbers of singular-looking dogs, resembling wolves, were sitting on the tops of the huts..... They made us comprehend that they were at war with the people who lived to the southward and to the eastward; but I could obtain from them no certain information. This morning the camp was thronged with Tlamath Indians from the south- eastern shore of the lake..... Spier: Klamath Ethnography Fremont also remarked on the scarcity of their horses.22 Fremont reached Klamath lake again May 6, 1846. He commented on the outlet, Link river, "this is a great fishing station for the Indians." His statement on the distribution of their settlements coincides with my information. They have fixed habitations around the shores of the lake, particularly at the outlet and inlet [Williamson river], and along the inlet up to the swamp meadow [Klamath marsh], where I met the Tlamaths in the winter of '43-'44, and where we narrowly escaped disaster ... . The Indians made me under- stand that there was another large river [Williamson river] which came from the north and flowed into the lake at the northern end, and that the principal village was at its mouth, where also they caught many fish. Fremont's party made a circuit of the lake. It is curious that he does not mention the settlements about Pelican lake on the western side, which he could hardly have avoided. On the other hand it is not clear how closely his memoirs follow his diary. Attacked while on the western side, his party reciprocated by raiding a village on the opposite side, perhaps one of those on Williamson river just above the present highway bridge. Behind the sage-bushes where they had taken their stand every Indian had spread his arrows on the ground in fan-like shape, so that they would be ready to his hand ... . Quantities of fish were drying, spread on scaffolds, or hung up on frames. The huts, which were -made of tall rushes and willows, like those on the savannah above [Klamath marsh], were set on fire, and the fish and scaffolds were all destroyed.23 The first invaders remembered by the Klamath came from the north about 1835. The stories about them have now a legendary flavor. The first group was mounted; the Klamath dubbed them mpe'tko-qks (from metc!a', to pack about) from their roving habits. An alterna- tive name, pdsai'iuks (French-Canadian in Chinook jargon), more surely identifies them. During their brief stay at the mouth of Williamson river they bought fat dogs of the Klamath, which to the latter's amazement, they ate. For these they gave buttons and metal disks with a central perforation (wika'lkal). Almost as early was a white contingent moving northward on the west side of Klamath lake. The Indians watched them unseen. They were impressed with the horses, axes, and the food of the whites. 22 Fremont, Narrative, 122. 23 Fremont, Memoirs, 483, 484, 494. 7 1930] 8 University of California Publioations in Am. Aroh. and Ethn. [Vol. 30 When the latter broke camp they left behind a little flour, which the Indians carried home and divided. They rubbed it on their palms and smeared their faces.24 Other early whites came from the southeast on horseback. This marked the beginning of the great migration by way of the Modoc country to the Pacific slope. On the whole the Klamath came off remarkably well in their relations with the whites. The full brunt of immigration was borne by Modoc and Pit River peoples. The Klamath were spared the Oregon wars of the fifties, which all but annihilated the tribes of western Oregon and northern California.25 The earliest white settlements in the neighborhood were about the present Jackson- ville, Medford, and Ashland. The first white settlement within Klamath territory was near the village of kowa'cdi at the northern end of the lake. This was about 1860-65, hence may have reference to the first agency established there. Subsequent relations have been on the whole quite amicable. The Klamath took no part in the Modoc war of 1872-73. Habitat The territory of the Klamath is on the high plateau of eastern Oregon, in the shallow basin wherein lies the group of lakes that give rise to Klamath river. The southern half of this basin about lower Klamath and Tule lakes is Modoc territory, extending southward across the lava beds in the direction of Pit river. The northern half, which lies against the Cascade range, is the home of the Klamath. The Klamath properly reside on Klamath lake (or as it was formerly known, upper or middle Klamath lake), on Klamath marsh, and on Williamson river, which connects them. At least, most of the permanent winter settlements are on these three waters, with a scatter- ing on their tributaries. Summer camps are established over a much wider territory, as far, it would seem, as the natural limits of this drainage area. 24 Salt was equally unknown to them at this time. 25 A somewhat obscure account was given me of the inclusion of Klamath in the great round-up of western Oregon Indians on Siletz (?) reservation in the fifties. These had been wintering at the Dalles and were herded with the Indians there. The Upland Takelma and Molala were penned in together and fed by the soldiers. Some died every day; the Klamath suspected that they were poisoned. The Klamath were kept apart, and, thanks to the strenuous arguments of their chief LEle'ks, were not only freed but supplied with beef and an escort of soldiery to conduct them back to their own country. % tp .- - a 0 . Pig. 1. Map of Klamath territory. t4 0 4b 4 Spier: Klamath Ethnography The western and southern limits of the range are easiest to define because of the topography and the proximity of other tribes. To the east and north the open deserts and broken country extend indefinitely. The nearest of the shifting settlements of Northern Paiute on the east at Goose and Summer lakes are sixty and more miles from Klamath lake. Those of the Warm Springs tribes are one hundred and fifty miles distant, more than halfway across Oregon. Teit's data show that more than a century ago the neighbors of the Klamath in this direction were the Sahaptins at no great distance and beyond them Molala and Cayuse. The wide plain south of Klamath Falls seems to have been unoc- cupied. The Modoc lived south of the Lost River hills about Tule lake and on lower Klamath lake. My Klamath informant doubted that they lived on Miller lake, or had permanent settlements on Lost river. Both tribes congregated at one place only for the early spring fishing, namely on Lost river, a mile or more below the buttes at Olene. The Klamath camp site is on the north bank, the Mlodoc on the south. The southern boundary of Klamath territory skirts the Klamath valley from this place as far downstream as Spencer creek, near the California-Oregon line, below the canyon of the Klamath. Below this on the river are the Shasta, whose northernmost outposts are on Shovel creek and north of the river on Jennie creek. The Klamath have no settlements on the river, but fish, hunt, and gather roots at such points as Keno, Spencer creek (ma'lbuk'is), and at laik'elmi, five or six miles upstream from Dorris (?). The western boundary is the Cascade watershed, including, how- ever, territory on the northwest side of Mount Pitt (Mount McLough- lin) and again southwest of Crater lake. The Klamath hunt north- west of this conspicuous mountain in a region of many streams called la'ndjli (possibly an American name). Fifteen miles southwest of Crater lake is Huckleberry mountain (iwumka'ni, huckleberry place) where the Klamath meet the Molala similarly engaged in berry pick- ing. The latter regularly occupy the highlands of the upper Rogue river drainage above Trail creek or some point below Prospect. These Molala form a southern extension of that tribe from the upper Ump- qua drainage. The northern and eastern boundaries are somewhat indefinite. Palmer stated: 1930] 9 10 University of California Publications in Am. Arch. and Ethn. [Vol. 30 East of the Cascade mountains, and south of the 44th parallel, is a country not attached particularly to any agency. That portion at the eastern base of this range, extending twenty-five or thirty miles east, and south to the California line, is the country of the Klamath Indians.26 This is essentially correct although the boundarv might better be set somewhat farther south to exclude the headwaters of the Deschutes, in earlier times the home of Sahaptin peoples. The country north of Klamath marsh is considered too open and cold for settlement. On the eastern border are a series of lakes (Silver, Summer, Abert, Goose) and rims which give on the desert to the east. There are Northern Paiute settlements on the eastern sides of Summer and Goose lakes, perhaps those of the Walpahpas mentioned by Meacham. The meeting with theKlamath and Snakes was one of interest to all parties, from the fact that they had been enemies, and the chiefs had not met in person since peace was restored. Living in the country intervening [presumably Summer to Goose lakes] was a small tribe of Wal-pah-pas, who are half Snake and half Klamath. They were mediators, though sometimes fighting on alter- nate sides, as interest or affront ga-ve occasion.27 Sycan river, Sycan marsh, and Yamsay mountain are summer resorts of the Klamath, who had their easternmost settlement (hicdic- lue'lukc) somewhere west of Gearhart mountain. Chewaucan marsh, between Summer and Abert lakes, was probably also visited, although I wa.s not told so. Its name is Klamath, probably tcuwa'k'an, arrow- head place, derived from the abundance of arrowhead or wappatoo roots (tcuwa') gathered there.28 SETTLEMENTS The occupation of this territory is determined by the manner of life. The fixed villages are the wiinter residences to which people return year after year. Early spring finds them leaving for favorable fishing stations where there are successive fish runs. Through the summer they move to the prairies to gather edible roots and berries or to the mountain and desert to hunt. During most of this time families are widely scattered and the winter villages quite deserted, but with the ripening of pond lily seeds in the marshes during August and September they again congregate. The pond lilies grow in enor- mous quantities, the seeds (wo'kas) constituting a staple second only 26 Palmer in Report, Comm. Indian Affairs, 262, 1854. 27 Meacham, Wigwam and War-path, 232. 28 Coville, Notes, 90. Spier: Klramath Ethn,ography in importance to fish. Summer is also the time of travel and trade, taking people as far as the Dalles where they may elect to spend the winter. Winters are exceptionally severe at this altitude (4000 feet), with deep snow and lakes, marshes, and even streams solidly frozen. Movement is stopped and unless a winter residence has been chosen which gives some possibility of fishing, starvation may be faced before spring. These winter sites are in sheltered, sunny spots on running streams or at springs warm enough to prevent ice from forming. The families return in the fall, rebuild the earth-lodges from the remains of the previous year's structure, and are all snugly housed by mid- December. The only permanent settlements are the winter villages. The summer residences are shifting, and while reoccupied from year to year, there is no feeling that particular camping localities belong to certain groups. The summer camps are rather indiscriminate aggre- gations of any Klamath, but the winter locations are so fixed of habit as to give a measure of political separatism to the several localities. The case of Tom Lang's family is typical. Their permanent winter home is at iu'lalo77E, a large village on the west side of Link river at the foot of Klamath lake, hence Tom is considered one of the southern lake people, iu'"laloqkni gen- erically. His family is accustomed to m,ove for the spring fishing to ditk!aks, Barclay spring, or to Modoc point, both on the east side of the lake, where they erect temporary lodges. Crossing the lake to d&'ktcoqks they leave their canoes and travel overland to woksa'lks on the north shore of Wokas marsh. Here Tom was born in July. When the berry and wokas harvest is ended they retrace their steps and take canoe to Algoma to pick wild plums, berries, etc. From there they drift back to their winter home. The towns for the most part lie along Williamson river, on the southern side of Klamath marsh, and along the eastern shore of Kla- math lake. As these are winter settlements, the open lake is avoided, and occupied only where warm springs outweigh its disadvantages. They cluster along Williamson river whose sheltered valley is dis- tinctly warmer than the lake front a few miles distant and where the running stream contains fish most of the year. The towns are not isolated, compact groups of houses, but stretch along the banks for half a mile or more. In fact, the settlements on Williamson river below the Sprague river junction form a practically continuous string of houses for five or six miles, the house pits being, in many spots, crowded close together. Informants insisted that many of these were occupied at the same time. When we consider that these earth-lodges may have housed several families, there is strong suggestion of a considerable population. 1930] 11 SccTI/e 0I / 2- .1 . 4 Mig. 2. Settlements in northern Klamath territory. Triangles represent cremation piles; rectangles, mourners' sweat-lodges. ' 1<16/17ath Spier: Klarmath Ethnographfy The towns listed below are mostly winter settlements. Each held one or more earth-lodges. There may be more than are listed on Sprague river, the only part of the region I did not visit. With this exception I think none have been overlooked.29 The towns are grouped in the list according to tribal subdivisions. I. Klamath marsh-Wiliawson river group (ai"ukckni).-Settle- ments are in two principal groups: along the southern margin of Klamath marsh and on the lower Williamson river. Although a few more were indicated in the latter region, I was told that the winter population at the marsh is the larger. They do not live on the western side of the marsh, as the snow-fed streams there are too cold for fish. The upper part of Williamson river, a rugged canyon, is unoccupied, but the towns are practically continuous from above the Sprague river junction down to the point where the river leaves the hills. Several others are on the lake front near-by. Evidently the middle Sprague river settlements should be included with this subdivision, but there is some doubt. Towns there are widely separated and extend a con- siderable distance up the river; the easternmost is somewhere west of Gearhart mountain. I am not certain that Sycan river is regularly occupied; roots are dug there. Sycan marsh is not occupied; it is a summering place for a'ukckni, for those of Yainax in particular. 1, muitcuia'ksi, near the bridge toward the eastern end of Klamath marsh. 2, k'Etaiwa's, along the eastern side of the marsh. A cremation pile lies south of this town. 3, guipgu a'ksi, houses stretching for perhaps two miles here, with a cremation pile near-by. 4, i'wal, and 5, kla'djoksi, houses for some distance along this tongue extending into the marsh. A third cremation pile is east of i'wal. 6, du'llkuit, a very large town, perhaps a hundred house pits in an area three hundred yards by one hundred. Others extend along the marsh to the east.80 All the tribelets congregated at this spot in the summer to gather lily seeds. 29 Gatschet lists sites (1:xxviii footnote, 142, passim). Many of his items are only place names; no permanent town is included that is not also in my list. In a few instances he has misinterpreted his information. 80 This would be an excellent site for archaeological exploitation. House pits here are more distinct than any others I have seen; elsewhere they have been leveled by the plow. 1930] 13 14 University of California Publications in Am. Arch. and Ethn. [Vol. 30 7, awa'lwask-an, at least five earth-lodges here. Houses scattered between this site and the last. 8, wa'ktale's, and 9, la'laks, slope steps,3' two settlements on the higher ground where Williamson river leaves the marsh. The second had twenty houses on both sides of the stream. 10, lobo'kstsoksi, on the bluff on the left bank of the Sprague river at the railroad bridge, with a few houses on the opposite side. There is a dam for salmon here. Gatschet gives the locality name as ktai-tiu- pakshi, standing rock.32 11, A site on the south side of Sprague river below the dam. 12, l!otcwa`e'ts is a large winter settlement about two miles above the dam on the south bank of this river. 13, koma"eksi is a site on both sides of the river south of Braymill, four miles from Chiloquin. There is a cave here which was the home of KEmiu'k-umps, the culture hero. Near-by is wo'kstat, the site of the only first-fish rites among the Klamath. 14, la'umkain is about six miles above this cave. Many people lived on both sides of the river there. A cremation pile is there, probably the only one on Sprague river. The place takes its name from a spring (ka'u-m) on the western bank. Whether there were permanent settlements above this on the river is difficult to say. At least there were settlements (15) of some sort on the river near Yainax before the coming of the whites. 16, hicdiclue'luke, a settlement west of Gearhart mountain. 17, bEzUikse'was (from bEziiks, obsidian); there are many people in this town on the right bank which extends for a mile below the Sprague river mouth. On the opposite side, ha'potgOs, is a dam in which spring fish are caught; near-by are a few summer houses. 18, takalma'keda, a town half a mile long on the right bank. The houses are on the bottomland along the river with quite a number on the opposite flat (19), called k'tai'di, lava. A little south is a part of this settlement, called gl'atspal'is. Of the distinct pits at takal- ma'keda, four are about twenty-five feet in diameter, one, fifteen feet. Northward from the town are a considerable number of storage pits and a fish dam. 20, djigia's, houses on the river level on both banks and on a higher bench on the right. There is a ford here. 31 Gatschet, 1:74. Several other camping places in the vicinity on upper Williamson river are also mentioned. 32 op. oit., 1:76. 1]Spier: Klamath Ethnography 21, ]kc!o'ltawas, a town of ten or fifteen earth-lodges on the left bank, with houses on the right bank extending to the next settlement downstream. There are otter (k!olt) holes in the right bank, hence the name. A mythical otter that lived here is seen in a rock standing in the water; on his back are his children, smaller rocks. 22, at'aw1kc, "to catch fish in still water," six to a dozen houses on the right bank; a few on the left. The water here is quiet; fish are caught with the trolling net. 23, ya'ak (ya'ac, willows), two dozen or more houses on the right bank. This was the largest river site in early reservation days, the road from Link river (Klamath falls) fording the river here. Of three house pits together, with their margins nearly in contact, two are twenty-five or thirty feet in diameter, the other fully five feet more. A cremation pile stands on the hill above this town. It is now leveled by marauders, as are all the others in this country, but it apparently covered an area forty feet in diameter. 24, tsa'k'wi, houses on the right bank with a few on the left. 25, wlt-amiumpsi, bear's place (wvt-aem, black bear). This is a high bluff on the right bank above an eddy in the sharp bend in the river; the scene of vision questing. A large shaman's earth-lodge is located here. Fish are caught all year long above this town. 26, goyEmske'Egls, crawfish crawl out, or kiEke'trius, small lizard, a town half a mile long beginning a quarter-mile above the highway bridge. The houses are on the abrupt slope right to the river's brink, and some on the little bluff above. There are pits of perhaps forty or more houses here, but at least half are dome-shaped cooking houses, for they are for the most part only six to ten feet in diameter, with few as much as fifteen feet. One of eighteen feet has its dome-shaped kitchen lodge clearly indicated on the north side. 27, wEla'lksi, named for a big quaking-aspen (wFla'l) there, is on the eastern shore of Agency lake. A few aS'ukckni live here in winter, in five or six small houses. 28, lok'o'g t, warm, a little warm spring near which are two houses on the higher land near Agency lake. 29, tco'klaluimps, service berry, which grows here abundantly, is a town of one or two small houses overlooking the lake where the Chilo- quin road meets the lake highway. Other towns may have been at ya'mzi, on the western side of Yamsay mountain, and kokena'okE, Spring creek, a large northern affluent of Klamath marsh. 1930] 15 16 University of California Publication in Am. Arch. an4d Ethn. [Vol. 30 II. Agency lake group (kowa'cdikni).-The winter home of these people is on Agency lake, the northern arm of Klamath lake. My questioning did not make it wholly certain that this group is inde- pendent of 'aukckni. In fact they join them at k'Fftai'di (19) on Williamson river for the spring fishing. 44, kowa'cdi is close to the mouth of Wood river. The site is a quarter-mile long. One informant stated that it contained twenty houses, another two earth-lodges (one a shaman's), not incompatible statements. A spring here permits fish to remain all winter. A few houses are at another spring a quarter-mile east. III. Lower Williamson river group (du'kwakni).-These settle- ments are close to the mouth of Williamson river. This region, du'kwa, is one of marsh, separated by a considerable section of open and higher plain (pEno'`s) from the nearest atukckni settlements. PEno'vs seems not to have been occupied in pre-reservation days, although du'kwakni and a'ukckni now live there. Du'kwa people seem also to live across the lake at Pelican bay, the home of the gu'mb6tkni. But whether this is a regular thing, or only a matter of my du'kwa informant's personal history, I am not sure. At any rate they go to dig roots in the marshes of the western and northern sides of Klamath lake in summer, not to Klamath marsh. Their winter settlements are on the isolated bits of higher ground within a mile of the river mouth, and on the marsh front immediately south. Most of these settlements, too, are on the right bank of the river. 30, mo'aksda, five or seven earth-lodges on the left bank nearly a mile above the mouth. Five more houses stood on the opposite bank with numerous springs behind them and the du'kwakni cremation pile a half-mile above. 31, wickamdi, four or five houses on the right bank below this, one or two a little below, then a cluster of four and another of two farther downstream. 32, la'wa'lstot, three houses on the point forming the right side of the river mouth. 33, mo'gihkiunks, the largest settlement in this district with five or six earth-lodges and as usual a number of small structures. This is on the left bank a quarter-mile above the mouth. One or two houses are on this side immediately at the river mouth. 34, djqgUis, four or five houses at a spring on the lake front to the east of the mouth, with one or two more farther in the marsh. A. 0 K/lalnol /1 f&# Fig. 3. Settlements in southern Klamath territory. Triangles represent cremation piles; rectangles, mourners' sweat-lodges. 18 University of California Publications in Am. Aroh. and Ethmn [Vol. 30 IV. Pelican bay gronp (gu'mb0'tkni).-This includes the Pelican bay district (gu'mbot) on the west side of the lake, the settlements along Four Mile creek, and those on the marsh north of the lake. Runs of fish are plentiful on Wood river in the spring when the snow melts on the mountains. People come up along the river in the sum- mer digging camas. But above Fort Klamath there is too much snow to live during the winter. 35, sle'tsksi, a group of two or three earth-lodges on the west side of Seven Mile creek near its mouth. A little spring here keeps the water warm enough for fish even in winter. 36, wudok'kan is a settlement in the marsh a mile from the last, and east of Seven Mile creek. 37, iwuinau'ts, two earth-lodges on the western side of a little creek emptying into the lake two miles east of Recreation P. 0. Houses are scattered along the marsh shore to the northern side of Pelican bay. 38, din6'ksi, an open space overlooking the northern end of Pelican bay and containing many earth-lodges. Others are on the lower level, lEl'aksti, close to the bay. This is immediately north of the mouth of Four Mile creek. A cremation pile stands a hundred yards north of this town on the slope above the bay, beyond an outcrop of rocks said to be transformed shamans. A big lodge of a shaman on the lower level has a pit twenty-eight feet in diameter; another earth-lodge pit is nearly twenty feet across. 39, e'o 'kai, four or five earth-lodges a few hundred yards up Four Mile creek on the left bank. 40, walo'kdi, four to eight earth-lodges, one a shaman's, somewhat farther up the creek on the opposite side. 41, Wak'a'k, a winter settlement south of the high ridge south of Odessa. North of the settlement is a sweat-lodge used by mourners. 42, gai'loks or gaila'lks is on the point south of Odessa or more probably between Howard and Shoalwater bays. No winter settle- ments here but a camping place where du'kwakni and gu'mbotkni roasted camas in the spring. 43, sto'kmatc, Eagle point, is a similar spring camp site. In the hot springs here are harmless black snakes (kamdila'k) which crawl over the swimmers. Curiously, in view of the Klamath tendency to localize their tales, there is no story concerning this locality. Two other localities may belong to this group: 63, e'o'kalk, gulch, on Wood river toward the mountains, and 64, e'ukwa'lksi on the east Spier: Ktamath Ethnography side of Wood river in much the same locality. These may be the sa.me site. V. Klamath Falls group (iu" laloqtkni) .-All the people of the lakne from Modoc point southwa.rd are named for their principal locality, iu"lalo1qF, Klamath falls. Despite the large area, they are relatively few in number. In summer they resort to the marshes west of the lake or fish on Lost river. The district along the eastern shore of the lake from Modoc point nearly to Algoma is. called ne't luksi (ne" luks, a late sunrise), from the circumstance that a low mountain here cuts off the morning sunlight. 45, kEt!ai'ksi, point of rock, a promontory two miles or so north- west of Modoc point, is their northernmost settlement. 46, suwiak6t'ks, jump over, is a large village at Modoc. point. It extended on both sides of the headland and for three-quarters of a mile south. The place takes its name from a large rock (about three and a ha-lf feet high) which stands a half-mile south of the point. It was a trial of skill to run uphill and jump over the rock. Bare island (au'cmi) opposite is their camp site, and has no permanent settlements. 47, iula'u, promontory, is a fishing village occupied in early spring. Pictographs are painted on the rock slide north of this. 48, diu'wiaks (named for its little spring, di'wik) is a winter settlement at the railroad point Ouxy. Little fish ran in the spring early in the year. 49, 1au'omot, a half-mile south, is, like all these lake sites, crowded along the foot of the mountain on the lake shore. There are many earth-lodges about a large spring here. The lodge rings are so clear that this would be a good place for archaeological excavations. Its name is from kau'6om, a place where fish run. 50, dli'tk!als, boiling spring, is now known as Barclay spring. There are many earth-lodges north and south of this. Two cremation piles are to the north on the highway, one distant a hundred yards on a rock slide, the second twice as far and directly on the southern boundary of the reservation. Unlike the cremation piles on William- son river, no glass beads or iron was to be found at this second site. South of Barclay spring by three hundred yards is a deep nameless spring into which pups are thrown to train them for hunt.ing. A large sweat-lodge used by mourners is situated at the foot of the slope near the angle of the marsh east of Algoma. 1930] 19 20 University of California Publcationm in An. ArcIh. and Ethn. [Vol. 30 51, kolwa'l, a point, is Rattlesnake point at Algoma. Only a few live here during the winter but it is a favorite resort in the autumn when chokecherries and wild plums ripen. 52, wuk!o'twas, wood knot, Buck island, is so called because the island in the lake is likened to a knot in a piece of wood. In addition to old house pits here not occupied by any group of the historic period, there are a few winter houses on the southern end and summer camps at the northern end where plums could be gathered. It was a favorite sport to dive from the rocks (called sohowae'ks, diving) on the northern shore. 53, lama'teksi, a place on top of the end of something, is on the point east of Buck island, and 54, k!su'nk!si, grassy place, is three- fourths of a mile south on the shore. Many people are at both localities during the winter. There may be old house sites here like those on Buck island. Wa'to-qks is a hill a half-mile east where wild plums are picked. There is an outcrop of white rock high on the hillside above iu'li (meaning deep in), the end of the marsh east of these settlements. This is said to be the net of KEmiu'kUimps, the culture hero, spread to dry. 55, iwau'wonE, meaning against the side hill, is a large settlement on both sides of Link river at the highway bridge. A crematory is on the hilltop on the east side a quarter-mile north. Shasta sometimes swam the river south of the village or crossed by the islet below the little lake Ewauna to attack the group on the east bank. This settle- ment is at the lower end of the falls (Klamath falls) that constitute Link river in its brief passage through its gorge. This is one of the principal fishing places of the region and advantage is taken of the ledges that form the falls for the construction of fish dams. But these, like all others in Klamath territory, are attributed to the culture hero, KEm u'k u-mps.33 56, iu" laloqE, meaning on the end, is the largest village of the southern lake people and gives its name to the whole group. There are earth-lodges on both sides of the falls but particularly on the western bank from the river mouth to within half a mile of the present power dam. There are many at the point at the river mouth (hence the name of the site) and thence extend westward about a little bay. A group of tall upstanding rocks at the point, called k.!taig'' kls, hole 33 Gatschet's reference (1:94, 97) to the Indians scooping up fish from the dry bed of the stream when a strong south wind drives back the waters in the lake seems like a purely mythical reference, but is confirmed by Clarke (Roek Piles, 41). Spier: Klamath Ethnography through the rock, had only a narrow way between them to allow passage along the shore. On a point of rocks on the hill above stands a cremation pile. A hillock west of the bay is called lalau'k!ot, flat, for the many flat rocks on top. The opposite point of the river mouth also has a settlement, called tc!a'tc!6msi, for some unidentified bird. 57, wekae'ls, fallen tree, is a settlement of moderate size on the shore a mile west of the river mouth, now intersected by the road. 58, wut!ana'k6lks, a low gap, and 59, iup!a'tona, end place, are summer sites respectively at the ends of a little marsh (now drained) on the west side of the lake. 60, woksa'lks, lily seed (wokas) picking place, is a summer camp on the north shore of Wokas marsh. When leaving in the fall it is customary to sink the canoes at the edge of the marsh, hide the grind- ing slabs in the juniper thicket, and pack by a regular trail to de'ktco6qks, whence they return to the winter villages. 61, de'ktcoqks, netting place, is a camp site on the west shore of Klamath lake oppdsite Buck island. Midway between this site and the island is a favorite place for catching fish at night with a tri- angular scoop net. The point Skillet Handle is called iuwa!'lE, end of a point, and a flat-topped mountain west of Wocus bay, seoloplis. 62, sa'stitka'wals, Shasta stand on end, is Squaw point where a few Klamath falls people might be found in winter. There are holes in the rocks here in which they pound to make the wind blow, pounding in those on the south side, for instance, to bring a south wind. TRIBAL SUBDIVISIONS The Klamath are not a single political entity. There are four or possibly five subdivisions or tribelets, each occupying a distinct dis- trict, and practically autonomous. This is separatism of the familiar Californian order. Nevertheless, the cohesion rising from a common dialect, common culture, and a, uniform reaction against all non- tribesmen, which on occasion leads to jointly taking the field against them, produces a tribal solidarity resembling that of the Plains people. The separatism is immediately clear in any discussion with them. They invariably reserve the modern name Klamath for the Klamath marsh-Williamson river division; all others are klamath only by extension. The groups are readily defined by the districts they 1930] 21 22 University of Californmi Publcation in Ainm. Arch. and Ethn. [Vol. 30 occupy, the names of which they bear; this, despite the lack of any sharp geographic division, rivers or ranges, such as we should expect to figure in political separatism. There are only three or four miles of open, unoccupied country between the middle Williamson river group and that at the river mouth, for instance, or between the latter and the northernmost outposts of the Klamath falls people about Modoc point. The cases of doubtful divisions really stress this point. Kowa'cdi on the lake shore may belong with the middle Williamson division; though only three miles away over a low range of hills, it is somewhat isolated. Similarly the Sprague river settlements toward Yainax, usually included in the same group, are in quite a different district. The third feature pointing to their separatism is the frequent feuds between them. It is quite clear that these involve the tribelets; they never occur between the settlements of a single division. The point will be clearer if we contrast Yurok sentiments. There feuds are between the towns or with only a group of kinsmen within a town, and at most when several towns are involved in a joint cause, it appears rather that kinsmen and their friends have been drawn in. 4 Further, feuds are carried on much as warfare with foreigners; property is destroyed, women and children enslaved. Such slaves can be retained indefinitely or sold again. We may wonder however if they are not restored when a settlement is bought, or at most retained only to even the score in lives. The identity of these groups resolves itself solely into such aspects of group consciousness. There is no political mechanism within the group. When it comes to war with outsiders, each group can act for itself, others may join if they wish. At the other extreme, the towns are not sharply separated. A town is probably in large part a body of relatives, although not so clearly so that towns are exogamous. In short, the tribelet is a body with a community of interest which hardly extends to the Klamath as a whole. Only one informant, Pat Kane, suggested dialectic and cultural differentiation. The dialect groupings he gave are purely geographic. He also assigned shamanistic tricks, medical practices, and games to the several divisions, which all other informants emphatically denied. I could observe no linguistic differences and Gatschet recorded none. The subdivisions or tribelets are: 34 Kroeber, Handbook, 49. Spier: Klamath Ethnography I. a'ukckni, the people of Kilamath marsh and middle Williamson river, the Klamath proper. Evidently the Sprague river people are included here; there is some doubt, but I am even more doubtful whether there were settlements as far away on Sprague river as Yainax in pre-reservation times. II. kowa'cdikni, those of a village of that name on Agency lake. They may be A`ukckni. III. du'kwakni, on the delta of Williamson river. These are more closely affiliated with the Pelican bay people than with a'ukckni. IV. gu'mbotkni, those of Pelican bay and the marsh to the north. V. iu"laloqkni, the people of Klamath falls (Link river) and the eastern shore of Klamath lake. The names are those of the districts they occupy with the suffix -kni, those of, people of: a'ukei, Klamath marsh; du'kwa, the district about the mouth of Williamson river; gu'mbot, Pelican bay, possibly taking its name from Rocky point; iu" lalo0qE, on the end, the village at the outlet of the lake into Link river. It is now difficult to get a true estimate of the size of these groups. The a'ukckni count twenty-nine or more settlements, half of the total; of the others iu"laloqkni have thirteen, gu'mb6tkni seven or more, du'kwakni five, and kowa'edikni, one. The towns vary considerably in size, yet this mere count agrees closely with the populational pro- portions as my several informants phrased them. The a'ukckni are by far the largest group, perhaps as large as all the others together. Nancy Phillips recalls the large number of children who played together at gu'mbot, Pelican bay. At this place the settlements do form a large compact group. Next in order she places the du'kwa group, then those clustering about iu" laloqE at Klamath falls, and goyEmske'Egis on Williamson river. The settlements on Williamson river are small and scattered, but she places their joint population above the gu'mb6tkni. The iu" laloqkni were comparatively few despite their large territory. Of the a'ukckni more lived at Klamath marsh even in winter than on Williamson river. The huge size of a single site, du"Illk-ut, bears this out. Consonant with their unequal size is it that the principal chief, at least in early reservation days, was the principal chief of the a'ukckni; all others were distinctly subordinate in prestige. According to one informant, himself a du'kwa man, the a'ukckni are the highest people. KEEmu'kiUmps, the culture hero, made them first, best, and strongest; later he made the others. 1930] 23 24 University of California Publ4cation in Am. Aroch. and Ethn. [Vol. 30 WARFARE AND FEUDS Klamath warfare is deserving of the name in contrast to the con- flicts of the northern Californian tribes. Raids are carried into foreign parts for plunder, slaves, and revenge, and received in turn. No single town or tribelet is alone concerned: warriors are drawn from the Klamath at large. This is in marked contrast to the Yurok and their neighbors, e.g., whose foreign wars seem but an extension of the private quarrels that set town against town, and in which the large body of Yurok may remain neutral. Warfare is here not mere retaliation, murder for murder, but is engendered by patriotic motive and tangible benefits. * There can be little doubt that the Klamath attitude is due to their contact with tribes of warlike habit to the north and east. Raids from that direction had to be met and countered. The interesting circum- stance is that the Pit river people were under a similar compulsion but apparently made no effort to cope with the raids by reciprocating. So far as I can learn they never ventured against the Klamath despite their equivalence or even superiority of numbers. The cause for the difference seems to be that the Klamath had reason to raid; slaves and plunder were important to them. There was a market for these articles at the Dalles with much-coveted horses to be had in return. The Achomawi had no such outlet. I would not underestimate the difference in tribal solidarity, which may have been entirely lacking among the Achomawi, but the suggestion is that such solidarity may in this instance be a function of a common cause against a common enemy. The Klamath had their share of internecine warfare. Besides the close bond with the Modoc, normally friendly relations are maintained with the Molala across the Cascades, the Wata'nkni (Warm Springs people) of the Deschutes, and the Wishram-Wasco at the Dalles. Warfare is directed toward all other directions. Raids are exchanged with the Shasta several days' journey down the Kla- math river, the Upland Takelma on the Rogue river, and the Northern Paiute (Snake) of the desert to the east. The Klamath fight the Kalapuya and take horses from the Warm Springs Indians.35 Since horses were obtained by them, raids may be expected of the Ya'mokni two hundred or more miles away; northerners generically, but in this 85 Gatschet, 1:18. Spier: Ktmath Ethnography connection the Sahaptins of the Columbia and the Cayuse seem meant.36 Both Ya'monkni.and Shasta made slaves among the Kilamath. How infrequent were these raids may be judged by the circumstance that my informants independently mentioned the half-dozen recorded years ago by Gatschet, and no others. The Pit river people, possibly only the Achomawi, may alone be raided with impunity. Reciprocity on their part would lead them only into the intervening territory of the hostile Modoc, who may well have served the Klamath as a buffer. For their part the Klamath were stimulated by the lucrative traffic in slaves established with the Colum- bia river tribes, and the majority of captives were by all accounts Pit river people. Gatschet writes of these raids as a regular occurrence. "In April and May the Klamath Lakes and Modocs would surround the camps, kill the men, and abduct the women and children to their homes, or sell them into slavery at the international bartering place at The Dalles", and again, "the Lake men enslaved many every spring time."37 To be sure, consonant with Kroeber's view,38 raiding was undoubtedly accelerated by the appearance of the horse in the north. CGontacts were then more rapid, and the beast itself became an important and desirable article to possess. But there can be no doubt that the practice of taking slaves on Pit river was already established. This is not only implied in the informants' statements, but it must be kept in mind that slaves were an important element in the society of the Wishram, Wasco, and their down-river neighbors. This means that the Klamath were not only turning to slave-taking as a lucrative traffic, but that slave-holding as indicative of social distinction was becoming their habit by imitation.39 How frequently the Klamath made war as a whole is unknown. It seems doubtful that concerted action was frequent. Perhaps large- 86 Henry Renier, an old settler, told me of such a raid up the Deschutes about 1876. 37op. Cit., 1:liX, 19. 38 Handbook, 319. 89 While the Klamath do not distinguish among the Pit river peoples raided, Kniffen 's recent account of the Achomawi (Achomawi Geography) makes it clear that Klamath and Modoe descended most frequently on Achomawi settle- ments north of the Pit and also crossed it to Atsugewi territory in Hat Creek and Dixie valleys. The Achomawi proper about Fall river were subject to con- tinual attack; the Hewisedawi, south of Goose lake, and the Astariwawi, near them on the Pit, bore the brunt. The Atsugewi divisions, Atsuge (Eat creek) and Aporige (Dixie valley), were also freely raided. The Atwamsini, on the middle Pit, and the Kosalektawi, on the extreme headwaters of the river, were reached much less frequently. The other groups of Achomawi were free from attack probably by reason of their inaccessibility; the Ilmawi, Itsatawi, and Madesi at the western end of their range, and the Hammawi on the extreme east. 25 1930] 26 University of Califoria PubUcations in Aim. Aroh. and Ethn. [Vol. 30 scale expeditions were undertaken only after the slave traffic was established on an extensive scale. On one occasion at least all the chiefs were at a dance preparatory to going to war: this implies that all the tribelets were engaged. All the tribelets joined similarly in a. raid for revenge against the Upland Takelma. It is said that the Klamath marsh division (a'ukckni), by far the largest group, always extended the invitation for war, but I doubt that this is literally so. It is true that they led in aggressions just prior to reservation days, but this seems because of the initiative of their chief LEle'ks, the most prominent ever known, who was singularly interested in the slave trade for his personal advantage. Internecine warfare is probably as significant as that with for- eigners, for quarrels and blood feuds are quite common. Feuds exist between the several subdivisions, not within them, evincing a solidarity which but rarely extends to the tribe. Thus the du'kwakni of the mouth of Williamson river are known to have taken action against members of each of the four other divisions, even the gu'mbotkni of Pelican bay with whom they are closely linked. There are cases of reciprocated action by each of the five against the others indiscrimi- nately. Yet Pat Kane stated that the lesser divisions avoided conflict with the a'ukckni. Yet there is the case of his own people, du'kwakni, fighting with an 'aukckni settlement. Nancy Phillips said it was true that she knew of no cases involving gu'mbo6tkni or iu" lalo7kni (south- ern lake people) with a'ukckni, but this does not mean they would not fight. I think it hardly likely that the others are awed by the numbers of the a'ukckni, for the feuds are essentially between groups of rela- tives alone. Yet others besides relatives might be induced to join. A characteristic circumstance initiating a feud is the murder of a man residing with a wife of another tribelet. Ten or twenty of his male relatives take canoe and demand the delivery of the murderer or satisfaction from his relatives in the form of women, slaves, beads, dentalia, etc. The parley (nde'na) is held from a safe distance at the top of their voices. Instead of acceding, the offenders may offer fight. But an effort is usually made to reach an agreement by offering pay- ment. They may forestall retaliation by dispatching as a go-between (gako'`kwa) a relative of the offended group who is paid for his trouble, a canoe, a shirt, net, skin armor, or a woman's dress. If they have neither women nor money to offer, they give nets, weapons, any- thing, even to the last things they have. If no amicable adjustment can be reached they resort to fight. Even though the score be evened Spier: KlEmath Ethnography by the death of several on each side, the feud continues until payment is made. Continued feuds prove disastrous, for retaliation is carried out with all the intensity and destructiveness of a foreign war. Vil- lages are attacked at dawn, houses and property destroyed, women and children carried away into slavery. A chief may intervene; so I was told, though I doubt it. Chiefs do accompany parleying parties and other men'may echo their words. Beyond that there seems to be nothing to their position but personal prestige which has been enhanced by developments since reservation days. LEle'ks and his successors among the a'ukckni seem to have made much of their position. Pat's father and Tc!i'wi, both du'kwakni, killed a iu" lalonE man at sto'kmatc, Eagle Point. They were taken to Klamath marsh and each beaten five times at the order of this chief. This was in pre-reservation days, or at least prior to the establishment of the agency in 1864 ( ?). Nevertheless this is not aboriginal, but, according to Nancy Phillips, these "laws" were brought by her cousin from the north. The group that owes recompense, having lost fewer men, goes to the other offering payment to even the score. The value of a man's life is not a fixed item, as it is for the lower Klamath river tribes; this despite the similarity of marriage payment and the slur on children not born of a purchased mother. My informants professed ignorance of what should be paid, yet it can hardly be doubted that recompense accorded with a man's wealth and prestige. Children who are captured are kept; settling the score does not mean that they are restored, though this may well be only in case they have been sold. Their fate depends on their captors, who may elect to rear them as their own children. More commonly they are miserably treated, made to sleep with the dogs, fetch water, etc., and beaten if they are lazy. They are as quickly sold at Warm Springs and the Dalles as are foreign captives. With the latter at least there is an advantage in immediately disposing of them, that they do not grow up to carry back to their own people precise knowledge of the habits and topography of the Klamath world. K6'kdlinks of du'kwa was a chief (16'ki) in warfare. He would tell his tribes- men to get ready to fight with the iu" lalolkni, for instance. He was always in the forefront of battle. When a youth he went out to seek power by bathing and wandering through the mountains at night. His mother had told him that long before those of iu" lalonE had slain his father, L" 1' wis, a du'kwa shaman accused of poisoning one of them. He wanted to become a warrior to have 1930] 27 28 University of California Putblications in Am. Arch. and Ethn. [Vol. 30 revenge. While he was swimming in deep water at du'kwa he found a human leg in his grasp; it had been given him by the ghosts.40 Then he became "mean," revengeful, and led his people down the lake to iu' laloqE for vengeance. He grew aggressive and attacked gu'mbft, kowa'cdi, and other towns. He was a hard man to hit; they could not kill him in open battle. His du'kwakni band went up Williamson river to the A/ukckni town at bEzuikse'was. Before the fight opened, Po'moks, an a'ukekni from far up Sprague river, came to him and announced that he would join him. K6'kdlqmks agreed because he was a relative. During the battle Po'moks slipped behind K6'kdldiks and shot him. That was the only way they could best him; since then there has been no feud with du'kwa. Quite in accord with Klamath customs, K6'kdlqks band was accompanied by a du'kwa shaman named ekts'as. When one of his side was shot, this shaman ran to him, extracted the arrow, and sucked the wound. Pat 's uncle of gu'mb6t and Nancy 's father of du'kwa went to iu" laloJn where they killed a man. Some gu'mb6t men went to Buck island and killed a man there. Imme- diately a party followed them back and killed a man at gu'mb6t. A brave man from du'kwa lived at gu'mb6t. The gu'mb6tkni traveled with him to kill iu" lalo?7kni, &'ukckni, and kowa'cdikni. He stole women everywhere, carrying them to gu'mb6t as his wives. He tried to kill some gu'mb6t men whom he accused of tampering with them, but they dispatched him. After he was cremated all these women returned to their homes. An attack might be expected at any time. The Upland Takelma (Ha'nis or Wa'liumskni), for instance, raid the Williamson river dis- trict. Crossing by the ridges south of Cra.ter lake, they travel down the ridge separating that river from Klamath lake. From this emi- nence they can see the smokes of a large part of Klamath territory. Gatschet describes such a raid, plausibly dating it between 1848 and 1855.41 The Takelma descended on bziUkse'was in the middle Williamson valley at salmon fishing time. Most of the warriors were away at Klamath marsh to welcome some northern Indians. Crossing the river the Takelma attacked at dawn, killing a few old people, but they were put to rout by the fierce onslaught of but five cuirassed Klamath, men of niluiksi, the Modoc Point district. They were pursued by the whole force of Klamath, hastily summoned from the marsh. Some escaped, others were found dead of starvation or wounds where they had hidden in the mountains. Five old women were once left behind at Klamath marsh to gather lily seeds. They brought their canoes down to the falls, as far as they could go. All they could not carry they buried there to leave through the winter. A Paiute lurking on the opposite bank followed them down to the falls. He killed all five with his arrows. When their relatives found them overdue, they went to search. The Paiute, having stayed in the neighborhood to steal horses, was tracked and killed.42 40 So my informant said, but she surely meant spirits. 41 The Klamath Indians, 1:16-18. 42 Gatschet records a similar Paiute raid on Williamson river or perhaps the same (1:28). Spier: Klamath Ethnographmy The topography directs Shasta attacks against southern settle- ments. Coming by the open country along the Klamath river, they attack iwau'wonE at Klamath Falls, sometimes swimming the river there. Or they cross by the little island to the south to surprise the settlement on the east bank. Many years ago a large body of Ya'mokkni, northerners, descended on the Klamath on Sycan marsh in the northeastern corner of their territory. These may have been marauding bands of Shoshoni and Bannock, fought by the soldiery of Fort Klamath in the summer of 1864.43 Henry Renier, an old settler, told me of Indians coming south about 1876 or earlier by way of the Warm Springs country from the Snake river range east of Lewiston, Idaho. Northern Paiute scouts use va.rious calls in order to keep contact with one another, imitating the crow by day, coyote in the evening, and owls at night; unlike Klamath scouts. When these are heard, women and children take to canoes, staying on the water until morn- ing. The screech owl is thought to call, kokEde'l skiu'ki'u, who sent me? A thrush (sko"'le) names who is coming, wi'mpis Ege'pkwa, they are coming. Klamath hunting in the Cascades might meet Upland Takelma from across the range. On one occasion (perhaps about 1850) a party of five or six was wiped out with the exception of a boy who made his way at night, arriving footsore at Klamath marsh. Those who fetched the bodies for cremation found that the Takelma had taken scalps, hands, and feet as trophies. Two of those killed were brothers to the maternal grandfather of my informant, Coley. A wife of this grandfather was an Upland Takelma woman, perhaps a slave, who was a sister to the Upland Takelma chief T6kt6'kli 'red). They killed her in revenge. She had been living among them at the marsh for years but had no children. The reciprocal raid can be reconstructed from Gatschet's informa- tion44 and my own. The expedition was undertaken at the call of LEle'ks, the most prominent a'ukckni chief and a distant relative of Coley's grandfather. Men of all the tribelets took part. Crossing the mountains on foot, they found the Takelma encamped in Molala territory at the head of Rogue river. They attacked as the Takelma rose at dawn, scalping many and carrying off women and children. Among those killed was T6kt6'kli, whose hands and heart they took. After the man was shot, LEle'ks lanced him through the throat. LEle'ks invited all the Klamath to a huge scalp dance at Klamath marsh. During the dances a wife of Coley 's grandfather danced about with the heart in her teeth. The captives were dragged by their arms and foreed to dance close to the fire. A prepa.ratory performance, called nl'keba, is an indispensable preliminary to a raiding party. It can hardly be styled a war dance, 43Gatschet, 1:lxii. 440p. Cit., 1:16. 1930] 29 30 University of California Publications in Am. Arch. and Ethn. [Vol. 30 as it is doubtful that it contains any true dance form. It is held in the open for from one to three nights depending on the size of the group engaged. It consists essentially of two things, mimicry of an onslaught by the warriors and prevision of their success by a, shaman. The men are dressed and armed for battle. Possibly they dance to the songs by jumping with both feet together. In their midst is a shaman who predicts their fate. Possibly only women sing and jump about, pretending to stab at imaginary foes with spears. Early on the morning tha.t the war party sets out, the warriors rush at their women and children to show how they will act in the attack. This sham attack is called t'k!oVla.45 The shaman exercises his clairvoyant powers before they set out or while encamped on the march. He jumps about in front of the men, "looking through them," and predicting who will be shot, and how the enemy will be slain. One informant ha.d it. that those for whom disaster was predicted would stay at home; another that they were too brave to do so. They dance and go to war nearly naked, wearing breech clouts or the short fringed skirts, but not the more recently a.cquired skin shirt. Their ha.ir, and that of the dancing women too, is drawn into a single loose bunch to fall down the back, wound close to the scalp with strips of skin: deer, otter, mink, or weasel. Only women shamans may use the last. Neither sex wears basket hats with this peculiar headdress. (Incidentally, quarreling men who come to blows seize one another by the long ha.ir.) Their faces are pa.inted; men's white, women's red, sometimes with blood from the nose. The shaman's garb is slightly more elaborate; his hair is tied with weasel or mink, around his neck a rattlesnake skin, and the skin of a biull snake across his chest a.s a bandolier. On the march with the war party, he wears skunk skin and eagle feathers. Shamans accomparNr war parties not only to fight but especially to use their clairvoyant gifts in their behalf. Wounds and broken limbs are treated by them, a distinctly unusual practice for shamans, but so well attested by circumstantial detail in this case as to be beyond doubt. Wounds and fractures are sucked and the patient turned over to lay hands for further treatment. 45Gatschet (1:196) describes this for the Modoc. "The whoops were chanted and howled while going round in a circle for one to two hours; even now they are heard on solemn occasions. This uniform performance was, however, inter- rupted sometimes by feigned attacks on a supposed hostile force lying in ambush or marching past. A scalp-dance tune is added, also battle cries."I Spier: Klamath Ethnography Women also accompany a war party according to one informant, but this is denied by another. They cook or paddle; that is easily understood, but I am inclined to believe that they actually participate in conflict. Perhaps the solution is that they accompany feudists, but do not go on extended raids into enemy territory. Some are said to fight with the men, shouting and jumping about to add to the terrify- ing effect of the surprise. Armed with short spears by preference, they help to catch women and children and to slay the aged as they run to hide. Leaders of the party are appointed by the chief, usually men of affluence and commonly his relatives. There may be four or five leaders of a single large party. The qualities expected of them are bravery, ability to maintain their good sense in the excitement, and to go without sleep. Chiefs became such largely by the prestige of successful leadership in battle. The war party proceed well-nigh naked and afoot no matter what the weather. It is of some interest that even after the advent of the horse they prefer to go afoot. This is reminiscent of the Blackfoot, but in our case at least may indicate how recent is the acquisition of the horse. My information suggests the second quarter of the last century for the coming of the horse, and the wars were over by about 1865. The raid against the Upland Takelma mentioned above was undertaken afoot. But this is not really a case in point because it is doubtful that there were then horse trails across these timbered moun- tains. The introduction of the horse facilitated long-range raiding. Gatschet's informant describes its use against the Pit river tribes about 1848-50; even then it was a ride of three days and a night.46 Canoe travel was of little importance in any but expeditions within their own territory. A warrior's armament includes bow, club, and sometimes spear, with shield and body armor. Fighting is wholly individual; there is no sort of battle array. After the first rush of a surprise attack, they fight by preference from cover. No line of bowmen ever fights in the open, nor do lancers make a concerted charge. The sinew-backed war bow is a powerful short-range weapon.47 Arrow shafts are wooden, the heads of stone loosely affixed so as to remain in the wound when 46 op. cit., 1: 19-27. 47 Fr6mont observed that in 1846 he saw Indians behind the bushes with their arrows spread fan-wise so as to be ready at hand (Memoirs, 494). 1930] 31 32 University of California PutbUationw sn Am. Arch. and Ethn. [Vol. 30 the shaft is withdrawn. The tips are not poisoned.48 However they believe that those made of obsidian obtained on a mountain west of Klamath lake are poisonous. A tale has it that an Upland Takelma shot a presumably mythical man, called Tcewls hu'liums (alighting flicker), there. Crow laughed and he turned into the pillar of rock still standing. This is presumably the widely known tale of the shat- tered rock-monster. A war leader, a rich man, carries several quivers; one under each arm and a third across his back. The short stabbing lance is stone headed; the club simply a straight stick or one bearing a knot for a head. Armor is the familiar long skin tunic and wooden slat armor of waistcoat shape. The round hand shield is used, but not the curtain shield, i.e., a skin hung before the body. Sock-like skin gloves sometimes protect the forearms. It is planned to arrive in enemy territory at the end of the day. Several good runners are sent forward as scouts (gau'uliu'kdja) to look for smoke from enemy fires. Scouts do not use animal cries like those of the Northern Paiute. The camps are surprised at dawn, for, in addition to the confusion, there is the advantage of a long day in which to make good an escape. Yet the fight may last several days, I was told, but it is more likely that the raiders withdraw at once unless beset by a rallied foe. One thing sets off the intensity of this conflict from that of the Arizona Havasupai as I found it; there never occurs here the gesture of bravado of drawing off for refreshment before returning to the attack. Yet there is something formal about these fights too; the leaders make a point of engaging with men of like rank in the forefront of battle, although they do not disdain lesser men. Warriors do not pat the mouth as they charge (go'lgi) ; this is done only in the scalp dance. The foray usually ends with the fleeing of the ablebodied among the enemy, the killing of the infirm, and the capture of women, children, and such property as beads and quivers. These are kept by their captors; perhaps some division of the spoil is made. Disputes are inevitable: Gatschet's informant describes how he wantonly shot down a girl whom he claimed as his due rather than let another have her.49 Pat Kane told me that all the booty is carried to the principal chief who asks for an accounting of the dead and slaves. The leader enumerates how each man took slaves or sealps and how the opposing chiefs were killed. The trophies of these chiefs are exhibited. I doubt his statement, unless he had in mind the prestige 48 Gatschet 's informant credits the Modoc with poisoned arrows (op. cit., 1:90). 49op. cit., 1:23. Spier: Kkimath Ethnography peculiar to the chief LEle'ks, to whom they were accustomed to trade their slaves for transportation to the north. On the other hand this may be a reflection of the Plains custom of reciting coups. Scalps are of liberal size, including the whole skin above the ears and brows, as is generally the case among western Indians. Bodies are dismembered for trophies; hands, feet, ears, heart, the lips cut away as one piece, genitals, and the whole head. (One informant denied that the head is taken entire.) *Trophies are never taken from women nor from any men except leaders, the best fighters. There may then be some truth in the statement of Gatschet's informant that they disdained scalping Pit river people.50 Kroeber has pointed out that the northwestern Californians as far east as the Shast.a neither scalp nor hold a victory (scalp) dance.5' Klamath practice is in accord with more general custom. The scalp dance (sa'tca) is one of rejoicing over vengeance achieved. Hence the invitation issues from a relative of the avenged, not of necessity from a chief. Dancing begins immediately on the return of the war party and lasts two to five nights depending on the size of the group that collects. A minor raid is of course a ma.tter of only local concern, but an affair of consequence brings everyone to a. celebration for the full. five days dictated by tribal pa.ttern. Gatschet's informant recounted how on one occasion they made their captives dance before them at the first camp on their homeward journey.52 Older people of both sexes,take part in the dance as well as the returned warriors. The young are barred and children especially must be kept at a distance. Quiet, decorous conduct is expected of those who do not dance, else they would be shot, for at this time the dancers are violent. Faces are painted white or red; hair is tied as for the war dance. They fast or at least eat no meat nor fish. This is the taboo that the Klamath impose during every crisis. There is no special purification for those who have slain an enemy. The dance takes place in an open place, the very one on which the war dance was held, according to one informant. A huge fire is kept burning at the middle, where captured bows and quivers hang on a pole. The captives are huddled at its biase, the older ones with their wrists tied before them. 50 Op. Cit., 1: 19. 51 Handbook, 844. 52 op. Cit., 1:20. 1930] 33 34 University of Calfornia Publications in Am. Arch. and Ethn. [Vol. 30 The dance begins in the evening. The women are provided with sticks a meter long from which the scalps are pendant. As instanced above, one carried her enemy's heart in her teeth. Two women com- mence a sinistral circuit; others, men and women alike, fall in behind them two by two until everyone is engaged. They step forward slowly, the body swaying from side to side as they stamp in unison with the right foot, bend forward with sticks thrust rigidly and obliquely before them, drag the left foot up, and stamp forward again with the right. All the while they are singing a meaningless refrain, wiha'ho wiha. From time to time someone leaves the ranks for a moment to shake the pole, crying aloud. The captives are sometimes made to dance inside the circle with the same step; forced to rejoice over their own defeat. A big boy may be struck with a club, a little one is car- ried on a man's back. They are whipped with the scalp sticks as they dance. When the song ends the women jump about with feet together, crying a'ho, a'ho, and whipping the huddled slaves with the scalp sticks, they cry alalalala, as they pat their mouths (a practice called gababa'a). During the intervals of rest the sticks are set upright in the ground. Pat Kane drew a slightly different picture. The dancers form a circle, each with his hands on the shoulders of the one in front; first the leader of the war party, then a file of men, followed by the women. They begin by crying even though no Klamath have been killed, but soon begin to sing. Scalps are sometimes set on the central pole. On one occasion they had taken the scalp and the store-bought shoes of a Shasta which they hung on the pole and danced around. Gatschet describes a dance, perhaps for the Modoe, with the implication that it is the scalp dance, but it seems identical with the Ghost dance. "A peeled tree, sometimes twenty feet high, was planted into the ground, otter and rabbit skins fastened on or near the top, and below them the scalps of the enemies killed in battle. Forming a wide ring around this pole (w'alash) the tribe danced, stood or sat on the ground, looking sometimes at solitary dancers, moving and yelling (yAika) around the pole, or at others, who tried to shake it, or at fleet horses introduced to run inside of the ring. Circular dances are of course performed by joining hands. "53 a The last dance was held on Williamson river about 1865. Two old women, wives of a Paiute named l!oila`wokt, were the captives. After the dance they were freed. 53 op. Cit., 1:196. Spier: Kkamwth Ethnography After the dance, scalps and trophies are usually preserved by those who called the dance, relatives of the avenged man. They a-re dried and stored far from their homes to be brought out and danced over at each new scalp dance. Sometimes bits of the scalp are sewn in rows down each breast of a woman's dress, a Plains custom impossible, of course, before the introduction of these Plains-like garments. Both practices are un-Californian. There is a question whether the notched rattle (ulo'ks) is played in the war and scalp dances. Neither drums, which the Klamath formerly lacked, nor rattles are used. Only one man, Ma'gai', whose day was well before the middle of the last century, knew the notched rattle. He knew all about war. He had overheard the Upland Takelma singing to it, a.ccording to one informant, and always used their songs when the Klamath made war on them. No resonator is used; the notched stick is rested directly on the ground and rasped with another stick.54 CHIEFTAINSHIP Chiefs are of but minor significance in Kla.math life. In contrast my informants invariably stressed the importance of the shamans. As an instance the largest house pits pointed out were always those of shamans. This relative unimportance of the chiefs was true at least until just before reservation days. At that time contact with the Columbia river tribes, whose concepts of chieftainship were sharp and deeprooted, coupled with the developing trade which began to make considerable differences in wealth among the Klamath, operated to enhance the position of the chiefs, the rich men. It is significant. that LEle'ks, their most prominent chief, had spent several years of con- tinuous residence at the Dalles. All this was sharply accentuated on the establishment of the reservation by the government's recognition of lay leaders while frowning on the shamans. It must be distinctly understood that the picture Gatschet draws of chiefly powers is non- aboriginal.55 In fact, his principal informant, David Hill, was one of the newly raised chiefs who, we must believe, was not inclined to underrate his new importance. Both my best informants, Nancy 54 The familiar association of this rasping sound with the voice of the frog and its use in a social dance (the Bear dance of the Basin Shoshoneans) are both unknown to the Klamath. 55 Gatschet, 1:58-62. 1930] 35 36 University of California Publications in Am. Arch. and Ethn. [Vol. 30 Phillips and Tom Lang, doubted that there had been anyone who could properly be called chief in pre-reservation days. As good a test of this as might be was the difficulty I had in getting the names of pre-reservation chiefs, let alone their functions. Shamans on the other hand were freely named and the many cited came far from exhausting the list. Gatschet encountered the same difficulty. "Many times I attempted to obtain a list of the former head chiefs of the two chieftaincies [Klamath and Modoc]. I succeeded only in learning the names of two chiefs recently deceased, and no biographic details were obtainable." This he explains was due to their strong repugnance to name the dead,56 but I found hardly a trace of a name taboo. This situation seems much like that of the Great Basin tribes and the particular functions of chieftainship reenforce the view. Some settlements have chiefs, others do not. It is very doubtful that they feel that political solidarity demands a chief for each sub- division. The three chiefs alone named with some certainty were all a'ukckni men (Klamath marsh-Williamson river division): TcV'lokin who lived at the site of the present town Chiloquin, LEle'ks at bez-ukse'was across the river, and Motcuinka'sk.Et at Yainax. Others of this subdivision were L''lu at ka'umk-an on Sprague river and Do'tklEm sa'rdjent (literally middle sergeant) at about the same place. In other tribelets were my informant's grandfather at mo'gijkuinks, and Ko'kdl.dks, both du'kwakni; a chief and another unnamed war leader at gu'mbot. It is quite in accord with this that none were named for iu" lalo"kni, southern lake people. One informant suggested Link River Jack (nelo'lotk) for this group, but Tom Lang, himself one of them, was certain that this man was a post-reservation leader and doubted that his people had ever had a chief. All these chiefs were contemporaries of the period 1850 to 1870. Their predecessors, if they had any, are unknown. It proved quite fruitless to enquire for which groups or districts these men were chiefs. Tc'loklin was chief for the towns about the mouth of Sprague river (lobo'kstsoksi, taka'lmakeda, etc.); LEle'ks for the whole middle Williamson valley, which includes these towns, and for Klamath marsh as well. In other words, each was a chief so far as people of his subdivision chose to follow his leadership. Such is the separatism among the subdivisions that there can hardly be a tribal chief. How far any of the above functioned as such 56 op. Cit., 1:xli. Spier: Kkrm4th Ethnography must have been purely a matter of personality. Pat Kane had it that the principal chief is always one of the a'ukckni. This is by far the largest subdivision, hence chiefs in other divisions may well have less influence. Yet I believe he had the instance of IFle'ks in mind. He said further that the a'ukckni invariably issued the invitation to war and their chief selected the leader and received the booty. I doubt it. A chief is one who has acquired his position in war, with some background of spirit experiences, wealth, ability to talk, and a sugges- tion of hereditary interest; this beyond natural ability for leadership. As one informant summed up his qualifications: he must be brave and have property; he must have presence to meet and talk with anyone, must be competent to arrange matters without dispute, and must have the interest of all his people at heart. I have described above how one leader sought power to lead a party to war. Since power is implicit for undertaking all things difficult, it must be assumed to be part of every chief's make-up. Several anecdotes of Tci'lokin will make some of these qualities clear. Tci'lokin was a short man, deep-chested, powerful, and exceptionally hardy. It is related that he overtook a party camped in the deep snows atop the Cascades. Not being properly equipped, he lay beside a fire covered only with a single blanket. In the morning he was covered with frost but apparently had not suffered. Somewhere in the north, possibly Warm Springs, was a man who owned a big slave who was very much a bully. While the slave was absent Tci'lokin traded three horses for the master 's best horse. When the slave returned he inquired for the horse and sent a man to demand it. That man went twice but each time Tci'loliin sat quietly sewing and did not reply. Then the slave himself went out and demanded its return but Tci'lolkln paid no attention. When the slave went to untie it, Tci'lokin tripped him. The former struck at him with his hatchet but missed. Then Tci'loR!n wounded the other and he retreated. Tcautcau, a northerner, wanting Tci'lokin's daughter, offered horses and other valuables. The latter refused. He came to Tci'lokin at night and fought with him till daybreak. Tci'lokin clung to the other's hair despite the beating he was receiving. He came again next night and again they fought. This performance kept up through the whole winter. When spring came, Tcautcau acknowledged his defeat "I, Te.'lokin, kelo's; yes, Tci'lokin, you are fierce, " and they were again friends. Again in the north some northern shamans went into a sweat-lodge. They wanted a shaman among their Klamath visitors to accompany them. None was present so the Klamath insisted on Tc'lolin joining them. One after the other each northern shaman sprinkled water on the hot rocks as he sang his song, hoping to force the others out.57 Some overcome by heat and steam had to be dragged out. Once outside they wondered what had become of Tci'lokin; 57 This recalls the well-known test theme of western mythology. 1930] 37 38 University of California Publications in, Am. Arch. and Ethn. [Vol. 30 he was nowhere to be seen. They thought he must have died. After a long time he began to talk, throwing water on the rocks. When he was quite ready he came out and walked to his own people. He had bested the shamans. One informant completed the humorous tale of the lizard who killed five grizzlies with the comment, "HEe was braver than Tci'lokin."I Chiefs are wealthy. All informants gave the same term, loki, for rich man and chief, and when asked for the rich men named the chiefs. They own many slaves; other people may have slaves, but. many have none. Some chiefs have a number of wives. LEle'ks had seven at one time, not slaves, but one a Modoc, the others native. This, the largest marrige group known, argues considerable wealth. LIEle'ks was accus- tomed to make frequent trips to northern tribes with captives to exchange for horses. Everything points to the rise in wealth of these men as dating only from the comparatively recent. period of contact with the northern tribes. Up to that time at least shamans rivaled chiefs in wealth. A chief becomes such primarily by his leadership in war. When a group is gathered preparatory to war, chiefs address them and under their guidance war leaders, prominent, brave, and wealthy, are agreed on. Chiefs themselves are among the war leaders and no one alone captains the party. Chiefs make a. point of engaging prominent adversaries in the forefront of battle. In addition to their other qualities, it is assumed that they are bra.ve, unexcitable, and able to do without sleep. Under the circumstances there can hardly be inheritance of chief- tainship, largely because it is not. an office to be regulated by definite succession. A chief's relative may be chosen as war leader, but not of necessity. Since wealth figures, it may be that at least the possibility of succession is heritable, but a man who is personally disqualified will not be considered. Pat Kane's paterna.l grandfather was a chief at du'kwa; further vision experiences had made him a good gambler but no shaman. But his son did not succeed because he was not com- petent and, in addition poor, nor has Pat any pretensions. On the death of LEle'ks, his son Solomon was passed over as lacking caliber, and in reservation days Alan David, previously a minor figure, was recognized in his place. Their relationship is unknown. My inform- ants did not know whether the fathers of any of the chiefs named above had been chiefs before them. The Kilamath have no feeling of class strat.ification as on the Columbia river and northward. A chief is a leader and nothing more. Spier: Klamath Ethnography His wife and children are commoners and not known by distinctive terms. The emphasis among the chiefly functions is clearly on leadership in war and haranguing the people. Every morning before daylight he addresses them and again in the evening until the sun is well down. I am somewhat dubious that this procedure is as standardized as my informant had it. The address runs on fighting and acquiring spirit aid. He warns against hostile raiders; tells them to rise before day- break; tells the young people to seek visions. " If you get a spirit, you -cannot be killed,; you will uever fall sick; you will live to be old. " But he has no word as to where they shall hunt or fish or carry on other routine activities. A chief has no speaker as a shaman has. Such might have been expected here, for Paviotso, Wishram, and Nez Perce chiefs have. A chief is not looked to to preserve peace. Old men may interfere in quarrels, but these are no especial concern of a chief. Nevertheless the word of an influential chief might end a feud. Just prior to reser- vation days the practice was introduced of tieing up and lashing quarrelsome men at the word of a chief. This was brought from the north by the cousin of my informant Nancy. It has a decided un-aboriginal flavor. A chief is under no obligation to help his people pay blood money, nor to assist them when starving. These are matters for relatives alone; they may die for all he is concerned. SLAVES, TRADE, AND WEALTH Sla.ves are ca.ptives of war, some of whom are kept by the Klamath but the majority are carried to the Dalles to be sold. They seem to have been numerous. Ga.tschet quotes a statement of Taylor in 1859 tha.t in the raid on the Pit. river people two years previously fifty-six -of their women and children had been taken and sold on the Columbia river.58 Doubt has been cast both on the number of slaves and the frequency of the raids as described by early writers. Kroeber, for instance, scouts the idea of yearly raids on the Achomawi and sug- *gests that a few actual cases have been exaggerated.59 Whatever the *case be with regard to annual raids, I am certain that the evidences are too many and too circumstantial to permit of doubt that raids were 58 1:lx. 59 Handbook, 319. 1930] 39 40 U1niversity of CaUformia Pubcationm in Am. Arch. and Ethn. [Vol. 30 frequent and had slave-taking as a primary object. It is probable that the introduction of the horse stimulated predatory habits and made possible frequent trading visits to the Columbia river people. From them they adopted for themselves the practices of slave-holding on an enlarged scale and of measuring wealth in terms of slaves, and were encouraged to raid more frequently, and to carry their spoil to the river to exchange for the highly desired horses. Prior to the coming of the horse, we may suppose the very mild interest in holding captives found in any Basin or Californian tribe. Most slaves (loks) are from Pit river (Achomawi and Atsugewi); others are Northern Paiute and Shasta, with a few Upland Takelma from the Rogue river drainage. A few of these Takelma are known to have been made captive on a raid into Klamath territory. The Klamath sporadically raid the Pit river people in spring or summer when the latter are scattered. Young children, women, and even men are taken.?? Other plunder (du'ta) consists of blankets, bows and arrows, dentalia, and other beads. Captives are similarly taken in the course of feuds among the tribal subdivisions, and perhaps as freely sold. One informant had it that they are not sold but grow to man- hood as foster children and visit back and forth with their own people on the friendliest terms. This view implies that they are taken and cherished as recompense for a loss by murder. Slaves are commonly traded to individuals who make a practice of taking them to Warm Springs and the Dalles to sell again. Such was LEle'ks in his day, who, having got three or four together, would set out for the north to trade for horses and blankets. Some he would keep to treat much as his own relatives. It is quite likely that a slave's life is much like that of any poor Klamath. They commonly live in dome-shaped mat-lodges set up beside the family's earth-lodge, and there the women are visited by their masters. How far they are abused or suffer from uncertainty, I do not know. Gatschet records the escape of Pit river captives while their captors were hurrying back from a raid.6' When the agency was established by -the government all slaves were freed, but most elected to stay, and of those who returned to their own people, some came back. Evidently they had not suffered from too harsh treatment; at any rate, by my informant's account, those who had 60 Gatschet says adult men were not enslaved but killed if captured. The instance he cites bears him out (op. cit., 1:24, 25). 61 Op. cit., 1:20. Spier: Klamath -Ethn-ograpIy grown up among the Klamath, Klamath in speech and custom, were not. tempted to begin life anew elsewhere. Some, at least, had con- siderable prestige... Pit River Charley, for instance, .a freed slave, returned from his people in 1873 with novel rites which were engrafted on the Ghost dance then newly come to the Klamath.62 Trade is probably of no great. consequence within the tribe although it figures intertribally. Contacts were few and frequently unfriendly until after the coming of the whites. The exception is the.neighboring Modoc groups; others are too distant. Winters are too severe for travel and trade, but summers find the Warm Springs people in resi- dence with the Klamath. These similarly set out for Warm Springs and the Dalles when the grass begins to grow, camping there in the open during the summer, and building mat-covered structures of earth-lodge shape (wake'ploks). if they stay through the winter. Slaves, Pit River bows, and beads are taken there to trade for horses, blankets, buffalo skins, parfleches, beads (probably dentalium shells), .dried salmon, and lampreys.63 Two. sla.ve children are valued at five horses, several buffalo skins, and some.beads. This is.not far from the values recorded by Taylor in 1859; one woman was worth five or six horses, a boy one horse.64 The. Molala are also visitors and are met on the headwaters of the Rogue west of Crater lake. From them buckskins a-re obtained for pond-lily seed and beads. Relations with the Shasta have -not always been unfriendly; some of the Klamath Falls group and Modoc married with them, possibly only since the white advent. The nearest Shasta settlements at Shovel and Jennie creeks can be reached afoot several days' journey down the Klamath. Laik''elmi is a jointly occupied fishing site somewhat upstream but below the' Klamath canyon. Skins and skin blankets are traded for Shasta beads. Dixon' states tha.t Shasta trade with the Klamath is slight.65 In the period 1860-70 the few Shasta remaining in their territory congregated on the lower Shasta river and near the frontier town, Yreka. They were reaehed -by cros#ing the Black Mountain range. At that timse. this was a much mixed group, principally half-breeds. My informant Tom Lang, taken there as a youngster, remarked how he was afraid of their highly tattooed faces. Intercourse and marriage went on freely. with -the Modoc. They were visited on Tule and Lower Klama.th lakes, a.nd Joined for the 62 Spier, Ghost Dance of 1870, 45, 51. - 63 Cf. Gatschet, 1:93; 2:436. 64 Op. Cit., 1 :lx. 65 The Shasta, 436. 19303 41 42 University of California Publication in Am. Arch. and Ethn. [Vol. 30 fishing on Lost river near Olene. A shallow bowl-shaped twined basket, called k!we'lu, is obtained from the Modoc, who in turn trade it from Pit river. There is no direct trade with the latter people. Horses were first acquired in any number about 1840. They had been previously acquainted with them in the hands of whites who crossed their country about this time; dates are uncertain. Ogden in 1826 remarked on a single horse seen as though it was a rarity among them.66 The first sources were Warm Springs and perhaps some group immediately east of that place. These were obtained by chiefs; others were too poor in slaves to buy them. The name given the new animal follows the common habit, wa'tcag from watc, dog. Saddles (pa'olas) were first made of antelope hide. Here, as in the Plains, horses served as a stimulator of contacts, especially with the Columbia river peoples. As I have remarked, their introduction seems responsible for the rapid increase in wealth, the flow of northern ideas concerning social stratification, importance of wealth and of slave- holding, Plains costume, and those predatory habits which furnished the withal to maintain the new status. Buffalo (yo'ho) skins are obtained at second hand; the Klamath know that their origin is east of the Umatilla, that is, in Idaho. Buffalo never ranged into this country, although they reached the Harney Lake district. But very few skins ever make their way into Klamath hands; only men with many slaves can afford them. They are wanted to cover a gambler's hands in the stick game, an oppor- tunity for ostentation. Parfleches (sko'psoli, folded up) are traded from Warm Springs people and Sahaptins beyond them on the Columbia. They are not made by Klamath. Incidentally, many older people have fluent command of Chinook jargon. They, or somewhat doubtfully the Modoc, are the southern- most people with whom this trade jargon is current. Kroeber remarks on the fragments in use by Yuki and Shasta.67 Naturally enough, it served a purpose with the Columbia river peoples alone and was of but small value with the whites coming from California. It is interesting to know what constituted wealth among the Kila- math, but with the old life now gone it is difficult to get direct evi- dence. However, certain things alone were spoken of when describing 66 Elliott, Ogden Journals, 210. 67 Handbook, 308, 319. 13]Spier: Kla4nath Ethnography wealthy individuals. In approximate frequency of mention are slaves, horses, beads-and not always denfalium-food, archers' equipment, furs.and hides, especially elk hides, Plains type garments, armor, large houses, buffalo skins, canoes. It is just as interesting to observe wha.t things were not mentioned as -valuable, or so to a lesser degree: baskets, nets, mortars, etc. In a word, articles of value in foreign commerce, articles of dress and adornment, of war, ritual, and sustenance, have value; domestic utensils have not. Some of the phrases describing wealthy men are illuminating. A war leader is chosen in part for his riches: he would own five quivers (five is the pattern number); he goes into battle with a quiver under eaeh arm and one on his back. If his bowstring is cut he has another. A rich man has fur blankets, coyote and wolf pelts. Dentalium shell beads do not function as money here as on the lower Klamath river. They. are strung in fathom or half-fathom lengths, but neither length of string nor size of shell is standardized. The longer individual shells are more valuable but no one is known to haggle over the precise length as Yurok and their neighbors do. MARRIAGE A marriage payment is a social obligation. It is not a negotiated transaction as with the Yurok, Hupa, and Karok, for example, where indeed the social status of the family is measured in exact accord to the sum paid for the bride. Here failure to pay, which can be only a confession of inability, leads only to a lowering in social esteem. It is consonant with this that the bride is in no sense a chattel; she may leave her husband at will, and she certainly cannot be disposed of as a possession. Bride purchase is as inapplicable a -term as can be found for the Klamath practice. In the interchange of presents which fol- lows marriage the advantage lies most frequently with the husband's family. Payment constitutes an obligation, a seal of respectability. This is the well-nigh universal situation in North America if payments enter at all; it is a far cry from the commercial, contractual bride purchase of Africa, for example. The peculiarity of the Klamath is that failure to pay constitutes a social stigma. This is clearly explic- able as a function of their geographic position, lying between the tribes to the west, with their clear-cut monetary evaluations of social relations, and.the tribes of the Great Basin to the east, utterly devoid 1930] 43 44 Uni.versity of California Pub1ications ':in Am. Arch. an4 Ethn. [Vol. 30 of any such sentiments. It also seems that, the reciprocal. visits by: which the marriage is consummated, with their exchange of presents,. is a reflection at a distance of the Northwest Coast potlatch, perhaps its. most southerly extension. Beyond the ability to pay, the choice of a mate is restricted only by kinship. No relative of any degree may be married; the children, they contend, will not survive. Common residence is for example no bar to. marriage. Step-brothers and sisters, being unrelated, are free to marry even though they have grown up together., Several men now living here married their step-sisters. Sibling terms are not used between step-siblings, though they have half-brothers and sisters in common. There is no village exogamy despite the relationship of most co- villagers. Yet there is a great deal of shifting residence especially among the several settlements of each division. Marriage within or without the tribelet is immaterial. Such evidence as exists however suggests that the. tendency is for marriage within the subdivision. The data are few but summarize in this fashion: Husband Wife a'ukckni du'kwakni gu'mb6tkni iu"laloakni Modoc Molala l'ukckni ........ 8 1 ...1 1 1 du'kwakni ...... ...... 1 ...... ...... ...... gu'm b6tkni 1...... ...... ....... ...... 2 iu"'lalo,qkni ...... ...... 2 1 ...... Of the a'ukekni marriages among themselves, two above were with co-villagers. I have no data for other endogamous marriages. Explicitly the data are the following: Pat Kane's father and his paternal grandparents were du'kwakni, his mother and her parents gu'mbo5tkni. These tribelets are more closely linked than with the others. His father lived at the time of his marriage at wud6'kAn, a gu'mb6t village north of Klamath lake; his mother with her people lived at sle'tsksi just across Seven Mile creek. Pat himself was born and grew up at mo'gitqkuinks, the du'kwa village at the mouth of Williamson river. His father 's sisters married a man at iu"l lalo,jE on Klamath faIls. Tom Lang's father was white; his mother lived at iu" laloqn. Both her parents were also iu" lalotqkni; her mother was related to Modoc and Shasta. Nancy Phillips is a'ukckni, born at k 'taidi on Williamson river. Her father was a'ukckni although he lived at du'kwa. His mother was of the du'kwakni. Nancy 's mother was Wlukckni of k 'taidi village; her parents in turn both A'ukckni, the father from k'taidi also, the mother from gla-'tspak 'is on the opposite river bank. The chief LEle'ks had seven co-wives; four, A'ukckni, one each, du'kwakni and Modoc, and one from ne" luksi hence probably iu" laloiykni. Two of the Spier: Klamath Ethnography a'ukckni were sisters from his own settlement on Williamson river above bzuikse'was, a third was from that town, and the fourth from gla'tspak 'is, a mile or so down river. His daughter by the Modoc wife was counted a'ukckni; while living at k'taidi she married an a'ukckni of Yainax. A gu'mb6t man, Wokau'klalls, married two Molala sisters. Girls usually marry within a year or so after puberty, which is marked, at least among the well-to-do, by a big dance. Betrothals may have been arranged while the couple were yet pre-adolescent, presents having been exchanged by the families. They are expected to carry out the agreement after the girl's dance, payments being made again. Should one of them die another match may be arranged with the same family, but not of necessity. Poor and rich alike arrange such betrothals. How far great disparity of age occurs I do not know, but it does happen that an older man marries a. young girl. A shaman for instance may be given a daughter by a man too poor to reward him otherwise. But it must not be thought that there is no choice. On the con- trary, the majority of marriages follow the desire of the couples. Elopements take place and although the levirate is the rule, much depends on the widow's willingness. Yet a girl's consent is not neces- sary. Despite her protests she may be carried to her husband's home. If she runs home she is almost invariably taken back again. That she would be driven to suicide is doubtful; one informant never heard of such a thing; another stated she would hang herself and in this event her parents had to return the gifts of the groom. An orphan girl of du'kwa lived with her mother 's brother. She was married against her will to a man of k'taidi. She clung to the ladder of the earth- lodge but they pulled her away and dragged her up-river to k 'taidi. Next morning she ran home and never went back. Nothing had been paid for her at the time but she was given in return for valuables contributed to the cremation of a relative of hers some time before by her husband's relatives. Immediately after this she married a man at Klamath marsh; whether payment was then made is not remembered. Elopements occur but payment is so much insisted on as a sign of respectability that parents guard their daughters with great care. This is at least true where the family wealth makes a suitable payment imperative. It is unthinkable that a chief's daughter should be per- mitted to enter an unceremonious, clandestine union. The poor and forward have little to lose. Modest demeanor is expected of a girl. 1930] 45 46 University of Caoifornia Publications in Am. Arch. and Ethn. [Vol. 30 She dances with downcast eyes in her adolescent rite; she sits apart and does not talk with male visitors at home. A man seen prowling about the camp is shouted away. A girl who does not wish to be molested will go over to her mother to sleep. A father will whip a daughter who has clandestine visitors, and his son for misconduct. No obligations may be assumed for which no payment is made. Despite the fact that elopements are frowned on, clandestine visits and elopements are frequent. The girl's parents may follow a run- away cduple and force her to return with them. It is usual however for the families to accept the fact and make gifts in proper form. The most lavish marriage display ever known was that of Lzle'ks' only daughter, Go'soluilb. They were living at taka'lmakeda during the fishing season when Ski'liils of Yainax eloped with her. They fled riding double; the horse fell with them while fording the river, but they escaped to Yainax. Soon after the girl returned alone. It took a month to get ready to conduct a proper marriage. LEle'ks made the trip across the mountains to Jacksonville to get store goods. A great many people went with them to Yainax, the majority taking gifts of horses, blankets, clothing, or anything they had. One man with nothing better stripped himself of his only shirt. It must be remembered that such Plains shirts were then newly fashionable and hence valuable. They gave the horses they rode there, hoping for reciprocal gifts on which to return. Horses to a total of ten were given the groom's people. They gave horses and other gifts in return. Payment having been made the affair was considered as honorably ended, and as though the groom had sent an emissary in proper order. Marriage without payment brings a slur on the children to be cast in the teeth by quarreling women. " Your children are worth nothing; you were not bought." Children taunt each other in the same way. They are respected in proportion to the amount changing hands at their parents' marriage, but this does not mean fhat a man's life is assessed at this sum. Nor is there any suggestion here that husband and wife are financially accountable to each other for injuries to their children. Marriage payment is prescribed not alone because it fixes social status in a general way but because, as one informant phrased it, it is decidedly unlucky not to make payment. So it is, for the couple have made an unsuccessful start of their social career. The amount is not prescribed but varies with the wealth of the individuals. A chief gives a slave and more; a poor man buys with food. A chief will not give his daughter for nothing; the groom must continue to bring gifts even after marriage. The valuables custom- arily exchanged are slaves, horses, blankets, beads, food (pond-lily seed and roots), elk and other hides, and strings of beads. Pat Kane's father gave five fathoms of glass trade beads. It is important to Spier: KKlamath Ethnography recognize that the marriage is consummated by a series of gift exchanges and inasmuch as it was said that in the long run the man and his relatives have the advantage, it is clear that this is no outright bride purchase. Giving a girl in marriage is sometimes a means of paying off a social obligation. Considerable property is destroyed at a cremation. This is normally furnished by the relatives of the deceased but in order to do honor a non-relative may contribute beads and other valuables to be consumed on the pyre. The recipient family is under obligation to reciprocate. Should one of the contributing family later want a wife from among them, they must acquiesce. The groom must make payment at the time of marriage, however. That no payment had been made in the case cited above undoubtedly explains why the runaway bride was not forced to return to her husband. Possibly the sum must be repaid in the event of separation, at least if the wife leaves the husband. Probably the time element is impor- tant; a marriage soon terminated means restoring the status quo; in one of long duration this is impossible. A man sends a friend to sound the girl's people on their willing- ness and the price expected. This man is chosen for his forwardness; he may be a relative but never a brother or father. He carries a bundle of sticks representing the gifts offered, which if successful he leaves, though they are not used further. The girl's father consults his wife before agreeing. Next day or as soon as the girl's people can get ready their gifts and contributions of food they take their daughter to the groom.68 The groom's relatives have gathered mean- while, collecting their gifts in the groom's home. The exchange takes place within the lodge, each party putting down its gifts beside the other seated opposite. The gifts from the bride's side, at least, com- monly include clothing (presumably the Plains type), pond-lily seed, and roots of various sorts. The payment now made on behalf of the groom includes the promised articles. Speeches are made by the parents of both, admonishing them to live in harmony. Payment for a bride is called woko'wi, the reciprocal gifts, i'wi. It is difficult to say just how formal are the visits that follow marriage; at least one or two are usual, if not prescribed for the well- to-do among whom further payments by the groom are expected. 68 In a Modoe (or Klamath?) song recorded by Gatschet (Songs of the Modoc Indians, 27, 29) a bride is carried on her brother's back. As Gatschet points out, carrying the bride is a custom of northerly tribes. 1930] 47 48 University of California Publications in Am. Arch. and Ethn. [Vol. 30 After a month of marriage the couple visits the bride's former home, accompanied by the groom's relatives, for a further exchange of gifts. Soon after this, the girl's father and his family reciprocate. Beyond this it is a matter of mutual solicitude; undoubtedly the visits grow less frequent as the passing years bring diverging interests. Later visits are of some duration, the couple even staying through the winter. Marriage is normally patrilocal, but if a man is poor, or has no close relatives, he may take up residence with his wife's people. When a young wife is about to have her first child the couple return to her mother to stay until she is strong again. This must be the circum- stance of temporary matrilocal residehee as it ordinarily occurs else- where. First marriages among primitive peoples usually follow closely on a girl's nubility. The need for advice and care during this trying period of pregnancy and delivery, the natural solicitude of the mother, are surely the reasons for the frequency with which residence is tem- porarily matrilocal. It will be observed that this is formally pre- scribed as the proper procedure almost solely in cases of first marriage. There is no need to search for a more esoteric reason such as the older writers had, matrilinear grouping, maternal authority, and the demand that the young husband prove himself to the wife's family. Among the Klamath this is the rule only on the birth of a first child. A series of terms describes the facts of residence. A man and his wife in matrilocal residence are non-vocatively wuko'b6ks and ska'boks respectively.69 A woman with her husband's people is t!e'niwas.70 On separation a woman who returns to her former home is ko'stbombli, a man returning to his, wuko'tbombli. I was told at least that a child's spouse is treated like an own child, that parent and child terms are used between them for there are no parent-in-law taboos. This is the usage of intimacy; there are specific terms for parents-in-law. There is the familiar Indian usage of a term between the parents of a couple, "my child 's parent-in-law. " Marriages occur any time during the year. There is no evidence suggesting their greater frequency during any season, nor any mating period7' 69 The latter term was given me at another time as meaning a newly married woman (probably the form ska'b6kak was intended). This comes to the same thing. These two words apparently correspond to Gifford's wapakc and kapakc, earlier and later co-wife (Californian Kinship Terminologies, 42). Gifford 's record is doubtless due to a misunderstanding; I enquired about these words and found that there are no terms for particular co-wives. 70 Gatschet (2:698) gives teina, a young woman. The meanings are not incompatible but I believe mine to be correct. 71 As among the Yurok (Kroeber, Handbook, 44). Spier: Klamnath Ethnography The number of plural wives depends on wealth; hence chiefs have more than others. From my notes I glean two instances of two wives, another of three. The wives in both bigamous cases were sisters. The greatest number known was the case of the chief LEle'ks, who had at one time seven wives, two of them sisters.72 I have described their derivations above. These women bore him many children. Whether the wives are sisters is a matter of complete indifference. To have two wives is called la'bwewa'nsg'tkt (lit., two females), three Enda'nwe- wa'nsgitkt, and so on. Collectively the wives may be referred to as slpkska'ltk; they call each other p'tciki'p. Plural wives are paid for, but whether there is reciprocal visiting I do not know. Slaves are also taken to wife. Polyandry is unthinkable to the Klamath; it would be adulterous. Barrenness is as usual a reason for taking a second wife, but not always. Pat 's maternal uncle had a barren wife. He kept her throughout his life and never took another. The case of three wives noted above was one in which there were no children; the children of this family always died in infancy, I was told. Whether this implied that two of the three were barren, I do not know. One word, ts'e'do, means a sterile man or woman. The sororate is not obligatory on the death of a wife, nor is there any notion that a girl must marry an older sister's husband when she comes of age.73 It is thought better to marry a deceased wife's sister, and from her family's point of view it is evidence that they care for the husband when they provide another daughter. There is however no feeling that they are under compulsion to make good the loss he sustained because of his marriage payments. A bereaved per- son waits a year or more before remarrying. The second wife has to be paid for like the first, but the amount is less. Again this depends on the man's affluence. The sororate is called splino'la, the sister, spl'nuwlp, i.e., sibling-in-law when the spouse is dead (cf. splnu', to cremate, spl'nuks, cremation place). The levirate is customary but not obligatory. It is desirable to keep a woman in the family, especially one with children. This is perhaps nothing more than an expression of household solidarity, for 72 Gatschet states ten, but I am certain there were not more than seven, at least at one time (2:126). 73 There is just a suggestion. When asked for the kinship term used by a woman for her sister 's husband, my informant gave the term for husband, but somewhat in confusion corrected herself with the woman 's term for brother- in-law. 1930] 49 50 University of California PubUctions in Aim. Arch. and Ethn. [Vol. 30 a woman with young children is as likely as not to be a member of a joint house group with her husband's immediate kin. We may also wonder if she would not ordinarily continue to reside with them even though she does not remarry in the group. It might be illuminating to know how residence with or apart from this family affects the practice. In default of a husband's brother, the widow should marry a nephew, either fraternal or sororal, or another of his kinsmen. A payment is made but not so large an amount as at her first marriage. Since only close relatives are paid, a widow may not be paid for if she has none. Much depends on the woman's willingness; she may not care to be a plural wife. Nevertheless, it is in some degree compul- sory, for if she should marry another, her deceased husband 's brothers may kill the man. While their objection is voiced loudly enough, they do not announce their purpose but await a favorable opportunity to surprise him. Quarrels and adultery lead to separation, which is frequent; there seems no compulsion to endure an unbearable situation. Incom- patibility and jealousy (suki'kssle'a) bring on quarrels as frequently as they do in any other group of human beings. Typical causes of domestic friction cited to me are adultery, as with a slave woman of the household, a woman's continual anger with her hus- band, or her demand for sole attention in a polygynous household. It is always the wife who leaves. My informants could imagine no situation short of adultery which would lead a man to send his wife away. Nevertheless he can force her to act. A good, stout-hearted woman stays, so they said, no matter how badly she is treated; only a mean-spirited one flies home. Inevitably the situation is complicated by children. A wife may take all her children with her or only the little ones. This seems a question of who is at fault, for a woman taken in adultery may choose to submit to beatings by her husband rather than be separated from her children. If a wife leaves of her own volition and for just cause, her husband must pay again to have her return, must assuage the feelings of her family by gifts. If he does not desire her, he may ask for the return of the marriage gifts but they are not automatically forthcoming. If the wife is at fault he may choose to demand the return of the bride price. Its return is called seldonga. If a man is impotent his wife may leave him for another. But he cannot cast his wife aside because she is barren; he adds another to the household. Spier: Klamath Ethnography In the matter of adultery it is quite clear that a man has no recourse but to kill the lover; demanding payment for the wrong is unthinkable. lIe waits his opportunity to ambush the other. A man of no consequence may have to put up with the situation, but at least he would never think of turning to a chief for help.74 The husband may send his guilty wife away, may beat or even kill her. Sometimes he will send someone to demand the return of the bride price, and her people return what still remains. The woman is then free to marry her lover who must pay for her in regular fashion. Nothing goes to the former husband. If a woman finds her husband unfaithful, she may beat (sato'mdoka) the other woman; her final recourse is to return home. TRANSVESTITES Transvestites or berdaches (tw!1nnae'k) are found among the Klamath as in all probability among all other North American tribes. These are men and women who for reasons that remain obscure take on the dress and habits of the opposite sex. Their number is small. Five men who lived as women were cited by an informant, two women who lived as men, and others are known. One of the former was from the Dalles. This is a very minor fraction in a population num- bering upward of two thousand. Kroeber has assumed that such individuals are invariably psychologically abnormal, homosexual ;75 I am not sure. The case of LEle'ks below rather argues against it. At any rate their abnormality is socially canalized: they are permitted to live as they desire despite the distaste of the normal Klamath for the practice, and the scorn and taunting to which he subjects them. It is to the point to note that the men who turned women adopted woman's dress, pursuits, personal habits, and speech, and in one case at least married. Neither of the two women affected male garb; one married and made some pretense of men's habits, the other is rather simply a case of an irregular sex life. The cases of two of the trans- formed men which follow seem to turn on adolescent experiences. While these are phrased with relation to esoteric situations, they are more adequately explained as states of psychological disturbance rationalized in these terms. A third of the men-women turned shaman, but I do not know that this followed directly from the transformation. 74 Punishment of adultery and other torts by a chief, as described by Gatschet (1:58 et seq.), are post-reservation customs. 75 Handbook, 46, and elsewhere. 1930] 51 52 University of CaUfornkia Publication in Am. Arch. and Ethn. [Vol. 30 As both sexes are shamans there was no need for the transformation had this been the -goal. M-uksamse'lapli, also called White Sindey, was the berdache sha- man. He tried'to laugh like a woman. He was twice married, first to a Molala man named Tci'ptci, later to a Klamath, Tc!o'moks. The latter is known to me as a normal old man who has since raised a'family of his own. White Sindey made considerable pretense to shamanistic powers, but was rather scouted. He performed and reacted like an ordinary shaman. One of my inform'ants now in his sixties had been a berdache in his youth. I had observed his general unmasculine behavior but it was not until my visit of the second year that another informant gave the reason. He lives alone, takes pride in his cooking and the care of his house, weaves baskets for sale as women do, and still wears a basket hat, which no Klamath man now will. He is fleshy, slow, and dis- tinctly feminine in gait and gestures, and speaks with the higher pitch of women s voices. Yet his transvestite state was given up nearly half a century ago. He is shrewd enough, though imild mannered, yet I confess that I suspected no abnormality. When about sixteen years old he was at the Dalles where an old woman shaman of the place took a fancy to him. He however was afraid that she would harm him so he put on woman's dress. He wore' his hair long, prepared lily seeds, wove mats and baskets; all typical women's pursuits. He has never married. After learning his history I brought up the subject of mar- riage with him. He claimed that he had once been married, his wife having died twenty years ago, and in answer to the question why he had not remarried replied: 'S'ome men like it that way.'' The most striking case is that/of the- chief LiEle'ks. As an ado- lescent he wore women's dress and performed the appropriate tasks. One day after returning from the lily seed harvest he lay .down to sleep. -They placed a bow and arrow and a punt pole by his side, the latter a symbol of women's work. T'hey stood ove& him and shoutted. As he jumped to his feet he grasped. the bow, thus determin- ing his future. He ran off to Mount Scott '(t!u'msumEni),' staying through the night on the mountain in quest of a supernattral expori- ence. As he was returning he smelled the 'sweat-y odor of 'running horses. He left off,women's dress and actions from that! time. He became lucky in every way, acquired wealth, especially in horses, and prestige, and became the foremost 'chief the tKiamath have, known. His Spier: Klamath Ethnography subsequent career seems quite normal unless exception be made of his seven wives! A woman named Co'pak lived like a man although she retained women's dress. She married a woman who lived with her a long time and finally died. She observed the usual mourning, wearing a bark belt as a man does at this time to prevent the back from growing bowed. She tried to talk like a man and invariably referred&to herself as one. Another woman still living has had relations with both women and men. She never adopted men's garb but told them that she was a man. She is today a common prostitute, an abnormal, irascible person. Those she lived with were all older women, not young girls. This practice is known as sawa'llnaa, to live as partners. They say of such: they have lots of partners, friends (snewa'ets doma' sawa' l'naa'sgltkt). Other Klamath of any standing have always avoided her and her women partners. She was never married to a man. HOUSE GROUPS It is now difficult to get evidence on household composition. Houses vary in size, to judge by the existing pits, from twelve feet in diameter to as much as thirty-five feet. The larger houses must have held a number of families and as many as ten or twelve persons. One described below had five families, numbering twenty. The majority of the inmates of a house can usually trace some relationship. This is also true of most of the inhabitants of a settle- ment. For example, the village mo'giqkuinks had half a dozen earth- lodges, of which three held the families of -two brothers and their father's male cousin. Residence is normally patrilocal. In the cases which follow the connecting relatives are for the most part related in the first degree. An earth-lodge on the lower Williamson river built about 1870, and measuring thirty feet in diameter, contained five men, their wives and children, whose respective portions of the house are shown in figure 4. This was known as the home of the shaman Boke's; the others were related to his wife. 1. Boke's, wife, and two children. 2. Lek1!o'c, wife and two children. 3. Wokau'1!alis, a widower, whose two wives had been Molala sisters.76 78Their Molala names were Slowa' and A'li'nk. 1930] 53 54 University of California PubUmationw in Am. Arch. and Ethn. [Vol. 30 4. Lo'l6klis, wife, and seven children. 5. Tc!ontc!on and his wife (childless). The wives of Boke's (1) and Lo'loklis (4) were sisters; Tc!6ntc!6n (5) was their brother. Lcik!o'c and Wokau'k!alls were distant cousins& of this trio. In front of this lodge stood another, a mat-covered house (wu'ke'-. plo'ks), housing the mother aAd three sisters of Wolkau'1!alls. This. was only t.en feet distant, measured about fifteen feet in diameter, and. a.lso had its entrance, this time at ground level, on the southeastern side facing the river. The inmates of both houses were gu'mbotkni from the western side of Klamath lake with the exception of Lo'lo6klisl a du'kwa. man, who alone properly belonged in this locality. STORMGED \ ST0RPR3r._.STP Fig. 4. Plan of an earth-lodge, showing the areas assigned to the beds of five families. One earth-lodge in the same district was known as wewalaeksumlo'- ltmalo'ks, old woman's earth-lodge. Its inmates were an old woman, her two daughters each in early life, the younger with three children, and an elderly woman, first cousin to the daughters, hence called older sister by them. The house was built especially for them. Data on the arrangement of polygynous households are of the slightest. It is clear that all the wives occupy one house with their husband. Their beds are separate. This is true where their number is few, but in the case of LEle'ks' seven wives, these were housed in two lodges. In one family of a man and three childless wives only one earth-lodge was used. Spier: Klamath Ethnography BIRTH The birth of a first child is an especially significant occasion. If it is the mother's first it will bring strong winds and thunder.77 The Kla.math also hold that in at least one other crisis, on the death and cremation of a shaman, the wind will rage. For the father the coming of the firstborn is one of those occasions on which to seek power in the mountains. Birth (wia'sla) must always ta.ke place outside the dwelling in a dome-shaped lodge especially prepared by the mother of the expectant woman, her mother's sister, or cousin. This is not the menstrual lodge, for this is used only by women of a shaman's household. This pro- scription of delivery within the dwelling is shared by other tribes of this latitude (Shasta, Northwestern Maidu, Lemhi, and Wind River Shoshoni) but is unknown to the south. A woman's first delivery is attended by four or five old women well-known for their skill as midwives and hence not of necessity her relatives. They are sometimes paid a fathom of beads. Men are excluded from the lodge. It may be that no man will chance con- tamina.tion. I do not know whether shamans are called in difficult cases. It would seem likely, since they ha.ve some genuine medical practice, were it not for this restriction. The procedures surrounding parturition are almost wholly prac- tical and physiologically well founded. The attendants press on the woman's abdomen and lift her repeatedly to hasten delivery. She lies on her back on a bed of grass with a soft pad (coyote or raccoon skin, or duck feathers) arranged to receive the baby. The umbilical cord is cut with an obsidian knife and the stump tied with the mother's hair. If anything else is used for the tie the child will con- tinually cry. The afterbirth (yepe.no'q ?) is deeply buried so that no animal may devour it and make the woman barren. Immediately after birth she is seated in the lodge on a bed of hot stones, covered with lea.ves. On this she must remain blanket-covered for five days. She drinks warm water freely; she may not eat meat nor fish, and must use a scratching stick, just as during catamenia and mourning. At the birth of a first child both parents wear sage-brush bark belts for 77 It is also the view of the Wishram that the weather is stormy or hot at the birth of a child, according to my information. 1930] 55 56 University of Caoiforni PubUcations in Am. Arch. and Ethn. [Vol. 30 five days. This is the belt also worn in mourning for a child and by a girl at puberty. I was not told of restrictions on the father's conduct, but if he is seeking a vision at this time he fasts of necessity for some days.78 The attending women pinch up the baby's nose and smooth its features to make it handsome. They are especially careful with a boy to straighten and massage his legs so that he may become a good runner.79 A tight belt of buckskin is put on a girl baby so that she will not become stout. The baby is put to breast immediately. Expect- ant mothers massage the breasts and squeeze out the first milk. The stump of cord falling off after five days, the navel is greased in order to help it heal. The stump, wrapped in buckskin, is put away by the parents with those of other children. Some precaution is taken to preserve it, at least during the early years, else the baby will die, but it is certain to disappear by the time he is grown. Should the baby die the cord is replaced to be burned with the body. The several references to five-day periods, and we may add the five- day monthly seclusion of -women, deserve a word of comment. To be sure there is some basis in fact for this; menstruation, post-natal placental flow and hemorrhage, and the sloughing of the cord do com- monly occupy some such period. However, in consideration of indi- vidual variation, it is clear that the stated five is but another applica- tion of the Klamath pattern number. There is no formal celebration at the birth of a child, as among the Wishram. Some other practices pertaining to childhood may also be noted here. Deciduous teeth are pulled out by the mother as they loosen so that the permanent teeth will erupt in their proper places. This has actually such an effect as well as stimulating the growth of the second dentition. There are no customs regarding these teeth; they are thrown away. A little girl's hair is made to grow long and thick by singeing it with a lighted splinter of a lightning-struck tree. This is not done for boys. There is no special attitude toward twins (Lob'als) so far as I could discover. This is noteworthy because of the fear of twins mani- 78 Modoc parents abstain from meat for ten days after the birth, eating only roots. They sweat for five days and then throw away their garments (Gatsehet, 1:91). 79 The Modoe method is described by Gatschet (Zoo. cit.). Spier: KlamathI Ethnography fested by several tribes of northern California and in view of their association with the weather and salmon on the Northwest coast. Neither of these attitudes appears among the Klamath. Nor is there any sympathetic bond between -twins, -nor is one of them destroyed at birth.' Informants had never heard of tripl-ets although they must occur in the usual biological proportions. 'Monsters are said always to be born dead-perhaps the midwives see to that. Like the stillborn, they are said to receive the universal cremation, but it is more likely that they are secretly disposed of, like abortions. The child of an adulterous union, if not palmed off on the husband, is aborted by pressing the abdomen heavily on a rock. Naturally this takes place in the seclusion of the brush. No medicines are in use for abortion. To abort is called hiso'koga; infanticide, siu'ga. How much of a disgrace barrenness is I do not know, but it does not lead to separation. To aid conception women grind a white shell80 (lo'ktas) and drink it in water. Men do the same for impotence. A small dentalium shell (t'u'tas) is also used. Klamath cradles show an interesting cultural amalgamation. The cradle for the newborn is the northern California sitting type; the carrying board for later use is identical with the oval wooden forms of the northern Basin.81 Through western Oregon and Washington, along the line of contact of the Basin board with the wooden trough cradle of the Northwest Coast, both forms are used or else a modifica- tion of one toward the form of the other. The basketry cradle (kwa'ls) for the first month or two of the baby's life is a trough with one rounded end on which the child sts. It is made by the mother of the expectant woman; woven of willow twigs, it measures eighteen inches in length. Its purpose is said to be to pre- vent the infant scratching its navel, with consequent illness, and to support its weak back. Its usefulness ended, it is thrown away. The carrying board (swens) is made by the baby's father or a near kinsman. It is a flat board of oval outline, carved from a single piece of wood. The lower end is rather pointed so that it can be stuck into the ground while the mother is occupied. An unexpected feature is that the sex of the infant is distinguishable from the shape of the 80 This is flat, two inches long, and spotted. It may have been got from the whites. 81 Barrett has it that the first cradle is of soft tules, followed by the sitting type of more rigid materials, with occasional use of the board (Material Culture, 257, pl. 14). 1'9307] 57 58 University of Calfornia Publication& in Am. Arch. and Ethn. [Vol. 30 cradle (fig. 5). Those for boys are relatively narrow and pointed, and with only one handle at the top, carved in one piece with the board. The girl's cradle is very wide, so that she will grow up broad, and is provided with two handles. Sex is distinguished in the cradles of some eastern California, Nevada, and Arizona peoples (Yokuts, Mono, Washo, Paviotso, Mohave, Maricopa) where it appears in the decoration of the hood. The angular board cradle of the Wishram and their neighbors is not. used here. a b c Fig. 5. Cradle boards; b, shape of boy's cradle; c, shape of girl's cradle. (Drawn from description.) The arch (tsi'dlls) is a single willow stick tied to holes on ea.ch edge of the board at its widest part. It is held rigid by two strings to a hole in the handle and two others to one near the point. There is no rest for the feet as in some cradles of this type, but a rest just below the buttocks permits the baby to half sit at times. This is of tules, possibly twined, placed transversely and in such a way that it can be turned down on the board and built over with padding so that the baby will lie flat. It is called sloteta's. There is no buckskin cover to the board; this is credited by my informants to the MIodoc, Northern Paiute, and the Upland Takelma. Instead, soft tule bark rubbed up fine is heaped over the infant, who is wrapped in a fur by the wealthy. The single lacing string is drawn back and forth through the five or Spier: Klamacth Ethnography six short loops on each side. A carrying band goes over the mother's forehead. Frontal flattening is customary but not universal. No reason is given. Hrdlieka was informed "that the Klamath regard a long head, i.e., a non-deformed head, with derision. They say it is slave-like, that their slaves had such, and that a man with such a head is not fit to be a great man in the tribe. Deformed heads are called 'good heads. ' 182 A buckskin sack, somewhat larger than the forehead, stuffed with dry cat-tail down, is tied to holes in the cradle. Hrdli-cka was told that this; was filled with seeds, such as water-lily seeds. This lies across the forehead but is not tied so tightly that the child cannot turn his head. Pressure is begun at birth 'and continued as long as the child continues in the cradle. According to my informant's representation the Molala flatten foreheads, sonme Pit River Indians do, while the Upland Takelma do not practice deformation of any sort. The cradle receives ornaments symbolic of the child 's future, in the shape of a tiny digging stick for a girl and a bow and arrow for a boy, tied to the edge of the board. Possibly they have more than orna- mental significance. These are no longer than the finger; the single arrow is as long as the bow. These are tied to the proper side of the board; the bow and arrow to the child's left, for a man holds his bow in that hand, the root digger to the right. Use of the cradle continues until the child can walk well. It is not discarded but kept for the use of another baby, unless the child dies, when it is burned with his body. NAMES The name is usually given in early infancy; always by the time the child talks. It is the choice of the parents, given without ceremony. Sometimes it is changed in later life, but there is no set occasion for this. Seemingly names change less frequently than their nickname character would give opportunity for. Perhaps these are always addi- tional names, for I learned of several men known by more than one name; perhaps this holds as true for women as men. But it was said that the first name (hiewa'ska) is the most important. There is considerable reluctance to the use of names, kinship terms being -substituted. This does not mean that there is a taboo; merely 82 Head Deformation Among the Klamath. 1930] 59 60 University of Caoifornia Publications in Am. Arch. and Ethn. [Vol. 30 they do not like to name their own relatives now dead nor tell. each other's appellations. On the other hand there was no hesitancy where reference was to things of the past., Children do not bear the names of dead relatives. As with most Indians,; naming another's dead relative is the common curse. Most na.mes are of the nickname type, given for personal char- acteristics such as. appearance,. gait, actions, habits. A few are named for their birthplaces, for birds and animals, or. for some circumstance connected with themselves. Children may also be named for spirits; for instance, a little girl is named Sodo'tRi for a woman spirit. Several men bear names of mythical characters, but given as nicknames. The following names are those for which I was given a meaning: Males KI okla'ndjamks, refers to its bearer 's big knees. K!6k!a'as, great blue heron (a shaman's name). LUpa'ktcis, he rolls when walking. Notco'ks, curled (as by singeing); given to a man as a curly-haired child. Ts!linitpaks (meaning unknown), given this man as a baby; he is also called P!asai'w6ks, big eater, because he was voracious as a boy. K6m6o's and Ske'lag, named respectively for the culture hero Knmfi'kfimps and 1k6l6m, marten, a similar trickster-hero of the myths. Both names were given because their bearers are liars-whieh gives some insight into their idea, of the tricksters.83 The man who bears the opprobrious name Skelag is more commonly called Ya'teus (meaning unknown). Lnle'ks, spotted, the chief named for the piebald horses he introduced to the Klamath, according to my informant and Gatschet 's. My best informant denied that the name has this meaning, however. Gatschet gives a second name for him, Shm6kaltko, bearded, for the little beard he wore.84 KEp!6tihas, short (?), the native name of the chief Tci'lokin, who was short. Tci'lokin is a name given by some northern tribe. One informant derived it from errors in his pronounciation, substituting kins in words, as another man " might put giO in every word," but just what these substitutions are is anything but clear. Females Blaia'kni, so-called because one of the blai'kni, uplanders, of upper Sprague river; a shaman. Ha'ntakl6tk, wide open, because she has a big mouth. Similarly she is called literally Mu'nsomgltk, mouth big. KoJWpli hole (kfipkfip), because she had small sunken eyes when a child. This is the only name my informant has ever had. LIme'lak, mule colt, because the woman is strong and can carry heavy loads. 83 "To act like Ske'lamtch is to do something not meaning to do it appar- ently . In speaking of somebody who acts on the sly, and differently from what he proposes and means to do, the Klamath Lake people will say, 'He acts like Sk6lamtch' " (Gatschet, l:cii, 115). 84 op. cit.. 2:335. Spier: Klamath Ethnography MWks, crying. This is given to children of both sexes but I recorded it as a woman's name. P`po'tcwa, to put feet (p6tc) in water, because she had done so, my informant surmised. Wokoilik&qks, leaning over, because this old woman as a child walked leaning to one side. Yatco'ks6p (a shaman), the mother (p'1i's6p) of a man named Yatco's. Another woman is called "greasy" but I failed to get the native word. STATUS AND KINSHIP TERMS Age Classification Terms A series of terms are in current use to designate individuals at the various stages of their careers. Unlike kinship terms, these are not used in direct address except for an occasional usage like "old man." They are descriptive of status, as much as the term l1o'ki, chief. There are exact equivalents for boys and girls, men and women. The outstanding features of the classification are: infants are sexless; thereafter discriminations turn on puberty, marriage, parenthood, bereavement, and old age. The curious category of a parent who has lost a child by death has real social significance. Like the loss of a spouse, the death of a child calls for a protracted period of mourning with special restrictions of conduct and is made the occasion of a vision quest. This is, of course, exactly comparable to our own bereavement within the immediate family but the idea of a termino- logical distinction is foreign to our thought. All the other distinctions have their counterpart among ourselves and are perhaps common to all mankind. MaZe85 Female Newborn baby muka'k86 Baby still at the breast muks Child of two years to puberty tc!a'ki ni'skak At puberty wa'k6ntk sto'pwitk (lit., menstruant) Youth after puberty tceI'loyak86, 87 siwalk Bachelor, spinster tcI'lwis si'u6mptc Young woman teina88 Newly married ska'b6ks89 ska'b6kak 85 Compare Gatschet, 2:140, 469, 509, 518, 560, 571, 602, 615, 697, 698. 86 Diminutive of the following. 87 Gatechet says before puberty. 88 Forms from Gatschet: cf. my informants' t I e'niwas, a woman living with her husband's family. 89 On another occasion given me as the term for a woman in matrilocal residence. 1930] 61 62 University of CaUfornia Pubiiations in Am. Arch. and Ethn. [Vol.30 After the birth of the first child skote sko'ksop Woman who has lost one child k'lekalas8 After the death of a child tc!6'kalatk l1pkla'ks Widower, widow tc!ima'ntk w6'nultk Old man, woman tci'k!a wilW&ks Old person k'mfitcha88,90 To misuse these terms dispa.ringly is an insult; e.g., to call an old man wa'kontk or skotc. Kinship Terms The peculiarities of the Klamath kinship system are to be found in the groupings of relatives in the parental and grandparental genera- tions with correlative terminology in the two descending generations. That is, the classification is made not only on the basis of generation but regularly according to the sex of the connecting relative. Thus grandparents are distinguished as they are father's or mother's parents, and the classification appears reciprocally distinguishing a son's from a daughter 's children. Again great-grandparents are either a grandfather's or a grandmother's parents and these terms are used reciprocally. Father's siblings are distinguished from the mother's a-and correlatively fraternal from sororal nephews and nieces. Similarly grandfather's and grandmother's siblings are separately termed and correspondingly the grandchildren of brother and sister. The con- necting relatives whose sex is siguificant in these classifications are father and mother for the avuncular relatives and grandparents, the -grandparents in their generation and the one anterior, anId brother and sister for the two descending generations derived from them. It appears further that this whole scheme revolves about the use of separate terms for siblings by male and female speakers. By exten- sion then a father's brother differs from a mother's brother just as a man's brother differs from a woman's brother, and so on. The nepotic relatives, are thought of in-terms of the siblings, the avuncular rela- tives as siblings of the parents, and the granduncle-aunt.as siblings of the grandparents. Add verbal reciprocity as the universal usage between those two generations apart and the Klamnath have four kinds of grandparent- grandchild and four kinds of granduncle-aunt-grandnep)hew-niece. ..It..is. consistent w'ith. this that, while the sex of the individual addressed 90 Hence the name of the culture hero KEmii'kuimps, old one. Spier: Klamath Ethnography is important in the parents' and grandparents' generation, sex is of no consequence among grandchildren, nephew-niece, or grandnephew- niece. The status is determined as a brother or sister of some connect- ing relative, or through a brother or sister, and in the latter cases the sex of the person addressed cannot matter. There are some particulars in which this system is not consistent. Thus one would expect from the above that there would be only four terms for siblings, a division only on the basis of the sex of the speaker and that of the person addressed. But in addition age dis- tinctions are made although not consistently. Thus a single term means a younger sibling, but it is not applied to a man's younger sister. Nor is this distinction reflected in the correlative groupings. For the rest, there are terms for father, mother, son, and daugh- ter, and all cousins (parallel and cross) of the speaker's generation are siblings. Hence parents' parallel and cross-cousins are avuncular relatives. Relatives beyond this all in the collateral lines are indis- criminately "cousins." Among relatives by marriage, we note a single term for a sibling- in-law of the opposite sex. Men and women use different terms for siblings-in-law of the same sex. The husbands of two sisters and the wives of two brothers have a term in common: this reflects the soro- rate and levirate. The levirate appears again in a single term for step-mother and father's brother's wife. Correlative groupings which might follow from these forms of marriage do not occur. The list of kinship terms follows. All are vocative so far as I know. This list was obtained independently of Gifford's and Gat- schet's published lists; cases of conflict were later resolved by discus- sion with the informants.91 p 'ti's6p, father. p 'ki's6p, mother. ko'clga, step-father (Gifford, a man's step-child. He also gives as meanings, mother's sister's husband and wife's sister's child, for both of whom I found no terms). p'sa'slp, step-mother; mother's co-wife; father's brother's wife (Gifford adds a woman 's step-child and husband 's sister 's child, but the latter is doubtful). These groupings follow from the sororate and levirate which are practiced. wfin6'k- son. bAt'p, daughter. 91 Gifford, Kinship, 41-43. 1930] 63 64 University of CaUifornia PubUcatios in Aim. Arch. aqnd Ethn. [Vol. 30 blui'ksip, father 's father; a man's son's child. pt !a'wip, father 's mother; a woman 's son s child. pk I a'sip, mother 's father; a man's daughter's child. pkI o'lip, mother 's mother; a woman 's daughter 's child. Gifford obtained these forms with the diminutive suffix-pak for young grandchildren, Gatschet and I without. tcuitecil, grandfather's mother; a woman's son's grandchild. wa'wiks, grandmother's mother; a woman's daughter's grandchild. (Gatschet gives this term as great-grandparent, Gifford as mother 's grandmother and a woman's granddaughter's child, which are included in my meaning.) p'tcoptcolip (from Gifford), father 's grandmother; a woman 's grandson's child. p 'se'ylp, father 's brother. p'ko'djip, father 's sister; a woman's brother 's child. blokootcep, mother 's brother. p 'sa'lip, mother 's sister. All parents' parallel and cross-cousins and half-siblings are included in these avuncular terms. p'a'ktis, a man's brother's child. b6tsga'lip, a man's sister's child. mako'k!op, a woman's sister's child. These apply as well to half-siblings' children and those of parallel and cross- cousins. baba'lilp, grandfather's brother; a man's brother's grandchild. blulo'clp, grandfather 's sister; a woman 's brother 's grandchild. koko'i (Gifford gives p 'kuyip), grandmother's brother; a man's sister's grand- child. (This was also obtained specifically as reciprocal between a man and his father's mother's brother.) bligi'p, grandmother's sister, a woman's sister's grandchild. Possibly there are separate terms for these granduncles and aunts cor- responding to the distinctions of paternal and maternal relatives in' other generations. It is evident that the cross and parallel cousins of the grand- parents are included in these terms. For example, in the genealogy below, bligi'p is reciprocal between a man and his mother 's mother 's mother 's sister 'a daughter, as well as between a man and his father 's mother 's sister. tke'unop, a man 's older brother. t6'piyfip, younger brother; a woman 's younger sister. tobo'kcip, a man 's sister. p 'a'nip, a woman 'a older brother. p 'tallip, a woman 's older sister. Half-brothers and sisters and all parallel and cross-cousins are called by these sibling terms, including the children of parents ' half-siblings. It is possible that cousins in the nearest collateral lines are called siblings as well (that is, children of parents' parallel and cross-cousins, who are themselves called by avuncular termg), but it is more probable that the following term, p 'o'mdjlp, is used. p 'o'mdjip, cousin, i.e., any distant relative. (In the genealogy below this is used by a man for his maternal grandmother 's sister 's daughter 's son.) hi'cw6ks, husband (lit., man). I also recorded a form xmbu'enip, which may mean something else. sn6wi'tec, wife (lit., woman). 1930] Spier: Klamath. Ethnography hosa'pk!6p (Gifford gives ceplugict), a child's parent-in-law (i.e., mutual parents- in-law). (Gifford gives the usage of hosa'pI16p as rare). pkasip, son-in-law (note: this should not be confused with pkla'sip). pto'd6p (or win6'kgumsnewai'tc, lit., son's wife), daughter-in-law. ko'sp6ks, a man's parent-in-law. pto'tkl6p, a woman's parent-in-law. Gifford notes the suffix -gewitk used in reference to the parent-in-law and child-in-law, but not in address, when the connecting relative is dead. mui'lguip, a man's sister-in-law; a woman's brother-in-law. My informant gave hi'ew6ks, husband, for a woman's sister's husband, but in evident confusion hastened to correct herself. Such a usage would be con- sistent with the sororate which they practice. p 'tco'kuip, a man 's brother-in-law. p'ala'mip, a woman's sister-in-law. p 'tc I ikip, wife 's sister 's husband; husband 's brother 's wife; co-wife. This classification follows from the levirate and sororate. There seems to be no term corresponding to this for a wife's brother's wife and a husband's sister's husband. spi'nuwip, sibling-in-law when the connecting relative is dead (cf. spilnu', to cremate). (I did not obtain this for a man's brother-in-law, but as no other term was recorded the use seems implied.) balocip, father's sisteris husband; wife's brother's child. p 'k! omklip, mother 's brother 's wife; husband 's sister 's child (possibly mother 's sister's husband but I doubt it because my informant at first denied that there is any term). There are no corresponding terms for husband's brother's child nor, I pre- sume, for wife's sister's child. The latter corresponds to the absence of a term for mother's sister's husband. The reciprocal of the first is p'sa'sip; this is given by Gifford for husband's brother's child. Evidently the feeling for verbal reciprocity is so strong that his informant gave the term, although, as will be seen by the genealogy below, there is no term for this child. sEa'muks, relative. GENEALOGY ?+9-1 or+ 9+OX c?3+92 'I I1 1 Henry+Elizabeth NANCY Nettie Allen+ 9 4 I I I Willie Lucile Fred Tom + Lola 95 Walter 68 e?Delford 9Lotella+Leonard eCoyote e%9 Benjamin Rosemary 7 dl6 1930]. 65 66 University of CaUfornia Publiations in Am. Arch. and Ethn. [Vol. 30 Nancy calls 1, p 'k I ollp; 2, p 'salkip; 3, no term; Elizabeth and 4 (both older), p 'tallip, who reciprocate with t6'piyiip; Nettie (younger), tS'piyiip, who reciprocates with p 'ta'lip; Allen, Henry, mui'lgiip, who answer mui'lguip; Henry (sinGe Elizabeth died), spi'nuwip, and he calls her the same; Willie and 5, mako'k!6p, who call her p 'salkip; Walter, Benjamin, bligi'p, and they call her the same; Rosemary, wa'wiks, and she reciprocates; Henry's siblings and father's brother's son, no terms. Lola calls Elizabeth, p 'sa'kip. Delford calls Lucile, p 'salkip; 8, p 'o'mdjip. Lotella calls 6, p'k!o'mkip; 7 (sex not stated), no term. The Klamath system resembles many others in this region, as I have previously pointed out.92 In general structure it belongs to the class I have called the Mackenzie Basin type, the distinguishing feature of which is the classification of both cross and parallel cousins as siblings. This is common to a group of tribes extending from the lower Columbia river through eastern California. and adjacent Nevada nearly to the Colorado river, and also in northwestern California. In the following table are listed all cases of the classification of relatives identical with that of the Klamath. This gives a somewhat different impression. The greatest number of identical categories are common to Achomawi, Atsugewi, Southern Maidu, Northern Paiute (Paviotso), and Washo; tribes lying consecutively southward of the Klamath, with only Northern Maidu omitted. There can hardly be any doubt of a common historic cause for this much of their terminologies. The Klamath system is set off from these, however, by the peculiari- ties noted below. There is no exact parallel to the Klamath use of four separate terms for parent's siblings, with reciprocal usage between father's sister and a woman's brother's child. As Gifford has pointed out,93 the use of four terms for these relatives is the most widespread type of grouping in California, but in the majority of cases these -terms are not used for parent's siblings alone. Four individual terms not used for other relatives (save sometimes for step-parents and spouses of parents' siblings) are used by tribes on the Columbia, in northwestern and east.ern California; Chinook, Wishram, Hupa, Whilkut, Wiyot, Atsugewi, Northern Yana, all Maidu groups, North- ern Paiute, Washo, Northeastern Mono, Yauelmani, Yaudanchi, and Paleuyami. None of these have the peculiar reciprocal usage of the Klamath. 92 Spier, Distribution of Kinship Systems, 77. Numbers preceding tribal names in the table below refer to the maps of this paper. 93 op. cit., 142. 1930] Spier: Klama&th Ethnography 67 As Gifford has also noted,94 the Klamath use of four special self- reciprocal terms for grandparents' siblings is unique. The general custom is to class them. with the grandparents. RESEMBLANCES OF' KLAMATH KINSHIP SYSTEm TO OTHERs System Nephews Parents as a and and Grand- Grand- whole Cousins nieces children parents children 71. Chinook .........X ? .... .... X X 69. Wishram .........X X .... .... x x 70. Wasco .........X ?............. ...... 6a. Shasta .....X.. .... ....X .... X 8. Karok .........X X X X .... .... 2a. Yurok . .......X .... X .... .... lb. Tolowa ....... . .... .... X lc. Hupa .........X X .... .... X .... le. Whilkut .........X X .... .... X .... 9. Chimariko ........................... . .... X .... .... lh. Lassik ......;. ... . .... .... X .... li. Sinkyone ........................... . .... .... X .... lj. Wailaki. . X .... X .... lk. Kato ................ . . X ..... X .... 4c. Coast Yuki .X X .... .... .... 4a. Yuki ...................X X X .... .... .... 4b. Huchnom ...................X X X .... .... .... lOa-g. Pomo ........................... . .... .... .... ? 16a. Northern Wintun .................. . .... X .... .... 6e. Achomawi ................... X X X X X X 6f. Atsugewi .X X .... X X X 7a. Northern Yana ................. . . .... .... X .... 7d. Yahi ................. . . .... .... X X 17a. Northeastern Maidu .................. . .... X X .... 17b. Northwe4tern Maidu .................. . .... .... X 17c. Southern Maidu .................... X X X X X .... 21a-84. Northern Paiute(Paviotso) X X X X X X 11. Washo ...............X X X X X X 21b. Eastern Mono . X X .... X .... X 21c. Western Mono ............... X X .... .... .... X 87. Moapa Paiute . ..........X ? .... .... .... X 88. Shivwits Paiute.X X .... .... .... .... 85. Uintah Ute ........ . . .... .... ? X X The Klamath terminology for siblings is exceptional: five terms, viz., man 's sister, woman 's older sister, woman 's older brother, man's older brother, younger sibling (for all except man's sister). The usual grouping in this region is by four terms (older and younger brother, older and younger sister) regardless of the speaker's sex. 941bid., 134. 68 University of Caifornia Publiations in Am. Arch. and Ethn. [Vol."30 I have few data on kinship usages. A child's spouse is treated like an own child and freely called son or daughter. Reciprocally, parents-in-law are often addressed by parent terms. The spouses of grandchildren and of nephews and nieces are not related, nor are step-brothers and sisters. There is no joking relation. The close bond between siblings and their families is attested by the opening of a tale: Wiwi'aak had five fathers, who were the five sons of Wile'akak. That is, one was his father, but the others were fathers because they were father's brothers. PUBERTY AND MENSTRUAL CUSTOMS The puberty dance is held only for the daughters of chiefs and others of wealth. For a poor man's daughter the rites are curtailed to the most necessary observances. The celebration serves as an announcement of nubility: girls marry within a year or so- of their dance. The dance (yo'k'Ela) occupies five nights at the first menstruation and is repeated at the four subsequent periods. It seems doubtful that any but the socially pretentious carry out full five celebrations. During the day the girl goes alone to sleep in the brush at some dis- tance, returning for the dance just before sundown. I am not certain that she then seeks supernatural power. At least the girl does not run about piling up rocks in lonely places as boys do. The dancing may take place in the lodge or outside, perhaps according to the season. It is simple enough; the girl trots forward and backward in the same course facing the east all the while. She is supported on each side by a man or woman whose arms are linked in hers. As they tire others take their places but the girl must keep on throughout the night. These others are young or old, relatives or otherwise, as convenience dictates. The singing is provided by a group of men and women who alone know the appropriate songs. These singers are in demand for every dance. They are paid by the girl's relatives, a horse apiece, a blanket for each man, a half-fathom of beads for the women, or an equivalent. Some are provided with rattles, poles to the upper ends of which bunches of dew claws are fastened, the butts of which they pound on the ground. At least the song leader has one; other informants spoke of two or three, of five men and five women, or of rattles for all. Spier: Klamath Ethnography Everyone is quiet during the singing, else the girl will die. My infor- mant could not say what would be done to the one who broke this rule. The singers are fed at midnight. The choristers sing about one another, men about women, women about men.95 The opening song is always: iwuine' sken djldja'lwis djaiyo'djai'wa toward sundown bachelors making a noise (?) The young men are having a jollification in the west. Toward daylight they sing: pe'kt kesRA'ni dji'lk!a ho'llidotqga Toward the dawn place a tiny brown bird came to meet us. The girl herself is expected to dance with downcast eyes, although there is no express prohibition against looking at the sky as among the Shasta. Common girls, however, will laugh and look around as they dance. There is no suggestion that the girl's glance is malignant. She observes three precautions during these five days, as she will at subsequent menstruations and during mourning. She must abstain from meat and fish, eating only lily seed or roots. She may, however, eat when she pleases and with the others. -Should she eat the forbid- den flesh, she will get "consumption" and die.96 She must not touch her hands to her face or head, using instead a scratcher (sado'ktcuts). This scratching stick or bone is tucked under the left wristband or hung about the neck. Every morning she -runs toward the dawn. In the Field Museum collections are several bone scratchers used during the girl's rites (nos. 61633, 61692, 63632). These are of two 'types; lozenge- shaped and oval. They are decorated with incised lines, set close together and filled with black pigment. Each is provided with a thong (fig. 6). During the day the girl retires to sleep in the bushes or in a dome- shaped mat-lodge a few hundred yards distant. She wears an old, dirty, fringed skirt, a blanket of skin (coyote, wolf, raccoon, or deer), and buckskin moccasins. The long buckskin dress commonly worn at dances may be used, although one informant stated otherwise. She does not wash nor comb her hair during this period. Her face is blackened with pitch and charcoal, but none is put on her hair as it would be in mourning. Instead the hair, hang- ing as usual in two braids in front of the shoulders, is doubled on itself to form two clubs, which are bound with braided cords of buck-brush 95 Some of these songs are given by Gatschet, 1:182, 189. 96 A bone drinking tube is not used. 69 19301 70 University of California Pubec&ations in Am. Arch. and Ethn. [Vol. 30 or red sage-brush bark. Buck-brush bands are worn around the head, neck, wrists, ankles, below each knee, and wound perhaps twice around the waist. This waist band may be a woven belt (k'ai'llis) of sage-brush. At the close of the ceremony the girl purifies herself by swimming; she does not use a sweat-bath. Other participants need no purifica- tion. Her clothing and bark bands are burned. Sometimes the cir- clets are wrapped around the branches of trees near the sweat-lodge where they are left to rot unless they are earlier gathered by someone who wishes to use them for a cautery. Fig. 6. Head scratchers, used by girls in puberty rites. A girl for whom no dance is held stays within her home, dressed in this fashion and observing the taboos for the requisite period. Menstruant women stay at home during their illness (ata sto'pwi) unless they are members of a shaman's household. Shamans must avoid them; seemingly other men are indifferent. Women of a shaman 's household occupy a little dome-shaped lodge near-by for five days. Menstruants eat only lily seed, camas, etc.; they permit themselves dried fish, but meat and fresh fish are taboo. Older women do not use the scratcher, nor wear the bands, nor burn their clothing.97 So much from my feminine friend, but a male informant had it that all women make use of a menstrual lodge, wear dirty clothing, neither wash nor care for the hair, and use the scratcher. However, women 97 Nor do they keep knotted string records of these events as on Puget sound. Spier: Klamath Ethnography do keep up their ordinary occupation in part during this time. They dig roots for instance, but others must care for the products further, else the women will never find any more. There is no comparable rite for boys at puberty. Nevertheless the time when a boy 's voice deepens is recognized as propitious for seeking power. He is sent into the mountains for perhaps five days, to wander at night, running continually, piling up rocks, swimming in the moun- tain pools and eddies of the rivers, fasting the while until a song comes to him in a dream. He must use a head-scratcher. Preadolescent boys are tossed into the river or whipped while they sleep. If the boy whimpers he will be a weakling; he is expected to jump up and shout. I have cited above the case of LEle'ks who was so treated at puberty, it seems, and thus lost his transvestite habits. When I [Pat Kane] was a little boy of seven or ten, my father sometimes threw me into the river while I slept. I did not ask why. Finally he told me, "I do not want to kill you. I do it so that when you are grown you will not easily get sick, so that if you are shot you will recover quickly." I always bathed in the river as a boy and you see that I have never even had rheumatism until recently. DEATH AND MOURNING Klamath funereal observances are personal, non-ceremonial, and somewhat unique in that they provide occasions for the seeking of power. There are no mourning anniversaries. Cremation is the universal practice, even for suicides, the new- born, and the stillborn. Only the bodies of secretly aborted infants are buried in the bush. The body of a man killed in war is brought home for cremation.98 Some Klamath men were trapping in the mountains west of Crater lake about eighty years ago when they were met by some Upland Takelma who killed four or five. One boy alone escaped and made his way back to Klamath marsh at night. The Takelma took scalps, hands, and feet of the dead, and held a dance. Many Klamath went to fetch the bodies, which they brought to Klamath marsh for incineration. Near each group of settlements is a spot where all cremations take place. It is somewhat apart from the settlements but not isolated. Three of these serve the Klamath marsh people; on the eastern side of the marsh near the towns k'ntaiwas (2, see maps, figs. 1-3), gepgiia'ksi (3), and i'wal (4). For others of this division of the tribe, there is one on Sprague 98 Cremation was given up forty or more years ago ("in about 1868," Gatschet, 1:86). The first burial was that of a Molala man, married to a Klamath woman, who was killed by a Klamath. 71 1930], 72 University of Caltifornia Publications in Am. Arch. and Ethn. [Vol. 30 river at ka'umnikn (14) and one on middle Williamson river near ya'ak (23). The du'kwa division crematory is near the mouth of that river. The Pelican bay division has a crematory at diin6'ksi (38). The remaining division, the Klamath falls people, have two at dl'tk!aks (50) on the east side of Klamath lake and two on either side of Link river. These eleven constitute the total for the Klamath territory. The body is dressed, with such beads as the dead possessed tied around the neck, wrists, and ankles. It is then wrapped in a tule mat. For ease in transportation on a man's back, the legs of a small corpse are flexed and tied. A larger corpse is lashed between two poles so that four men may carry it. Canoe transportation is of course neces- sary in most cases since the settlements are strung out along the shores. The corpse is taken out of the house by the regular exit. Cremation takes place five days after death, according to Gatschet 's Modoc informant.99 The pyre of green logs, three or four feet high, stands in a slight depression on the ash-heap (seka'lgi) of preceding cremations. These heaps are said to have been ten or twelve feet high, "as high as a haystack," built up of the ashes scraped together after each burning (splinu'). Gatschet states that the one north of the (middle?) William- son river was twelve feet high in 1877.100 White vandals have long since leveled these in search of plunder. All those I saw are now low mounds not more than a few feet in height nor more than thirty feet in diameter. Property of the dead, such as his beads, arms, and skins, is piled on the body as it lies on the pyre. Other valuables are contributed by relatives and others who wish to honor the dead. The recipient family assumes a peculiar obligation by accepting these; should the contributor later stand in need of a wife for himself or a relative, this family must provide the bride. The amount consumed is considerable, for even today the ash-heaps are littered with glass trade beads, iron halter rings, gun barrels, and fragments of obsidian blades that have withstood the flames. When the fire has burned down, several of the male relatives who stir it with long poles, roll the remains out to be rewrapped in a mat and burned again. It is said that bodies are hard to burn; the fire must be renewed several times. The heart is the last portion to be destroyed. Poor people might not be able to burn the whole body. 99 Op. cit., 1:85. The singing in the dead man's lodge described by Gatschet (1:90, 92) is apparently modern. I was not told of it. 100 Op. cit., 1:86. 1]Spier: Klamath Ethnography The rewrapping of a man is done by male friends, of a woman by women, but men alone tend the fire. They cover the remains as quickly as possible, all those about crying loudly the while. Little children are kept at a distance. By way of purification, those who tend the corpse leave their clothing out of doors for several days and swim, or bathe in the sweat-lodge. The house in which the death took place, even that of a child, is burned with its contents.'01 The survivors live elsewhere until, on a remarriage, for instance, a new lodge is built. If the house of a shaman is burned with the regalia it contains, a violent storm springs up in the night, hence they wait for calm weather before setting it afire. It is only safe for an old woman to fire it. When a shaman dies and again when he is cremated, storms rage for two days. Strict mourning is enjoined on the spouse of the deceased and his parents. It is quite clear that a son or daughter need not mourn so deeply for a parent. Those who observe the fullest mourning cut their hair short immediately on the death; this is called sla'mltk. After the cremation they sweat for five days, go out in quest of a vision, and observe the mourning restrictions throughout the following year. The mourner fasts inside the sweat-lodge (spo'kliks) from early morning until sundown during these five days. There are only four sweat-lodges proper to the use of mourners, or rather three. This restrict-ion is so inconvenient considering the scattering of Klamath settlements that I would doubt it, had I not definite assurance from all informants.102 There are none in the territories of the Pelican bay and Williamson river-mouth people. The three lodges are semi- subterranean, of timber and rock construction, and earth-covered. These are said to be the gift of KEmui'kuimps, the culture hero. A very small one stands a mile and a half north of Klamath Agency.103 A second is on the south side of a point called walka'ksi, near Odessa on the western side of Klamath lake.'04 The third and largest is on the hillside at the angle of the marsh east of Algoma. The Klamath Falls group have no mourners' lodge but make use of a hot spring at ho'kewalks, a mile east of the present Klamath Falls railroad station. 101 I was not told of cairns of stones tossed on the spot by passers-by as described for the Modoc (Gatschet, 1:85). 102Gatschet also noted three alone (op. cit., 1:xlii, 82, 142). 103 This may be the one mentioned by Gatschet (op. cit., 1:xxxi) although the name he gives resembles that recorded for site 64. 104 This is referred to in a tale of Mink and Weasel. 1930] 73 74 Uni.versity of Caifornia Puxblications in Am. Arch. and Ethn. [Vol. 30 There they place planks over the spring on which to lie. The common small sweat-lodges are not used in this connection. The sweat-lodge near Algoma is now in ruins but it was always in need of repair, I was told (fig. 7). It is now a U-shaped wall of stones enclosing an area ten by twenty feet. The walls still stand five or six feet high, composed of thousands of small rocks. The roof was a rough gable covered with dirt. The lodge faces the east, uphill on a rather steep hillside. Stones were heated outside and carried just within the entrance. Air seeped in along the ground at the back or lower side of the lodge, -so the mourner would retire there from the heat. A little spring is close by. There were no houses within several miles of this place. The little lodge north of Klamath Agency faces south. Fig. 7. Mourners' sweat-lodge near Algoma. The nearest man stands in the mouth of the U-shaped structure; the middle figure on the far wall; that to the right beyond it. At the end of the five days of sweating the mourner swims in the spirit places in quest of power or seeks it in the mountains. The reason seems to be not alone that he has lost power through his sorrow, but that this is an especially propitious occasion when the spirits will pity him for his grief. One such place is a pair of rocks, Wolf and his wife, north of Modoc point. A man will wrestle with the Wolf rock, his wife with Wolf 's wife, in order to become strong and run well. During mourning women wear bands of buck-brush or sage-brush around the head, neck, waist, wrists, knees, and ankles like those worn during the puberty dance. The bands are disposed of by wrap- ping them around trees near the sweat-lodge where they are some- times collected by those who wish to make cauteries of them. Both parents (or men alone) wear service berry or sage-brush belts like those of girls at puberty, so that the back will not become stooped. For a year or longer, pitch is smeared over the face and hair, and black- 193]Spier: Klamath Ethnoography ened with charcoal. This is strictly enjoined for mourning women; older men follow the custom and perhaps even children, except the little ones, but young men may be exempt. This pitch is not that exuding from the logs of the pyre as among the Modoc'0l but is gathered from any tree. The hair cut off is later tied in little rolls all over the mourner's buckskin dress or shirt by way of decoration. (Enemy scalps serve the same end.) These garments are burned with their owners. During the sweat-bath period and for perhaps a year longer, the mourner must abstain from fish for fear of sickness,'06 must not touch his hands to the face else it will wrinkle, nor may he touch his hair lest it turn gray. Instead, a bunch of grass is used when washing, and a scratching stick (wa'was), nine inches long, is suspended around the neck on a sage-brush band, or tucked under the wristband. The mourner spits into a hole dug with the scratcher. This may explain a song record by Gatschet: "This is my song [or spirit], that of the (sweatlodge) stick-hole."'107 These are the same taboos observed by a menstruating woman and by a boy at puberty. After the loss of a wife or child, a man abandons the first fish or game that he takes, gives the second to another, and succeeding ones he keeps; a gambler gives away his gains. This is called sapu'tsa. A woman does similarly with the roots she digs after her husband or child dies. Something of the same attitude is shared by a gambler after receiving spirit aid, a novice hunter or root digger; all give away their first gains. Similarly no one who has recently lost a child may fish or even cross a river for fear the fish will go away. They must wait a year and then again use the mourner's sweat-house. These may all be viewed as rites of purification or attempts to reestablish oneself as a normal being. Wolf lived opposite the cave at komh'6ksi on Sprague river. When his wife died he brought his children to the river bank. The people shouted to him not to cross the river because fish were running. But Wolf said, "I did not come to catch fish. I brought my children because they wanted fish. " They stopped there. Then Crow laughed at them and they turned into a row of rocks.'10 105 Gatschet, 1:86. 106 Gatschet (1:89) understood his informant to mean that after sweating for five days, a widow eats only fish and no meat. 107 Op. cit., 168. 108 Told by Tom Lang. 1930] 75 76 University of California Publica-tions in Am. Arch. and Ethn. [Vol. 30 The following myth brings out some of the mourning practices.109 Coyote's wife died. He set out to seek power. H.e burned up his house and all his belongings. He went along crying: he sang ha'newa ha'newa b6liik6'mhiikdino'l!t (a cry) It ought.be dusk quickly. He wanted the sun to go down quickly so that he could swim for power. There was a woman sitting by the road. He said, "I do not want a woman. I am crying because my wife died." Again he went on singing hh'newa..... Next he saw some moccasins on the road. He said, "I am not looking for moccasins. I burned mine when my wife died." He went on singing again ha'newa. Now he saw a heap of food on the road ahead. He said, "I am not looking for food. I burneAl all my food when my wife died." He went on, singing ha'newa. Then he came on a pile of ripe plums which had fallen to the ground. There was water there so he washed out his mouth with a handful [as is always done before a meal]. He had his scratching stick at his wrist. He stuck it into the ground and spat into the hole. He said, "This is the way the pse'odiwas (humans) will do." So he ate, ate, ate the plums until his belly got big. Then he got sick and could not swim. Thus he made the rule that mourning Indians must not eat plums, else they will get sick. GAMES Gamblers seek spirit power for success, men and women alike, by swimming in spirit haunts in the river or in mountain lakes. The first of his gains after receiving this power must be given away, just as he must give them away the first time he plays after the death of his child. This is of a piece with a lad or girl destroying or giving away the first results of their efforts at hunting or root gathering. Luck is also acquired by finding large obsidian blades and certain dead animals; mole, weasel, mink, or frog.1"0 Finding the mole is luckiest. "To find anything like this is lucky, so it gives you luck." These must be found dead; there is no use taking a trapped animal, or a recently manufactured blade. The premium put on the obsidian blades is reminiscent of their exaggerated value as symbols of wealth among the tribes of the lower Klamath river."-' All such lucky objects are bundled in white sage twigs and hidden some distance from the house. This may also be true of the gambling bones; at least Pat Kane keeps his at the crematory several miles from home. I was not 109 Told by Nancy Phillips. '10 Gatschet lists as similar Modoc charms, a hummingbird 's nest, the leg of a dead frog, a dead mole, a rare "fire-bug," and arrowheads (1:134). 111 Kroeber, Handbook, 26, 54ff. Spier: Klamath Ethnography told that this had any connection with the possibility of contamina- tion from women's presence in the house, but I did not enquire. A gambler must fast if he wants success, and must begin the fast before he goes to fetch his charm. The blade will disappear if not properly cared for. The gambler carries the stuffed skin in his hand or places the blade in water at his side as he plays, or wets it from time to time. The frog's foot, wrapped in an odoriferous leaf, or a mole's foot,1"2 is placed under the blanket or mat on which they sit while playing. My informant had never heard of putting the blades under the mat, as Barrett noted."3 Luck is called ditc satwa'ya; k!ai ditc satwa'ya is literally, indeed no good luck. Gamblers paint their faces, put strings of beads around the wrist, and attach little feathers to their hair, when they play in all serious- ness. They sing as they play. The stakes are piled beside them. The side that loses, bundles up its gaming sticks and articles, carries them off to bury at a crematory, and sets out to seek spirit power for luck. The principal game is the four-stick game, called so'kals, from so'kla, to gamble, hence the gambling game par excellence. Gatschet"14 also calls it spe'lshna, to put the fingers forward (from speiluish, index finger) and shulshe'shla. It is played by men seated in two opposing rows. Two pairs of sticks are provided, with a circular mat or basket or a skin to hide them under. The sticks were described to me as nine inches long, two an inch in diameter, marked at the middle (these called cu'lses), and two only half as thick (sko'tos). Culin describes a number of sets, of which the thicker sticks taper toward each end and are ornamented with parallel spiral lines at the middle, while the slender ones are closely wrapped with buckskin thongs for their whole length. From this they derive their name "wrapped up" according to Gatschet. The thicker sticks are made of mountain mahogany."15 The hiding basket (p'a"la) is flexible and fiat, hardly more than a circular mat; it is ornamented and has a bunch of deer dew claws attached to the center of the convex side. Blankets, mats, elk hides, or buffalo hides are sometimes used to hide the sticks; the last especially by rich men. Twelve short pointed sticks and two longer are provided for counters. These, contrary to the usual usage elsewhere, are not placed in a single group, but six of the shorter 112 The latter according to Gatschet (1: 134). 113 Material Culture, 253. 114 op. cit., 1:79. 115 Culin, Games, 328-332; Dorsey, Oertain Gambling Games, 23. 1930] 77 78 University of California Publ4ctions in Am. Arch. and Ethn. [Vol. 30 variety and one longer one are stuck into the ground before each group of players. These are called kee'ic, which, according to Gatschet, is derived from ksh'ena, to carry off.116 Three of the sets described by Culin have each six counters. The hider rattles the four sticks against the under-side of the hid- ing basket which rests on the ground before him and allows it to drop over them as he leaves them on the ground. The guesser must specify the order in which the sticks are arranged, which he does with a very characteristic gesture. If he is correct the sticks are thrown to him to hide. If he is wrong, the hider draws one or two small counters from his opponent 's stock. The longer counter is taken last. For two arrangements, B and C below, he takes two counters; for A or D he takes one. The game is won when one side has all fourteen (no'ga, to win all the stakes; go'le, to win). Each arrangement of the sticks has a name and a definite gesture which the guesser uses to describe it. He claps his hands once when his decision is made and brings his hand down or across in front as may be. Name Arrangement Gesture A. sko't6s ( ) Moves right hand to the right. B. yo'skls l l l I Moves right index finger down the center. C. skW'tc!6s Moves first and second fingers of the right hand down the center. D. woi'lis (or b6okc) l (or liii) Moves right thumb to the right to point out the thin sticks or the fore- finger to the left for the thick sticks (reversing the movements for the second arrangement). Dorsey gives a somewhat different procedure. First, he is probably right in stating that for the opposite arrangement of A (with the slender sticks to the left) the hand is moved to the left. But he states that the guess for this position does not matter, since whether right or wrong, the sticks do not change hands and the hider gains no counters. Further, by his account, after a wrong guess is made and one counter forfeited, the hider arranges only one stick of each kind for the next guess. Both Dorsey and Gatschet give the name vuish, vf'ish to arrangement A. The hand game is of recent introduction; of this all were in agree- ment. It was brought from the far north by a man called Spo'kan, possibly one of that Salish tribe, before the Modoc war. He came from Warm Springs and married a Klamath. He introduced the accompanying songs as well and its name nai'atia.117 116 Op. cit., 1:79. 117 My informant had never heard the name l6ipas recorded by Dorsey (op. cit., 22). Spier: Klamath Ethnography This is also a man's game played by about six in a row on each side. Two men are each provided with two short bones, one of each pair marked with a black band. These two change the bones from hand to hand as they sway from side to side erossing and uncrossing their arms, their companions accompanying them with swaying and song the while. The opponents guess the hands which hold the unmarked bones. Twelve sticks for counters are stuck upright between the contenders. As with the four-stick game the successful guess wins the right to hide the bones; at each failure the hiding party takes counters from the neutral pile and, when this is exhausted, from the opponent's winnings. Here too the guess is indicated by a particular and characteristic gesture. In the diagram below the bones are repre- sented as in the hands of two men seated side by side; the crosses indicate the marked, circles the plain bones. Name Arrangement Movement Counters taken by hiders yo'sKIs xoox Hand edgewise down middle 2 ske'tc! bs oxxo Prone hand forward 2 woi'is"18 xoxo Forefinger to right (or left for 1 (or oxox) the second arrangement). When each side has six stick counters, it is called otano'ga, now cooked. The side getting all twelve wins. Dorsey describes the bones as three inches in length, two wound about the middle with buckskin. The wound pair are called sku'tash, tied around, or hl'cwoks, men, the others solsas, female (?). The counters he renders kshesh.1"9 My informant reported that the Pit river people have short bones which they roll in grass between their hands, and toss into the air. The beaver-tooth dice game (sko'se) has its usual simple form. It is played perhaps more frequently in winter than in summer, and while primarily a woman's game, both men and women play together. The dice are the four crescentic incisor teeth of a beaver, ground hog, or ground squirrel, or bits of wood.'20 The upper teeth (called hic- 118 These names, which are identical with those for the four-stick game, were given by the same informant but on a different occasion. The names given here are probably correct for both games, hence Dorsey, Gatschet, and my own doubtful alternative for D above, are evidently all wrong. Arrange- ment A cannot occur in the hand game, of course, sinee each man must have an assorted pair. 119 Dorsey, op. cit., 22. In 1930 I collected two sets in which one pair was encircled by a carved and painted fillet. 120 The Northern Paiute use long sticks for this game, according to this informant. 1930] 79 80 University of California Pub4ation in Am. Arch. and Ethn. [Vol. 30 wo'ks, men) are ornamented along one face with zigzag lines; the lower pair (gu'lu, women) bear a row of dots or parallel cross-lines. The hollow ends are plugged with something red for decoration. They are swept up off the thigh and scattered on a flat stone, such as a grinding slab. If they lie so that the marked faces of two "men" or two "women" are uppermost, the player wins one counter; if the cast shows all blanks, he gains two, and if all marked faces, he wins the game. When he throws one blank, the dice pass to another player. This is called k'ai'dwa snl'kM, to win no counters. The first three of these scores agree with Gatschet's information.12' Another of my informants said that any two marked faces count one, all of a kind, marked or unmarked, two. The counters (snm'kal) are twelve (ten?) sticks, which are used in the usual fashion; each player draws from a central pile, then from his opponent's, until he has all. I am not certain, but they may play as two opposing sides. Both Oulin and Gatschet describe the use of woodchuck teeth as well. Gatschet writes of all four marked with parallel lines or crosses on one side. Dorsey describes stick dice for this game, seven and three-quarter inches long, flat on one side, rounded on the other, and tapering at the ends. Gatschet's information is that the game is played by women only, by two persons or two partners to the side.122 A form of the stick game from the Klamath at Siletz reservation is described by Culin128 under the name tuckinaw. This, the fanxiliar game of northwestern California, must have been learned at Siletz, since it is unknown on the Klamath reservation. Of a bundle of thirteen sticks, two are marked differently from the others. The bundle is divided into groups each wrapped in a wisp of grass. Evidently the guesser must avoid choosing the bundle containing the two special sticks. Double ball shinny (djima') is a-woman's game played in summer. The "ball" (djima') consists of two bones or hardwood billets, eight inches long, and somewhat more than an inch in diameter, fastened at their middles by a nine-inch length of buckskin.'24 Each player has a fathom-long straight stick (so my feminine informant states, but Gatschet says two) with which the ball is struck or rather thrown. The stick is called swe'kos. The object is to drive the double ball between the opponents' goal posts (Gatschet gives yuash; yo"a is to 121 Op. cit. 1:80. 122Culin, op. oit., 137; Gatschet, 1:80, 81; Dorsey, 25, 26. 123 COulin, op. cit., 247. These may be Klamath river people, not the Klamath of Oregon. 124 Gatschet describes the string as two to three feet long connecting two sticks or pieces of cloth (op. cit., 1:81). Culin figures a ball in which the cord passes through the center of each billet (Games, 659). Spier: Klamath Ethnography win). Carrying it beyond the line of the posts but not between them does not count. They start by throwing it into the air with a stick; this is called salk.!we'tka. The players, five or six to the side (Dorsey says two to ten play125) are scattered over the field. Chasing one another is called swa'tsna. -The ball must not be touched with the hands nor kicked. Shinny (SEo'tE) is a man's game played in the summer. Six men take part on each side and attempt to drive the ball between the goal posts, arranged as above. The same provisions apply here. In addi- tion tripping, holding, and hitting one another are not permitted; there is in fact a functionary, a non-participant, who sees to this. He also throws the ball up between the contestants at the center of the field to start the game. The ball is of hard wood, two inches in diame- ter. The sticks are about an arm's length, three fingers broad more or less, and crooked at the lower end. These are made of willow roots or other wood. Football is not played by the Klamath, but is played, according to my informant, by Modoc and Northern Paiute.126 The usual form of the hoop and pole game is not known to the Klamath, i.e., hurling poles at a rolling ring. Analogous to this, however, is a game (cil'klclka) of shooting or throwing darts at a hoop, played in several ways. It is also called ho'sako-ks from hosa'tena, to roll. A ring (ci'rkelka) of tules is made fourteen inches in diameter or sometimes much larger. One informant described a ring of willow branches as thick as one's arm and thirty inches in diameter. Dorsey describes a ring of six inches used by boys. A ground is prepared by burning the grass on a course seventy- five to one hundred feet long. At each end four or five archers take their positions. One rolls the ring by an overhand throw; the others attempt to make their arrows stick in it. One who misses is penalized by the others filliping him on the back of his hand or punching him in the small of his back. They always fillip the same spot until the hand swells. Another informant gave a variant in which the ring is propped up so that its edge faces the opposing side, who attempt to shoot it. If they miss, their opponents win; if they hit it, it must be rolled toward them while they shoot again. The stakes are bows and arrows. My informant's account of this game was not clear; I may have con- 125 Dorsey, op. cit., 19. 126 Cf. Culin, op. cit., 704. 1930] 81 82 University of Calfornia PubUZwtion, in Am. Arch. and Ethn. [Vol. 30 fused two games. A better informant denied that they shot at the edge of the ring but Dorsey records this.'27 Two rings are used according to his account, one beside each party. These rings are pecu- liar in that each has a bar across its diameter. In another form of the game recorded by Dorsey, two boys seated indoors roll a three or four-inch ring between them. The arrow shot into the ring must be dislodged by a shot of the one who rolled the ring. If the latter fails, the other receives one of the stakes, an arrow; if he succeeds, the trial does not count. In still another variety of the game the ring is only half an inch in diameter and the dart is a bone point set in a wooden handle. This is also played indoors by any- one. The player attempts to pierce the rolling ring with his dart. If he pierces one rim it counts one, both sides two. A dart game (slikne'ls, dart) is played by men in summer. The dart is a pointed willow pole, eight to ten feet long, or, according to a second informant, of fathom length, feathered at one end, and sharp- ened or provided with a bone point at the other. Two targets are set up some distance apart, perhaps one hundred feet. These are bundles of tules four inches across and three feet tall, or smaller bundles mounted on sticks. When one man has made his javelin stick in the bundle, the whole party proceeds to that mark and throws at the other target. No stakes are wagered, but the man who wins can punch the others over the kidneys or fillip with his middle finger the backs of their hands or their foreheads five times or more. Another informant had it that the players are paired; one on each side throw- ing his dart. in turn. He fixed the stakes as horses, blankets, and beads. A variant of the game described by Dorsey128 involves forked sticks as targets. If the javelin rests in the crotch, it counts five, otherwise the nearest throw counts one, the total to win being ten. KEmui'kuimps, the culture hero, played this game between Buck and Bare islands at the opposite ends of Klamath lake. Other athletic contests are frequent. Foot races (so'ina) are indulged in by men or women, but not mixed groups. The race is to a nearby goal (without returning) or over a circular course. Boulders are lifted to see how far the contestants can stagger with them. There are several large spherical stones, up to eighteen inches in diameter, at the village du"' llkut on Klamath marsh where various groups gather in summer. Men, and even strong women, of the various set- 127 Dorsey 's data are also confessedly poor (op. cit., 17, 18). 128 Op. Cit., 19. Spier: Kiaimath Ethnography tlements are pitted against each other. Wrestling is a sport for men and boys. The village suwiak-a"eks, jump over, at Modoc point takes its name from contests of jumping over a rock there. This is about three and a half feet high and the approach is up a hillside. The Klamath are fond of diving; certain places on the lake are favorites, as the rocks on the northern shore of Buck island. Poles are balanced on the forefinger; this nameless sport is indulged in by boys alone. Ball juggling is unknown.'29 The so-called ring and pin or cup and ball game (cok!o'k!os) among the Klamath consists of an elliptical or lozenge-shaped bundle of tule rushes to one end of which is fastened the cord which at its other end is attached to the butt of the pin. In a specimen I collected the bundle is very light, fashioned from the inner layers of tule stem into a mass an inch and a half long with a diameter half that dimen- sion. The cord is a mere filament eight inches long, twisted of dried nettle bark. The pin is a porcupine quill (smai'em) as long as the ball. The ball is allowed to hang pendant; by a quick plunge of the pin straight down, an attempt is made to pierce the point of attachment of cord and ball. This "breaks the month" (shapashspatcha, accord- ing to Dorsey) .130 It is considered great luck, for it will shorten the month and hasten the coming of the spring. It is a pastime of young and old of either sex during the long winter nights spent in the earth-lodge. Top spinning is not confined to any season nor any ages. The top (hesta'lgeas) is a pointed stick thrust through a disk of wood or bark or a lump of pine gum. It is twirled between the palms. The contest is to see whose top spins longest. A buzzer (ske'okee'wls ?) is made of an astragalus bone of a deer and a sinew from the deer's back. The looped cord passes through two holes in the bone. The loop is swung in the familiar fashion to twist it; pulling on it makes the bone spin. The buzzing sound pro- duced says, poke'wot poke'wot, drive the deer out, evidently a charm. Near Pelican bay are tongues of land extending into the marsh from the foot of the mountain. Deer were driven out onto these and killed, or forced into the water where women in canoes waited to kill them. The buzzing is a "sign of this." The buillroarer is a boy's plaything; it has no significance. This is a paddle-shaped piece of wood, about fourteen inches long, pro- 129 Gatschet records the term shikukang6tch for stilts (op. oit., 2:307). I am sure I would have been told of stilts had these been aboriginal. 130 Op. cit., 21. 1930] 83 84 UTniversity of California Pu,blicatiomz in Am. Arch. and Ethn. [Vol. 30 vided with notches at the narrow end in which the cord is tied. It is not ornamented. The sling (ske'wlc) is used only in sport. It is wholly of buck- skin; a wide piece to which thongs are attached. One is provided with a loop to slip over the finger. This is an aboriginal device. As a mere. plaything hummingbirds (p'i'sas) are fashioned from the hard interior of the tule. These are hung up in the earth-lodge. String figures or cat's cradles (soplo`ptcule`as, from k!o6ptc!a, little finger ?) are for amusement. Yet if played in winter the months will grow longer. Their number is legion. I was told of a few; the Fig. 8. String figures. stolen baby, Tca'kiak, sunrise, owl's nest, sick Coyote's knees, the girl who fetches water, a bundle of arrows, the abandoned boy who grows, and two coyotes who run from each other. The interesting feature of these figures is that they depict moving scenes; thus sick coyote sticks up his knees; the girl goes to fetch water crying omdja, omdja, water, until she spies an Indian when she cries maklaks, human, and flees to the other hand to begin again. Three of the figures were recorded. The figure of the stolen baby is called swensowa'tkeas, taking from the cradle. Put each hand through the loop so that it lies on the back of each wrist. Lift the radial13' string with each little finger from the ulnar side and the ulnar string from the radial side with each thumb, so as to form a loop on these four fingers (fig. 8a). Insert each 131 Following the terminology of Rivers and Haddon (String Figures, 147) the thumb-side of the hand or finger is called radial, the little finger-side ulnar. Hence strings on these sides are called radial and ulnar strings. A loop near the butt end of a finger is proximal, near the tip distal. Loops may lie on the palmar or dorsal side of each finger. Spier: Klamath Ethnography index finger under the intersection on the opposite palm. Using the lips, lift the loop from the back of each wrist so that it rests in front of all the fingers (fig. 8b). With each index catch the strings at A and pull them through the loops on each index while allowing the latter to slip off the dorsal side of the index. This forms a horizontal rectangle at, the center which represents the cradle. This is narrowed by bunch- ing the fingers of each hand. A little stick is thrust through this with the lips; this is the baby. Let slip the loops from each index finger. Pull the hands apart and the stick will fly toward one of them. (The resulting figure is shown in fig. 8c.) The hands represent two Old Women spirits (wile'akak) fighting over the baby; the one who gets the stick wins. A second figure forms Tca'kiak], Little Boy spirit. The loop is placed over the right thumb and the left thumb and index finger. Hold both strings against the palm with the three other fingers of the left hand. Insert the right index under the loop over the left thumb and index from the dorsa,l proximal side ,(i.e., from the direction of the wrist). Release the strings held by the other fingers of the left hand. This gives figure 8d.132 . . . . The final figure showing Tca'kialk is given by figure 8e. The third figure, sunrise (sa'basdlini'gi), is shown in figure 8f. Pull- ing the hands apart makes the middle loop stand as a triangular figure, representing the appearance of a mountain peak gradually emerging against the dawn. This construction is begun with the loop passing across the dorsal side of the left thumb and little finger and the palmar sides of the other three. Give the loop one complete twist (so that the ulnar string, e.g., of the left hand, is on the ulnar side at the right) and insert the thumb and little fingers of the right hand similarly. Pick up the palm loop of each hand from the proximal side with the opposite index. Hold the radial string on each index aga,inst the palm with the middle and ring fingers. Insert each index under the ulnar strings of each thumb from the proximal side, lifting the thumb loops off onto the indexes. Care must be taken to keep these loops at the tips of the indexes. Release the strings held by middle and ring fingers. With each thumb carry the radial string of the proximal loop on ea,ch index under (past the proximal side) the ulnar string on the little fingers. Bring back the thumb bearing a loop of this ulnar string. Slip little fingers out of the loops. (At this point my notes fail.) 132 Unfortunately my field notes play me false: I cannot give the interme- diate stages for either this or the following figure. 1930] 85 86 University of California Publicationm in Am. Arch. and Ethn. [Vol. 30 Children imitate not only the daily activities of their elders, but their games as well. Little girls build little dome-shaped lodges. Their dolls are made of clay with modeled eyes, nose, and mouth, but are not dressed.133 Both these clay images and those made of wood a-re called a'm'os, but the wooden figures, representing the spirit Tca'kialk, Little Boy, are not played with. Children play at shinny and double ball, and make cat 's cradles. They play a game called wo'yllitas wherein they run crying wo'yl wo'yl . . . . as long as their breath holds out and stand a.t the spot where they fail. The object is to see who can go farthest.. Or they cry dovgliu doygliu .... . Similarly, girls squat on their heels with hands on knees and jump up and down, singing la'masi, to see who can do it longest. They bet on this. A favorite spot to play this game, called salmasi'a, is at dtuna'ksi on Pelican bay, where the ground resounds when they jump. It is worth noting that the number of counters used in the guess- ing and dice ga.mes is always twelve. In the four-stick game-I so recorded it-the counters are thought of as twelve plus two. Another of my informants spoke only of the twelve and this is the number recorded by Culin and Dorsey.134 This number is unexpected because the pattern number of the Klamath is five. It ma.y be tha.t these games came to the Kla.math in a form with twelve units and have not been reduced to the normal five or its multiple. How the pattern operates is illustrated by the informants' statements that five or six on a side play double ball shinny, that the perfect score in the dart game is five, and the penalty there five or more blows. On the other hand, it may be significant that I was told of six men to a side in the hand game and six in shinny. There are three possibilities; six and its multiple may be an indigenous pattern number for games, it may have been acquired as part of the game complex from other tribes, or it may be only a chance reference of the informants. SMOKING Smoking is a pastime with little to suggest ceremonial usage. To be sure, the shaman smokes in the interludes of the curing ceremonies, using a special form of pipe, special tobacco mixed with reputedly- poisonous roots, and prepared for him by his speaker, but he does not. 133 The Northern Paiute make clay figurines representing horses and other animals, corrals, etc., according to my informant. The Klamath do not. 134Even the form of the dice game described for the "Klamath" of Siletz, involving thirteen sticks, two of which are marked, may really be intended for twelve plus one, since elsewhere only one stick of such a set is marked. 19]Spier: Klamatth Ethnography cure by blowing smoke on the patient nor blow it to the distant spirits as an offering. The impression gained was that this is a matter only of self-indulgence during his periods of rest. Gatschet translates a song as "The smoking pipe is my medicine-tool (m ulwas), the imple- ment for tobacco," but the word also means any esoteric parapher- nalia of the shaman.135 The lay use of tobacco is at any time that convenience dictates, e.g., when resting in the evening. Women use tobacco as freely as men. The custom of passing the pipe from one to another is probably absent though I made no specific inquiry. Tobacco (ko'teka)'36 is smoked, never eaten. A boiled decoction of it is sometimes taken for bowel disorders. The plant (Nicotiana attenutwta Wats.) is not cultivated, but com- monly grows in dry places.137 A bundle of these plants together with kinnikinick is hung to dry in the top of the sweat-lodge until the leaves turn brown. Coville states that the dried leaves of manzanita, also, are mixed with the tobacco.138 According to an informant the seed pods (which are sticky when fresh) and the adjacent parts of the flower are smoked; according to another the leaves and stem. The several plants are ground with any convenient little stone. The pipe (p!oks) is commonly clay with a spherical bowl and a short wooden stem, but is sometimes of stone and discoidal. The more common stone bowl is elbow-shaped with the arms at an obtuse angle, made of soft sandstone, shaped and hollowed by pecking. Barrett states that the discoidal form is less used than the angular but my information is to the contrary.'39 The stem is elder with the pith pushed out; Coville mentions the wood of the wild rose.140 The manu- facture of clay pipes by a non-pottery making people seems unusual; the only other use of clay is for making dolls. Yet the ease with which clay can be manipulated as against the difficulty of drilling and shap- ing stone must have favored its use. A shaman's pipe (which has no special name) is the common clay bowl fitted with a long stem, eighteen inches or more, wrapped with 135 1:167. 136 No word correspinding to the Shasta 6p, tobacco, Takelma 6Up, Crow 6p, etc., is known. 137 Coville, Notes, 104. One of Powers' correspondents states that the Kilamaths (Klamath river people?) cultivated it (Tribes of C'alifornia, 426). 138 Ooville, Notes, 102. What is here called kinni1kinick is probably Coville 's bearberry (ka-mai-mi, Arotostaphyto& nevadensis Gray); the common kinnikinick of northern latitudes does not grow so far south as Klamath lakes. '39 Barrett, Material Culture, 253, pl. 22. 140 Coville, Notes, 99. 1930] 87 88 UnPiversity of California Publioations in Am. Arch. and Ethn. [Vol. 30 rattlesnake skin. This may be used by the shaman only during the ceremonies, but I think not. The tubular pipe is not Klamath. A long tubular pipe, with a small mouthpiece, found by the Indians at Pelican bay, had been described to my informant as the probable possession of a shaman who resided there. This is probably a trade piece. MUSICAL INSTRUMENTS The only true musical instrument known to the Klamath is the flute. This with three varieties of rattle and the hand drum make up the sum of their instruments. The musical bow is a plaything. The flute (slo'lus) is played for amusement and in courting; curiously enough it is played by women as well as men even when courting. Its music is called su'lulu. Various sounds are imitated on the flute, such as the noise of flying pelicans' wings. The melodies include those of shaman 's songs, although it is never used in shamanis- tic rites. I do not know that song is ever accompanied by flute. It seems more probable that the melodies of certain songs are considered appropriate flute music.141 There are flute songs about the turtle and deer. The song for the antelope is: gaiyo' gaiyo'wa gaiyo' gaiyo'wa chewing tcewom aie'nol teli'do6ga antelope me am browsing slewi's ai'e'nol te'li'do7yga wind me am browsing. He browses along (toward me?) when the wind strikes him ('?). A flute song for the mountain sheep is: hotuiks nis sewae't geoge'lE wie'so;m that allows me that earth mo4untain sheep doston hoiha'nsE lali'sdot where I run on the ledge on the steep hill. These are not magical songs for calling game but solely for amusement. Another song referring to any bird is: djlkgasltgo djlkgasiltgo nenea flutterer flapping. 141 Gatschet gives the text of several songs, which, because they are neither appropriate to shamanistic or puberty rites, may belong to this category (1:189, 192). Spier: Klamath Ethnography * This is the open flute, lacking tongue or reed, and played by blow- ing across the open end while it is held vertically or horizontally. So I was told, but it is more probable that the vertical is the only posi- tion used. These are manufactured from branches of the koko'dom bush or the elder (slo'luisEm), from which the pith is easily removed, or from the long legs of cranes. The wooden flutes are three-quarters of an inch in diameter, one to two feet long, with three holes (or four or five ?) for finger stops burned at odd intervals. The deer-hoof rattle (so'kts-oks) is probably the only one of the three Klamath rattles that is of any antiquity among them. It is used exclusively in the girl's adolescence rites. The Klamath form is a slender pole four to seven feet long with a cord along the upper third to which bunches of deer dew claws (V'o'djitqks) are tied at intervals. The butt is pounded on the ground to jar the hooflets. Of the other two types, the clap rattle is the familiar cleft stick, the split end of which is struck against the palm. This was used in the Ghost dance of 1870 and in no other connection. It was derived from the Pit river peoples. The notched rattle (ulo!ks) is known to have been played by one man who learned it from the Upland Ta- kelma. It may have been used in the war and scalp dances. This is the familiar notched stick set on the ground and rasped with another stick; no resonator is used.'42 The hand drum has no great antiquity among the Klamath but was in use before the coming of the whites. It was derived from some northern tribe, probably those of Warm Springs. Its very name, bo6mbom, is the common word in eastern Oregon and Washington. It is used only in the social dances, which are themselves of no great consequence. The drum is the familiar shallow tambourine affair, with only one head. Thongs crossed on the open face furnish a hand grip. Drums are constructed of a flattened juniper bough bent into a ring of one to two feet in diameter. Heads of badger and antelope hide are best; they yield the loudest sound. The drum was not deco- rated until lately, when stars were painted on the head. This is said to have been learned of the whites, but is more likely derived from the modern version of the Smohallah cult, the so-called Pompom religion of eastern Washington and Oregon. The drum stick (wo'- tanotc) is a stick wrapped with mink hide to form a knob. 142 There is no association of the sound with the voice of the frog.nor with the rain, nor is it used in social dances, as in the Great Basin area. 1930] 89 90 University of California Publications in Am. Arch. and Ethn. [Vol. 30 Several instruments whose presence might have been ex.pected are unknown. Such are the cocoon rattle, the foot drum, or any other form except the hand drum. As to the cocoon rattle I doubt that there are any large cocoons in this region. It is rather surprising that those who accompany the singers do not beat time on a plank with clubs after the fashion of more northerly tribes. A peculiar Wish- ram drum, a horizontal pole slung from the roof by a rope and pounded against the wall, had been played at the Dalles by one of my informants, but it was never introduced among the Klamath, To be sure, their houses do not have plank walls but alternatives might have been substituted. The musical bow should be mentioned here only for comparative purposes. The Klamath lack any special form but sometimes toy with an ordinary bow, holding the bow in the mouth and tapping the string with a stick. There is no name for this. Whistles (stu'ta) are also playthings. These are made of willow, of a plant called wa'kum, or a jointed grass called watco6'kwls. The willow whistles are seven or eight inches long and provided with side holes like a flute. Whistles are blown direct into the open end. Whistling with the lips was indulged in even before the coming of the whites.'48 Songs are not whistled. SOCIAL DANCES It is not clear whether there were any dances of purely social intent in earlier times or whether they were so rare as to be negligible. Two good informants said that formerly there were none. Indeed the use of the hand drum with social dances (yeka') suggests their recency. Shamanistic performances, which were evidently frequent, may have provided a substitute entertainment. Some time during the year a good singer will busy himself arrang- ing a dance solely for amusement, called k'si'ullks. Men and women form a circle with hands clasped, but the fingers not entwined. The step is a sidewise jump, not a shuffling of feet as is usual in circling dances among other tribes; the movement is in a counter-clockwise direction, I think. The circling begins with the song; they do not wait to sing a phrase before moving. Formerly, according to an unreliable informant, no musical instruments were in use but recently the hand drum was introduced from the tribes to the north. 143 So, too, the Northern Paiute, aceording to my Klamath informant. 10Spier: Klamath Ethnography Another circling dance, called sko'tcp-us, may not be a social dance but may be used in the scalp and war dances in a way that is not clear. Here those in the circle do not hold hands and their step con- sists in shuffling along by moving the toes and heels alternately. The headdress for dancers consists of long eagle feathers, or a bunch of small eagle and gull feathers may be kept in place on top of the basket hat by means of cord passing under the chin. Downy feathers are hung pendant at the back of the head. New Year's observances, or new fire ceremony, such as the Yurok and Karok know, are not found among the Klamath.'" ETIQUETTE The Klamath are on the whole a hospitable and friendly lot. Good breedi'ng demands that visitors be invited to share one's meal. There is no studied effort at ostentatious feasting, however, nor is food forced on a visitor in the manner of our eastern Indians. If visitors, men or women, arrive at meal time, the children of the house- hold are sent out on the pretext, as they tell the children, that the food will stick in the visitors' gullets. A kind of grace is said before eating. Bits of food are cast toward the spirit places to feed them first. I am not sure that the accompanying formula145 is a prayer. Good form does not demand that the visitors clean up their portions, nor does one hear the smacking of lips in appreciation. There is among them no objection to noisy eating. Children are told not to talk while eating fish, for fear of their swallowing bones. It is usual to rinse the mouth with water before and after eating. Soiled fingers are washed in a near-by stream after the meal. Children are definitely instructed in good behavior. The common salutation to the visitor is i gotpi'iE, yes, have you come?, followed by go'li, come in. On leaving, the guest announces i ga'mbin, yes, I am going, which evokes the invariable response i ga'mbli, yes, go. One hears molnskanditc, thank you, but rarely. For a man to remove a woman's hat from her head is reprehen- sible; it is usually construed as a solicitation. Certain behavior is not so much a matter of general good form as of sex difference. Egress from the earth-lodge is by a ladder project- ing through the hatchway overhead. Men mount the rungs facing the ladder, but women place their feet sideways, mounting crab-wise. 144 Kroeber, Handbook, 53, 102. 145 See p. 168. 1930] 91 92 University of California Publications in Am. Arch. and Ethn. [Vol. 30 Sitting positions are of wide latitude; of course they sit directly on the ground. Almost any position is assumed except sitting on the crossed ankles. Both sexes sit with their legs crossed before them, Turk-fashion, or with the legs extending straight before them, in which case women cross their ankles. It must be remembered that the common dress of both sexes is the string skirt. They kneel while paddling, both sexes alike, sitting on their heels. This is quite possible because the gunwale of a laden canoe is only a few inches above the water. The Klamath sometimes use the lips for unostentatious pointing, but they more commonly point with the forefinger. In a common position of rest, a seated individual places one hand on the other over the abdomen. To curse a man his dead are mentioned, not necessarily by name; in fact the reference may be quite indirect. In a tale the skunks say to Coyote, "You are the son of a person always on the lake," knowing full well that his father was dead. I was told of two men jocularly calling each other tc!ima'ntk, Widower. One named the dead wife of the other and was covered with embarrassment because the grandson of that woman, who was present, was provoked. An old man may be called wa'kontk, adolescent boy, or skotc, a man with a first child, in derision. This is nearly equivalent to a curse. Gatschet gives the expression sko'ksam kiifla gen 1, go to hell! (or more literally, go to the ghosts) which may well be aboriginal. But watchhgalam weash, you son [child] of a bitch! and watchagalam pe-ip, you old bitch!, strike me as apt translations from frontier vernacular.'46 146 Gatschet, 2:319, 477. Spier: Klamath Ethnography II. KLAMATH SHAMANISM INTRODUCTION Klamath religion centers so largely in shamanism that it may be permissible to describe the whole in terms of shamanistic practices, the spirits with which they are concerned, and the acquisition of supernatural powers. There is little ceremonialism apart from the shamans' performances. There is a rite for girls at puberty, a dance in preparation for war and another in celebration of captured scalps, some formal feasting at marriage, and of recent date the Ghost dance of 1870. The religion is only weakly developed. Power is sought from a host of spirits whose characteristics are not sharply defined. These are predominantly birds and animals, winds, lightning and the like, and a handful of anthropomorphic beings. They are however defi- nitely localized in many instances and partly identified with the figures of mythology. But there has been no attempt to marshal the spirits into an ordered pantheon. Relations to the spirits have no ethical implication. The belief in a land of the dead, too, stands apart from the spirits and their functions. It has solely a folkloristic existence, for it little concerns the actions of the living. Power is deliberately sought by almost every KIamath. There are times that are especially propitious; at puberty, in distress, or on the loss of a wife, a husband, or a child. The mode of seeking and the revelation are stereotyped: one fasts at night on mountain tops, run- ning about, piling up rocks, or diving beneath lonely pools. Mourners at least must first purify themselves in the sweat-lodge. The looked- for response is unconsciousness, a profuse hemorrhage at nose and mouth, and a spirit 's song revealed in a dream. The spirit is not the seeker's guardian and exacts of him no special conduct, no offerings, no venerations. There is only one obligation; he must accept the blessing and use his power once the spirit makes itself known. The power is motivated by the song; the song and the spirit are one and inseparable. There is little specialization of spirits and their powers: curing, gambling, hunting, love-making, and sha-manistic trickery are fields in which any spirit may operate. 1930] 93 94 University of California Pu4blitions in Am. Arch. and Ethn. [Vol. 30 The shaman himself is, or was, the outstanding figure of Klamath society. He had no rival in the chiefs, the rich man, until the coming of the whites brought a redistribution of emphasis in Klamath life. Chiefs had previously been persons of little influence. All this is familiar enough among Indians; the Klamath differ only in particulars. In its several aspects, their religion, as well as its practice, is one with that of the Great Basin, Plateau, and Mackenzie tribes. THE POWER QUEST The quest for supernatural aid is open to everyone, men and women, boys and girls. It is a constantly recurring feature of their lives. Everyone, perhaps without exception, seeks the powers at least once during his lifetime and further experiences seem to be repeatedly sought. The first occasion is always puberty. I doubt that boys and girls are expected to be uniformly successful in securing aid at this time, because, it was said, at sha.manistic performances those who had power might sing their songs. This may imply tha.t the others either never acquired power or had lost it. My informants' frequent emphasis on the period of mourning for a husband, wife, or child marks this, too, as a peculiarly pregnant period when power may be successfully sought. At least. at such a time a man loses his power and must seek again if he wishes to regain it. A man also seeks power at.the birth of a first child. But it is not only the.. crises of life that -provide occasion; the quest may be undertaken a.t any time, and it is the prac- tice of such as gamblers and shamans to renew and enhance their powers by frequent attempts. The quest is called spiu'tu. The experience does not ordinarily include a vision, but is made manifest. in a song heard in a dream. In some instances where a vision is had, a dream song follows, but it is clear that the latter is the essential experience. In the following case a man and a woman had identical visions but to the man alone fell the experience leading to power., The grandfather of a Klamath Falls shaman went to stY'ku6Otc, the point of Eagle Ridge, in the evening. There is a spot in. the lake where bubbles rise.I Horses lived in this long ago (!). When this man reached the headland he' saw two pinto horses, with long manes, rearing against each other from the water. He was not frightened but took off his clothes and plunge-d into the water in order to secure ,ppwer.. When he recovered consciousness he found himself again on land.. From that time he was lucky; he began to get many pintos, large horses. Because. of this experience his grandson, the shaman, also acquired, in his turn, spi'rit songs of all kinds. Spier: Klamath Ethnography An old woman, gathering wood at this place, also saw two horses in the water. Turning to her husband she said, "There are horses standing over there," but when they looked again, the horses had disappeared. Such a disappearance (txko'tka) is a sign that a relative will die; one of her relatives died after this. Visions evidently follow one of two formulae. Either, as above, the spirit appears standing in the water and then disappears, or it drags the swimmer below the water for the purpose of devouring him, as instanced below. One word swi'is means song and spirit being. A certain dualism is involved here. Spirits are definite beings that dwell in the moun- tains and in the water. On the other hand all songs are called swi'is, including those for amusement and non-shamanistic dancing, which bear no reference to spirits. That is, spirits and songs are certainly recognized as separate entities. Nevertheless, when it is said that a man has a Coyote swi'is, for example, it means that he has the appro- priate song or songs and that this animal is his supernatural helper. The spirit never manifests itself but in the song; the singer is the vehicle, the voice of the spirit. Song and spirit are one and the same thing. Power is sought in lonely spots on the mountains, in mountain pools, in eddies in the rivers, in all places where spirits are known to dwell. A boy is sent into the mountains on a vigil of several days, perhaps five, at puberty, that is, when his voice changes. He seeks power that he may acquire property, be a good hunter, become rich, a chief, and be able to do all the things that are difficult. (One does not need power to be a fisherman, for instance.) He must fast and must not touch his hands to his face, but must use a scratcher instead. He must sleep without covering a'nd warm himself only occasionally by a little fire. He runs about constantly throughout the night, p'iling rocks into high piles'47 (called sEwa'l) and swimming in the mountain pools. He prays, calling loudly to the spirits, and finally gets an answer. At night he may see a spirit with blood flowing from his mouth; then he faints and when he wakes he too will have a hemorrhage (djakglekg"e'ka). He "nearly dies" before he secures the song. It is possible that girls also seek power at puberty. Their dance at this time takes place at night; during'the day they sleep in the brush or on the near-by hills'. Then a song may be' dre'amed. One informant had it that they spend the night piling rocks like boys, returning at daylight, but I consider it very unlikely that a girl 147 Clarke, Rock Piles, 40, refers to the many piles of four or five-boulders placed on one another which occur in this country. 1930] 95s 96 University of California Publications in Aim. Arch. and Ethn. [Vol. 3( would be allowed to go unattended. Women well along in years, how- ever, follow the regular practice, at least as to bathing in the spirit places. The same method is pursued by the mourner, gambler, or shaman who seeks power, but swimming in the deep river eddies was more frequently mentioned in this connection. These seekers are almost always men, but older strong-willed women will attempt it. Power may be sought at any time during the year, but they do not dive for it until the first willow buds appear (in the seventh month, March or April). I presume that deep snows and ice in the mountains preclude earlier attempts. During the day the seeker sweats and fasts, waiting in the brush until nightfall. He then goes to the river and dives to the bottom in search of the spirit. He must not be frightened even if he sees some- thing moving under the water. He prays before diving, " I want to be a shaman. Give me power. Catch me. I need the power." Some- times he comes up unconscious, blood streaming from his mouth and nose. He must wait until the flow stops before going home.148 This night he hears nothing because he has been frightened by the spirit, nor will he eat the following day for fear of falling ill. On the following night or soon after, he hears the song of the spirit as it sings beneath the water. He may dream of hunting deer, acquiring horses, of success in gambling or with women, and of marriage. One informant dreamed of being in an earth-lodge. When he wakes the seeker sings about the subject of his dream. One song dreamed by an old woman is e'ha e'ha e'ha e'ha e'ha (repeated four times) noai'meso spu',qg`tg`ple' sp-i'vg6tg6ple' I am the one pursuing you. It seems that, like a boy at puberty, this seeker, after swimming, may go to the mountains to fast and pile rocks. He may remain there for some time, but one exceptionally rapid quest was cited. Nancy's father having lost a child, he went swimming in Crater lake; before evening he had become a shaman. A power quest under unusual circumstances is that of LEle'ks. This man, the most prominent and successful chief of the last century, was a berdache in his youth. His people wanted him to give up woman's garb. One day when he lay asleep they placed a woman's, 148GOatschet also describes this (op. cit., 1:82-84). 193]Spier: Klamath Ethnography canoe pole and a bow and arrows beside him. Shouting suddenly, they woke him and as he leaped up he seized the bow, thus determining his future life. Running off to Mount Scott he spent the night there. As he was returning he smelled the sweaty odor of running horses. In later life he was lucky in everything; much wealth, especially horses, came to him. Dreams of the same sort came to men or to old women without being deliberately sought. Should a man have several such visions, he may decide to become a shaman, whereupon he undertakes the cus- tomary fire dance of the novice. But if he chooses to ignore them, the power and the opportunity slip away. Last night I [Pat] dreamed a song: I am nearly a shaman. Yesterday I tried to catch fish but it rained too hard. I then went to town but I could not get what I wanted. So I returned home, ate, and lay down. I dreamed that my horses fell in the mire when I drove near the mountain. I wanted to save them but I could not raise them. I heard a woman 's song behind me. dje'udjls snalwas feathers all over woman Helen sna was [his daughter's name] woman The fish did not like me; they did not let me catch any. They made the bad dream. This is the same way as a shaman dreams. Another dream, though not of this type, may be inserted here. Last night I (Coley) dreamed I was chased by a big, black man; not a Negro, just hard to distinguish. He had nearly caught me when I woke and heard the chickens squawking. There is only a slight tendency to the inheritance of the shaman's art. The case cited above where a shaman received spirit songs because his grandfather had an experience is hardly to the point, since these experiences are so common that every shaman must have had a close relative who obtained power. But my informant's statement that when a powerful shaman dies his child may become a shaman is borne out by her own case. His spirits and his clairvoyant powers (dode'itks, dreaming) are released by his death and must go to some- one else. If his relatives keep his pipe or other paraphernalia about, it is certain that his spirits will come to one of them. Naney's father, who died before she was born, was a shaman. When she was quite small and could first understand, she began to dream of Frog, a big woman spirit. Sometimes she would jump up at night and run, afraid of the dream. The spirits persisted in trying to make her accept until she was grown; finally she heard their voices whispering close behind her. She was about twenty and had several children at the time. Her relatives did not want her to become a shaman. 1930] 97 98 University of California Piublcation in Am. Arch. and Ethn. [Vol. 30 Her older sister (i.e., mother's sister's daughter) gave her medicine so that the spirits would no longer bother her. Two of Nancy 's children became sick. A sha- man, Alec Wilson, diagnosed their ease as due to Nancy 's refusal to answer the spirits, to begin the shamanistic performance. Another shaman was called in to relieve her of the spirits. He did it very quickly; he had three or four spirits to lead each of Nancy 's "to their side." He used the songs of these spirits imme- diately, for they were now his. These two spirits were Thunder (lEme'is) and Falcon (Endo'gis). Had Nancy wanted to become a shaman she would have called the people together in order to demonstrate to them in a fire dance. A partial list of spirit places (spiu'tuks) in Klamath territory is gleaned from my notes. As I made little inquiry about the southern and eastern parts of their country, it happens that most of these lie in the north. They are about equally divided between localities in the Cascade mountains and deep eddies of Williamson river. Undoubtedly some of the rocks described as transformed people, animals, etc., are thought of as places where spirits reside, but as I was not told that power was sought there I have not included them iin this list. Ge'wuis, Crater lake. People were stolen and taken down into Crater lake by beings there. Some say they have found no water in the lake. Instead there were rocks as big as trees and deep tunnels in the bottom.. There are animals, snakes, and a sort of people who live at (or in) the ocean. Men swim in the lake at night to get spirit aid. A second site is ma'kwalks, a point of rock projecting over Crater lake from the western cliff. The seeker clambers down and piles rocks on"the point. On the northwestern side of Ghost's Nest mountain, which lies southwest from Crater lake, is a saucer-shaped bed of rocks over- looking the whole countryside. Lads built their fires near-by and lie in -this in order to see a spirit. On the eastern side of the mountain are piles of four or five large rocks, erected by seekers. Ho'tkak6onks, Mare 's Egg spring, is six 'miles north of Cherry creek on the western side of Agency Lake marsh.' They'swim here at night. Once men found no water in this pool but a big man, with hair'covering his body, standing in it. When they returned he had disappeared and the pool again held water. 'Pau'usnekas, 'Harriman ('?) mountain southeast of Mount 'Pitt. The Klamath say this is higher than Mount Pitt; indeed, being nearer, it appears higher. - -Two accounts pertaining to this mountain may refer to the same iheident. Long ago a Klaiiath climbed it, but he never returned, for he died of the dold wind although it was summer. A PelicAll bay man, who died long ago, once went to a lake on this mountain at sundown to seek power by swimming. He saw men in it, the backs of some, Spier: Klamath Ethnography the heads of others. He was frightened and did not swim but went home. As soon as he started home the thunder began to roll; it pursued him all the way. He was afraid of it, so he jumped into Klamath lke. He sat in the water for a long time before the thunder left him. He became a shaman because of this; he had the thunder spirit in him and sang the thunder song. Finally people killed him because he became so powerful that they feared he would kill them with his spirit. Sto'kmotc, the headland of Eagle Ridge, is on the west side of Klamath lake. Spirit horses live in the lake near the point, as instanced above. North of Modoc point are two groups of rocks. The first, red rocks on the hillside a half-mile from the station, represents a bear's entrails. Two big rocks somewhat farther north and on the plain are Wolf and his wife. When people mourn for a dead child, men go to the Wolf rock to seek power, women to his wife. They seek to be strong and to run well. They wrestle with these rocks. There is an eddy in which they swim a mile and a half above the mouth of Williamson river, opposite the cremation pile on the right bank. An otter (k!olt), bearing stripes, is seen in this. Many spirit otters of this sort are seen at k!o'ltamtcls, north of Klamath Falls. A channel cutting through a bend of this river several miles from its mouth is called koka'k'uit. A man saw a water snake (called to'ke) in this place. It was as big as a tree and had two straight, pointed white horns. He told a shaman, who said that, the snake having caught him, he could not be cured. This man died in two days. While this story suggests it, I am not sure that they swim for power here. Another informant stated that at Tcoka'lwas, just above this, is a huge horned snake (wisi-k, snake generically). Possibly the same one is meant. In a tale the hero Wawiwo'k!os cuts the snake's head off. Wit'amiumpsi, black bear's place, is at a sharp bend in the lower Williamson river (a mile above the highway bridge) overlooked by a high bluff. Long ago when men came to this place to swim they found no water there. Standing on the bluff they saw several bears below. Another man came to swim at sunset. He saw a bear standing upright in the water. " That is what I like. I will swim there, " he said. The bear disappeared. The man dived in but did not come up. He was found lying on the bank, blood flowing from his mouth. The bear had caught him and made him a shaman. So people try to get power there now. 1930] 99 100 University of Caifornia Public5ations in Am. Arch. and Ethn. [Vol. 30 K'tai'di, lava, is a cliff of that rock two miles below Chiloquin on the left bank of Williamson river. Seekers slept on this. Ska'mdi is an eddy in Williamson river above Chiloquin. This is a spirit place but it does not seem that they swim there. Once I [Pat Kane] and my two brothers camped near there just before sunset. We were on our way to Klamath marsh to pick pond-lily seeds. My oldest brother started a fire while the younger fetched water. I went to tie the horses. When I returned, my brother who had been to the river said, " There are many beaver in there." We took our guns to the spot. Soon a beaver swam close to us, only a few yards away, but we could not hit it. Soon two more came swimming close by. We failed to kill these. Then five eame and soon so many more that they com- pletely filled the river. My brothers said, "We had better not stop here. Let us get away." We packed the horses and left at onee. I saw the water shakingisO and the sound of thunder. It took all night to get baek. My oldest brother got sick; no shaman could cure him. He was sick all summer and in the winter he died. Neither my younger brother nor I were sick. A man and boy were going to Klamath marsh afoot. The boy was six or seven. The man wanted to try swimming in Ska'mdi. He told the boy, "If those things bite and kill me, you go home." The man had lost his wife; the boy alone was left. "I do not want to stay here; I want to die," the man said. The boy cried to him not to swim there. He walked back from the river bank, ran and plunged in. Big animals reached up and bit him. They came up twice; then he appeared no more. The boy saw it and wept. He started home; it took all night for him to reach BEziikse'was below Chiloquin. At this place the water boils and cannot move down the river. No one can swim there; something always bites and kills them. SPIRITS, SOULS, AND GHOSTS There is hardly a mile of Klamath territory but has its mythical reference. Everywhere there are personages of the folk-tales, now transformed to rocks by Crow's laughter. Spirits are legion and in many cases are localized, so that one looking over the countryside finds it rich in religious connotation. But despite its comparative richness, Klamath religion is quite unsystematized. There are no deities and the spirits as such hardly figure in the tales. Most of the actors of the tales are not spirits and this is especially true of the principal ones. There is little suggestion of any fixed relation between such as do figure there. In other words, the Klamath lack all architectural feeling in conceiving their religious beings. Spirits form a uniform class; no one of them can be said to be especially more important than any other. Souls and ghosts are separate entities, and again are sharply distinct from spirits. Spirits 150 It is conceivable that an earth tremor dislodged the beaver. Abi l / \ 1930] Spier: Klamath Ethnography 101 (swi'is) as we have seen are beings that live in the mountains, in river eddies, and in lakes, where they are sought by those in quest of power. Souls are personal and after death go to the land of the dead, No'lisk'n. The soul leaves only at death; illness is not ascribed to its loss. Ghosts (skuks) wander from the land of the dead where souls. have gone. It was not said that ghosts are souls in other guise nor were the ghosts referred to those of particular persons. The juxta- position is rather like that of our own folk thought. One word hoki's means soul, breath, and life. It is intelligible then why one of my informants said that all animals as well as men have souls. It is the quickening of life rather than spiritual personality. that is implied by the word. They say of death, hoki no'tsna no'llskan, the life leaves for the land of the dead. Its seat is within the heart (staii'nas); outside is the flesh (djole'ks). When the body is cre- mated, the soul goes to the land of the dead. The flesh is destroyed; the ashes and smoke blown into the air are carried back to earth by rain, to the country of which the body is a part. Everything grows well then. But the soul never returns.'51 Ghosts return from the land of the dead because they want to live again. They are constantly about seeking to catch someone's soul to take to the land of the dead. (So I was told; but there was never a hint when they told me of curing that sickness is due to loss of the soul.) It makes no difference that one has spirit power, his life may be taken. This is most likely to happen to one in mourning as he sits alone weeping in his house at night, for ghosts usually come at night. Next day he will die. This happens commonly enough; even little boys at play see them and die at once. Shamans can discover this. Whatever they are like in the land of the dead, ghosts that return neither resemble humans exactly nor are they skeletons. Today some people talk without fear with ghosts, formerly they would have died. Possibly my informant meant by this that members of his own Shaker sect or adherents to the Ghost dance cult could talk to them with impunity, but he may have had in mind only the prohibition of the folk-tales against talking to the dead as they are being transformed to living creatures. 151 Gatschet records an obscure statement which he interprets: "The Maklaks [Klamath] believe that the souls or spirits of the deceased pass into the bodies of living fish; they become inseparably connected with the fish's body and there- fore cannot be perceived by Indians under usual circumstances" (1:130). The word used is however skuks, ghost. I received no information of this sort. 102 University of California Publcation.s in Am. Arch. and Ethn. [Vol. 30 Last month an old woman, who was very sick, was left alone. As she lay there she saw ghosts. They caught her; she died. When her children returned she was dead. Her daughter and grandson rubbed her chest to revive her. She was soon restored except that one ear, which the ghosts had touched, was "closed."I Her daughter asked, "What is the matter?" She answered faintly, "The ghosts caught me." Today she can barely talk; her sight and hearing are gone. A Klamath Falls man met a ghost on the trail. It was short and had long white nails on hands and feet. The ghost said, "Let us trade nails." The man said that he could not take his off. The ghost said, "I can pull them out," and exchanged nails with him. The man went home with the long nails and went to bed. When he woke, he looked at his hands and feet, but he had his own nails again. The same man saw ghosts dancing at the cremation place on the hill east of Link river (Klamath falls). He watched them dance all night. They gave him a pipe to smoke. Then they threw firebrands at him and he ran away. About ten years ago I [Pat Kane] had been swimming in the river. I was returning to my home with some horses when I heard crying. I saw an immense fire at the cremation place on lower Williamson river. Men, women, and children were wailing there. I held my hat over my face and did not listen. I did not tell my wife. For three or four days I was sick, so I went into the sweat-lodge. Later I went down there but found no signs of a recent fire. The names of the dead are mentioned rather freely, but no one likes to name his own relatives or hear another do so. In fact, such a reference is insulting. But this is not due to any fear of the dead; it is rather a matter of taste. The land of the dead, No'llskan,152 lies in the west, the sunset. Everyone goes there. It was created by the culture hero, KEmiu'kiumps, who made the humans that figure in the tales (called psaudi'was) and ordained that they should go to this land when they died. Nothing is said of the road thither but the characteristics of the land and its people are tolerably well conceptualized. Everything in the land of the dead is inverted from this living world. The ghosts sleep by day and live at night. In the day there is nothing to be seen but their houses and skeletons. At sundown they are clothed with flesh again; they rise and build fires. They dance through the night to fall asleep when dawn breaks. They feed on swana's eggs and perhaps they do not eat them in the manner of humans, since a tale suggests that KEmui'ku-mps is fed in the manner of the ghosts; the eggs are thrown at him. Further, the eggs are unnatural in that they are hot.'53 So far as the Klamath are con- cerned, swan's eggs themselves are mythical, for while swans cross 152I suggest that the word means, literally, land of the departed; compare no'tsna, departed for ; nutsnots, soul leaves; tc&kliin, service berry land; ca'stilkan, country of the Shasta. 153 Perhaps the Klamath do not eat eggs at all. I neglected to inquire. 0Spier: Klamath Ethnography the Klamath country in the winter, they never nest there. The eggs can be gathered only in bottomless baskets. The ghost's elks are water- bugs to mortal eye, and the flesh that they take is in reality bone. All their possessions are things that were destroyed with the dead, houses, baskets, beads, etc. Their hospitality also provokes them to throw firebrands at their visitors. Perhaps the very fact that KEmiu'kiumps is called out of his name by them in a tale is an attempt at inversion. There is a wide variety of spirits from whom power is sought, animals, birds, reptiles, fish, natural phenomena, and a handful of anthropomorphic and purely mythical beings. The animals, etc., are not individuals of living species, but mythical beings, either immensely large or very small. Such anthropomorphic spirits as are definitely conceptualized are also tiny. Although they speak of all of these as individuals-they say for example that a man's spirit is Weasel or it is Wlle'Ekak, dwarf old woman-nevertheless there remains the suspicion that they are rather genera, for the same named spirit is described as dwelling in several localities and belonging to several shamans at the same time. The number of spirits is indefinitely large; the following partial list is made up from various data given by my informants. The animals include grizzly and black bears, wolf, coyote, dog, mink, weasel, and fisher (?wa'lklotcka). The birds are the horned owl, bald eagle, golden eagle, fish hawk, a smaller hawk (tcl'ktu), a little hawk (wl'tgotgis), falcon, crow, magpie, buzzard, pelican, red and white shitepokes, butterball duck, a small sea gull (k!.otya'was), flicker, prairie chicken, an unidentified black bird (nao'llins), a blue and brown one (gau'wi), and a tiny speckled bird (dj'iske'gE). In the fish and reptile class are the rattlesnake, a big speckled sluggish snake (wut'mEnuiks), lizard, frog, a water snail (k!awi's), salmon, and trout. It is obviously the predatory animals that are chosen, with the omission of the cat tribe. The bird list also includes all the birds of prey and carrion eaters. Even though the list is incomplete, there are some significant omissions, such as deer, rodents, the song birds; and most fish are absent. The two fishes included are the largest repre- sentatives in Klamath waters. There is little suggestion of differentiation of function and power among these spirits. To be sure, it is only through certain spirits that curing and the various shamanistic tricks can be performed, but my informants never once volunteered the statement that these were espe- cially powerful. The exception is Frog, described by Nancy Phillips 1930] 103 104 University of California Publications in Am. Arch. and Ethn. [Vol. 30 as the most prominent and powerful of spirits, one on whom all shamans call, and whose presence is necessary before they can begin their curing ritual. This confirms Gatschet's information.'54 The Frog spirit is interesting on its own account. It is conceived as a big woman and so closely associated with pond-lily seed (wo'kas) as to lead to the suspicion that these are interchangeable concepts in the Klamath mind. It should be noted that the preparation of this staple is peculiarly woman's work among these people. In their mythology too, Frog figures as a woman. The last association is so widespread, perhaps universal, in North America, that the only special Klamath elements in the triple association are with wo'kas. While I am not certain, it is probable that this is the spirit which was described by Pat Kane as djEdje'llstcl, whereas the common word for frog is ko'wE. Gatschet credits a similarly important r'ole to another spirit.'55 "Another class of spirits embodies the spirits of those animals which have to be consulted by the ki'uks or conjurer when he is called to treat a case of disease. Such persons only who have been trained during five years for the profession of conjurers can see these spirits, but by them they are seen as clearly as we see the objects around us. To see them they have to go to the house of a deceased conjurer, and at night only. He is then led by a spirit called Yayaya-ash appearing in the form of a one-legged man toward the spot where the animal spirits live; this specter presides over them; there the conjurer notices that each appears different from the other, and is at liberty to con- sult them about the patient's case. Yayaya-ash means 'the fright- ener,' and by the myth-tellers is regarded as the Thunder or its spirit. " I received no suggestion confirming any part of this. The mythical beings are also probably genera rather than indi- viduals; to'ke, a horned water snake, Wile'Ekak, a dwarf old woman, TcaIkia'k, a little boy, and Goga'nE, a dwarf man. There is also a nameless wild man (the spirit for the trick in which blood is magically produced) and a woman spirit (for curing). Wawiwo'k!os, a mon- ster-killing hero of a tale, figures in a shaman's song and hence may be a spirit. Wile'Ekak is a dwarf old woman whose name is derived from wlla'e'ks, old woman. She is about thirty inches high, and wears a basket hat and long buckskin dress. In one form she appears as a rock on the northeastern slope of Mount Pitt. She controls the west 154 Gatschet, 1:ci, 180. 5S 1 :XcViii. Spier: Klamath Ethnography wind; it is really that she breaks wind. They shout to her to stop the wind when it blows too ha.rd, to give them a stern wind to drive their canoes along, or to blow the mosquitoes away from Pelican bay. She also lives at Rocky point on Pelican bay and at Kowa'cdi, at the mouth of Wood river, as the following tales156 relate. There were two young girls playing with pond-lily seed at Rocky point. This old woman came and spilled the seed all about. So they fought her, but while they were doing so, they fell over dead. When the two were found, shamans were called who cured them. These two told Nancy about this; they died only recently. There was an old woman sleeping at the same place. She felt something very heavy sitting on her chest. She put her hand there and caught the wlle'Ekak. Still holding it she started to light a fire, but by the time she had the fire lighted, it had vanished from her grasp. An old man at Kowa'cdi was going home. He saw the wile'Ekak sitting in his way, grunting. He was a shaman. He went right past it but it never harmed him. This sort of thing has happened many times. Tcalkia'k may be seen anywhere, probably most frequently in water. There is one, for instance, in lower Williamson river, above the cut-off channel. One of the upright rocks at the cremation place at Pelican bay is a tcakia.'k. The name is also derived from an age term, tc!a'ki, a small boy, but the spirit is said to be no larger than a baby. He has long hair which hangs to his waist in back and he is entirely nude. The wooden images of Tcakia'k set up by the shamans, however, have red-sha.fted flicker feathers around the neck. Goga'nE are dwarf men seen in many places but commonly fre- quenting the Cascade mountains. Their footprints are no longer than a baby's. While all my informants spoke of these beings, it happens that the following experiences (except the last tale) were told Nancy by an old Paiute resident of the reservation. This m,an saw a g6gna'nE while hunting in a canyon near Snake Camp. The little man stood up, shaded his eyes with his hand, and looked around. There were some Paiute at Snake Camp who were going to dig Indian potatoes. They saw a baby walking around. One woman picked it up. The g6ga'nE carried her far off to a cave in the mountains. A Paiute woman was cutting willows. She left her little baby near-by while she worked. A g6ga'nE came along and, devouring the baby, substituted himself in the cradle board. When the mother came back she thought it was her own. She picked it up and began to nurse it. The baby would not stop suckling. She pinched its mouth but it would not stop, it only smiled. When she got home she heated an iron and put it against the baby's mouth, but still it just smiled. They could not get the baby away from her. Then they built a big fire. They cut her breast off, but the woman immediately died. They threw the baby into the fire, 156 Told by Nancy Phillips. 1930] 105 106 University of California Publications in Am. Arch. and Ethn. [Vol. 30 but at once it jumped out and ran away. This happened after the whites arrived in this country. There are other "bad stories" about the g6galnE. Two little men, gbga'nE, made a big pit and chased a deer into it. They saw the deer standing in it. One of the goga'nE climbed down and held on with his arms around the deer's neck. The deer bounded out. The g6ga'nz had a bow and arrows on his back, but he lost them when the deer jumped. The deer started running. The other g6ga'nn followed. He found the first dwarf's hat, then his moccasins. He picked them up. The one who was following called, "The deer is carrying him." The one who was hanging on choked the deer to death. Gatschet undoubtedly refers to Goga'nE in the first of the follow- ing paragraphs, despite the apparent difference in name. That which he gives, na'hnias, was evidently known to him only from a song and sounds much like a form adopted for euphony's sake. The second dwarf, Gwinwin, is unknown to me, unless it is Goga'nE; the manner of Gatschet's reference suggests that it was named in a myth. "A miraculous dwarf is mentioned under the name of na'hnias, whose footprints, as small as those of a child, are sometimes seen upon the snow-clad slopes of the Cascade Range by the natives. But the dwarfish creatures who make them can be seen only by those initiated into the mysteries of witchcraft, who by such spirit-like beings are inspired with a superior kind of knowledge, especially in their treatment of disease. The name is derived either from nena, to swing the body from one to the other side, or from nainaya, to shiver, tremble. "Another dwarf genius, about four feet high, Gwinwin, lived on Williamson River, where he habitually sat on the top of his winter lodge and killed many people with his black flint hat. He is now a bird. '157 Various natural phenomena are spirits: clouds, thunder, lightning, the sun, moon, and stars, and each of the seven winds (one for each of the compass points except northeast). Rainbow is also a spirit, or rather it is WIle'Ekak's spirit, and, to add to this confusion, it is said that she swallows it like a conjurer's sword. Earth and earth's child, the snake, are also spirits. It is not quite clear how certain other things are spirits. A piece of wood which is fastened to the pole that stands before a shaman's lodge is a spirit called wo'kda'lls. I could not discover that it is any- thing but the stick itself. Ha'nowas is a pestle-shaped stone which stands at Diuno'kai village on Pelican bay. It is about eighteen inches high and five in diameter with protuberances called breasts. '57Gatschet, 1:xcix, 163. ]Spier: Klamath Ethnography This is a spirit. Doubtful as spirits are certain other stones; yati's which stands at I'wal, and n'e'knuk, a group of stoines back of Du" llkiut, both towns on Klamath marsh. These are "shaman stones"; evidently once living things, for it was said that they were transformed when Crow laughed at them. A Molala struck one of the ne'knuik stones with a basket, saying, "How can this rock have power?" He soon sickened and died. The ne'knuik is a shaman. SHAMANISM The shaman (kiu'ks) is one who has acquired more than usual spirit power. This is a relative matter; some shamans have consid- erably more power than others, and everyone who has got power is in* some degree capable of using it as a shaman does. One of my informants, for example, now avowedly a member of the pseudo- Christian Shaker sect, maintains that he too is a little qualified; he can make it rain. Shamanism is completely individualistic among the Klamath; there are no shamanistic organizations. It is clear that the shaman is the most important individual in Klamath society, taking precedence over rich men, the chiefs. Sha- mans' houses are the largest in the village and they have at least as much property as wealthy laymen. Invariably my informants spoke first of shamans in describing outstanding individuals. Their behavior and practices were readily formulated, whereas I experienced consider- able difficulty in learning the functions and even the names of chiefs. It is reasonably certain that the status of the chiefs, that is the rich men, was relatively unimportant until the introduction of horses in the middle of the last century stimulated trading with the Columbia river tribes and perhaps brought in some of the northern view of the importance of wealth and chieftaincy. There is no doubt that the position of chiefs was strengthened by the habit of the whites of deal- ing with chiefs and other prominent laymen. Shamans are both men and women, but unlike the Shasta, the majority are men. My informants mentioned many men shamans during the course of my visits, and this probably did not exhaust their memories; but on request they could muster only four women shamans and one berdache. Women never acquire the prestige of men for they are.rarely credited with such unusual powers as men. They cannot wander in the mountains nor dive for power so well as men; 1930] 107 108 University of California Publicatioas in Am. Arch. and Ethn. [Vol. 30 they are more easily frightened. Nevertheless two of my informants said that women could cure more readily than men. All four women held the regular shamanistic winter performance, and cured as well, but only one of them had power to find out if another shaman had poisoned her patients, although all male shamans can do this.158 The berdache White Sindey claimed to be a powerful shaman but no one believed her. There are no specialized types of shaman. The weather shaman, the rattlesnalke and bear shamans of Californian tribes are not known here. There is however a suggestion; wounds made by grizzly bears can be cured by songs of a huge spirit grizzly according to one inform- ant, although this was scouted by another. Weather is controlled by the songs of a special group of spirits, but, as elsewhere, this power is exercised by shamans in addition to their usual curative capacity. There is no special power for curing snake bites. It may happen that some shamans have only the power to control the weather or to cure bear wounds, but I heard of none. It is clear, however, that the Klamath would not single them out. A shaman has powers from many spirits and is always accumulat- ing more. I could not get my informants to specify which powers belonged to particular shamans. Beyond specifying what songs a shaman used in a particular instance, they implied that an experienced shaman knew an inexhaustible number or would doubtless claim to have them at his command. Hence several shamans have songs from a spirit at the same time. The songs are probably not identical. At the same time it is not likely that two shamans will be performing at once, so that the problem of conflicting claims to possession probably never arises. I am also uncertain whether spirits are thought of as individuals or genera; even in the case of anthropomorphic beings, I believe they imply a plurality. Hence they would doubtless resolve the difficulty by maintaining that they had the same songs from dif- ferent spirit coyotes, dwarfs, winds, or whatever. New spirits are always being sought or come unsolicited in dreams. A spirit, like that which gives the power to swallow fire, "comes stronger and stronger" until the shaman feels compelled to perform this trick. If he fails to obey the insistent spirit he will die. 158 This is the case described by Gatschet (1:6468) and also cited by my informant. Spier: Klamath Ethnography A shaman, kiu'ksmag (Shaman Man), was working in a hayfield for some whites. He lay down saying he had a headache. His spirit, the buzzard, was making him sick because he did not heed it. He told his companions to fetch him a snake. He took the little watersnake by the tail and swallowed it. He said he was cured; the snake was no longer in him but had left his body immediately. The shaman is possessed during his performances. He is the vehicle of the spirit; the spirit sings with his voice, sucks with his lips, and sees withohis eyes. It is not. in him at other times, but in its home in the mountains or under the water and must be called on to enter his body. The shaman's behavior was instanced as follows. A shaman, Alec Wilson, was singing of his coyote spirit during a winter per- formance attended by my informant Nancy, then a little girl. He chuckled for quite a time, saying, "There are lots of swamp mice here." Then he crawled across the lodge through the audience and stealing up behind Nancy bit her hard. He bit her because she was moving around. A trance state is not part of the shaman's behavior, nor was a frenzy described. The stereotyped physiological disturbance, real or feigned, is blood gushing from the mouth while he falls unconscious. The absence of the trance state suggests that the shaman is not of necessity psychologically abnormal. The shaman's house is always an earth-lodge and practically all performances take place within it. Not only larger than the ordinary dwellings, it differs from them in being decorated. It is the usual semi-subterranean structure with a conical earth-covered roof resting on four centra.l posts. The entrance is by a hatchway in the roof with the ladder at the east end. The fire is kindled directly under this opening. The posts are painted with horizontal stripes, alternately red and white. They are not carved. According to one informant, men's faces, snakes, lizards, animals of all sorts, but not spirits, are painted on the posts; but I doubt this. The grass-stuffed skins of various animals and birds, probably all of them representing the shaman's spirits, are hung under the slanting roof beams; mink, skunk, crane, tcauwilo'ks (a duck-like bird), butterball duck, white shitepoke, rattlesnake, a big speckled snake, and the dried skin of the lizard were mentioned. The birds ha.ve outstretched wings. Standing on the roof of the lodge, on the northern side of the hatchway, is a wooden image of teakia'k, the little boy spirit, made by the shaman. Head, trunk, and arms a-re rudely carved and pa.inted. Red-shafted flicker (yellowhammer) feathers are fastened around its neck (only shamans wear the feathers of this bird). 1930] 109 110 Uni.versity of California Publications in A4m. Arch. and Ethn. [Vol. 30 The Field Museum collection contains a carved wooden image (no. 61763) of a woman bearing a child on her back. This may be one of the figures set on a shaman's house, although I was not told of such. It is about forty inches high. The carving is rudimentary; there is nothing to indicate how the child is carried. The ear lobes of the woman's figure have been pierced and a bit of cloth inserted. The stuffed skin of a blue crane is also propped up outside. I presume these are pla.ced there only when the boy and crane are the shaman.'s spirits.159 Gatschet's informant mentioned a red fox skin hanging on an oblique rod set up outside the lodge.160 A tall slim pole, called wa"lasdiu'pka (wa"las, standing up), with horizontal bands of red, is erected in front of the house, that is to the east, about fifteen feet distant. A long cord pendant from the top bears at inter- vals mixed bunches of feathers from eagles and other spirit birds. A miniature gamblin-g tray tied to it belongs to a woman spirit, and near the top is a piece of wood representing the spirit wo'kda'lls, which so far as I could learn is wood. Sim,ilar feather-bearing cords hang from the house posts inside the lodge, but here the gambling tray is one of standard size. All such articles, his special decorations, regalia, pipe, and baskets, are known as the shaman's mtu'lwas, paraphernalia. A shaman's outfit "used in certain curing rites" in the Field Museum of Natural History (no. 61564) consists of a cord at intervals on which are tied alternately short sticks and wooden tablets. The short sticks are reeds, nine in number, four to seven inches long. These have marks burned (?) transversely or spirally on them. The tablets are rectangular, with rounded corners and with a short handle at one end by which they are tied to the cord. These measure about five and a half inches long by two inches wide, and are painted red except for the section nearest the cord, which is black. The two-strand cord is also red. The shaman's dress includes a band of woodpecker scalps worn on the forehead or around the neck, or a bunch of red-shafted flicker (yellowhammer) feathers attached to the top of the hat or worn as a necklace. Shamans alone wear these. These are the familiar orna- ments of northern Californian dancers. The forehead band reaches just behind the temples; the woodpecker scalps are fastened on a buckskin foundation. The quills of the flicker feathers are doubled over the cord on which they are strung. The shaman's body is painted red during the winter performances, but not at all when he is curing. That he is barefoot emphasizes the recency of moccasins. The shaman puts on special dress to cure only if the case is urgent.' He then dons a mink or weasel skin hat with feathers of all sorts attached, a special 159 The Modoc also place them on the lodge, I was told. 160 Gatschet, 1: 71-72. Spier: Klama.th Ethnography ('?) buckskin shirt, and wrist and neckbands of weasel. Some sha- mans wear a necklace of grizzly bear claws. (Other men never wear bear claws.) His face is painted black. Then he is powerful to cure, sawi'gan gina. Gatschet describes a band of otter skin worn round the neck and bearing various shells, feathers, bird scalps, etc.; hence it "must unite the magic powers of them all.'I161 Some articles of shaman 's regalia are in the collections of the Field Museum. A bear-skin cap (no. 61563 ?) consists of the head skin, decorated on two or three sides with circular disks of buckskin ornamented with quillwork. These are one and three-quarters inches in diameter, and bear three concentric equidistant rings of quillwork, the most central containing a cross of the same. A shaman's wand "used in curing ceremonies" (no. 61676) is a three-foot shaft bearing bunches of feathers bound on at two points, with a bundle of dew claws tied to the butt. The feathers are gray; they are bound at points about one-third of the distance from each end of the shaft. The bindings and the shaft have been painted red. The dew claws are pendant on individual thongs, each of which is wrapped with quills at two points. The prescriptions affecting the dress of the laity at the shamanis- tic performances are illuminating. Everyday dress, at least for wealthier individuals, is of the Plains type; moccasins, skin shirt, long leggings, and breech clout for the men and long skin dress for women. Poorer people wear skirts of tule or buckskin and are barefoot. But the latter garb is obligatory at these dances. This is clearly another instance of conservatism in the field of religion, for the marginal position of the Klamath to the area in which Plains garments are worn, and the known recency of manifold contacts with the Plains- like tribes of the Columbia suggest that Plains dress is a recent acqui- sition. The hair hangs loosely down the back but is tied about close to the head. As usual basket hats are worn. Dancers, however, wear hats made of animals which are spirits. Taboos and other restrictions on shamans are only feebly developed. A shaman must avoid a menstruant woman; in fact, women who live in a shaman's house must leave their homes at these periods, whereas otherwise this is not usual. Etiquette requires that anyone crossing the lodge must pass in back of a seated shaman, at least when he shows signs of communing with his spirits or is eating. This holds whether a performance is going on or not. Otherwise blood gushes from his mouth and he falls unconscious. One informant had seen this happen to the berdache shaman White Sindey while she was eating. The result is the same if anything falls in front of a shaman. There are l1 Op. cit., 1:172. 1930] ill 112 University of California Publicationm in Am. Arch. and Ethn. [Vol. 30 no food taboos; shamans and others can eat their spirit animals. The only taboo of conduct peculiar to any spirit known is the prohibition against wearing footgear by one who has a dog spirit. For example, a Molala-Klamath halfbreed, the wife of the chief L''lu, who dreamed of this spirit, never wore any protection for her feet even in the winter snows. The shaman is cremated and his house burned like other Klamath. Usually his regalia, feathers and skins, are destroyed with the house. This must wait for calm weather; then only an old woman dare burn them. Immediately a storm rises, with strong winds and thunder, but lasts only a night. When the shaman dies and again when his body is burned, storms rage for several days. THE SHAMANISTIC PERFORMANCE While curing of necessity takes place at any time, shamanistic per- formances are given at only one time during the year. This is a name- less period, the fourth month (December-January), in midwinter,'62 the coldest period when the lake is frozen and the timber cracks with the cold. By this time all the earth-lodges are reconstructed and the Klamath are settled in their winter quarters. This is a period of enforced inactivity, what with the intense cold, the deep snows, and their inability to hunt and fish. Furthermore, the seances cannot be held any later in the winter for then they begin to run short of pro- visions, I was told, which is undoubtedly true and yet may be a rationalization. This brings to mind the winter sacred period of the Northwest Coast tribes, with its quiet dictated by a fear of profana- tion. It is not sacred to the ordinary Klamath; spirits are sought at any time. It is significant only to shamans and others who have spirits. Shamans make a point of giving their performances every winter. A number of them may join for the purpose. The novice waits for winter to prove the coming of his power and only after this does he attempt to cure. The advent of power is called dodi'kalgot. Non- shamans who have a little power, perhaps a single spirit, may sing their own songs at a shaman 's performance if the spirit impels them. The shaman informally'63 invites the participants to his home for 162 There is a discrepancy here in that the Klamath reckon a year of ten months, and these lunar periods begin with the lunation of August. My informants made it quite clear, however, that, although they specified the fourth month, they meant the period about Christmas. e 163 Knotted invitation strings, for instance, are unknown. Spier: Klamath Ethnography the winter dance. They bring food for a feast, especially pond-lily seed contained in large boiling baskets. This is left outside the lodge until the close of each night's dance and may not be touched for fear of sickness. Children are excluded except from some of the sleight-of- hand performances. The invitation includes men and women, not necessarily relatives, who know the songs and are willing to help. A man is chosen to shout during the performance, calling the spirits from their distant homes to help the shaman. Another is more care- fully chosen to act as the shaman's interpreter. This speaker or interpreter (lodo6tka) functions both at this per- formance and at curing practice. The shaman always asks the same man to act for him; obviously an important element in his success. Nancy's husband, for example, was always speaker for White Sindey, the berdache shaman. Older men are preferred and never women. The shaman holds his clenched fist over his mouth while talking with the spirit's voice, and while he talks no esoteric language but good Klamath, he shouts and whispers so rapidly and incoherently .that his words must be repeated by the speaker. This assistant also prepares and lights the shaman's pipe. Evening and morning during the per- formance he stands atop the lodge and calls to the shaman's spirits in the mountains. He is paid in blankets and beads. Clownishness is not part of his role. The audience sits close to the walls, each family by itself. The shaman reclines on his sleeping mats which are usually on the northern side of the lodge, but sometimes on the opposite side. This is his place on everyday occasions. His family is grouped back of him and his speaker is at his side. He dances on the northern side in front of his bed or rather in the central space between the four posts.164 This is swept smooth and no one is allowed to cross it to reach the ladder, for instance, even when there is a lull in the singing. The singers are seated on the southern side. The performance, called wa'hla, takes five days and nights, of which the first day and night are the most significant. They begin in the early morning with the fire dance. Five piles of dry non-resinous wood have been placed at intervals outside the central space. The pile to the south of the foot of the ladder is first used. The shaman dances on the south ( ?) side of the huge fire as close as he dares; the men who help him, dance on the opposite side.'65 Their step may be a 164 There is no dance platform, such as the stretched elk hide of the Wishram, which my informant had seen. 165 Their positions are probably the reverse. 1930] 113 114 University of California Publication in Am. Arch. and Ethn. [Vol. 30 jump (sideways ?) with the feet together, but I am not sure. Women may dance if men are few, but all the audience joins in the singing. As the fire dies down it is replenished with each of the four remaining piles in sinistral order, the last being that north of the foot of the ladder. This occupies the whole day and is not repeated during the five-day period. Its purpose is to call the spirits to act through the shaman, to give him the strength to perform the feats of legerdemain that follow.166 The nights are given over to these shamanistic tricks. Two tricks, swallowing fire and arrowheads, are part of every shaman's stock in trade. They are performed by the novice during the first night, though they may never be repeated in any subsequent winter cere- mony. Until the novice has made the ceremony to show that he can carry off these two feats, he is no shaman and cannot cure. The fire-swallowing trick (ko'pg-us p'an, torches eating) is per- formed with the aid of an old lake trout spirit (tew'm or m y), or rather they say the spirit eats the fire. Five men stand in a row each holding a bundle of resinous wood slivers, ten inches long and two inches thick, called ko'pguis. The shaman dances in front of them, others dance vigorously with him while they sing, ko'pg s babi's gistsa'mpka I eat fire while I walk. geno gita gistsa'mpka This is where I have been walking. The trout is represented as singing. The torches are lighted, the shaman snatches them in quick succession, puts out the fire with a gulp, and flings them aside. The companion trick is swallowing arrowheads; this follows fire- swallowing after an interval. The fire has been allowed to die and is only rekindled as a little light is needed. Perhaps twenty obsidian arrowheads are tied at intervals on a cord, which one man feeds to the shaman as he dances. He swallows these with the aid of certain bird spirits, or rather the spirits eat the blades: gau'wi, a blue and brown bird, w6'tg'tgis, a spotted hawk, and k!o!t'ia!was, a small sea gull (noisy, white, with forked tail and black spots on head). A tule mat is arranged like a cape over the shaman who now seats himself. They sing the songs of the bear spirit, now represented in the shaman, to bring back the arrows. Soon they hear him cry out, "ha," as each 166 Briefly described by Gatschet (1:70). Spier: Klamath Ethnography blade flies back and sticks into the mat. This performance is called stopswa'lc. A variant of this was ascribed by Coley Ball to a shaman of years' standing. He performed it on each of the five nights, which a novice would never do. I saw a shaman dancing in an earth-lodge. Then two men swung him around head down through the hatchway, each holding by a leg. Two men below held the ends of a string to which five obsidian blades were attached at intervals of a foot. He swallowed the blades and cord, and became unconscious. They rolled him in a mat, bound it, and slung the bundle under the slanting beams at the north side of the lodge so that his head was to the east. Then the women sang. Suddenly he cried, "ha," five times and with each cry a blade appeared stuck in the mat. But the cord was not seen again. They took him down and laid him on his bed. Then he began to sing. The morning following this dance everyone except the shaman goes to bathe in spite of the ice that may be in the water. When they return to the shaman's house the food is brought inside and eaten. The shaman, however, must fast through the whole five-day period. The food, or at least the pond-lily seed, which has stood outside the lodge for twenty-four hours, is thought to be eaten first by the frog spirit, and until she has eaten no one else may. A young woman once went out to stea.l some; when she touched it, the pond-lily seeds turned to blood. This frog spirit is a big woman who is the principal spirit of every shama.n. When a novice performs, the following four nights are given over to singing and dancing like that of the first day. Experienced sha- mans, however, show their power by filling all five nights with such performances. Each morning there is bathing and feast.ing, and on the fifth morning they return home. One informant said that beside the speaker the dancers were pa.id for their assistance, but this was explicitly denied by others. Pat Kane stated that the shaman's winter performance, which he called ba'pie'nE, lasts only two nights and is solely occupied with singing. I probably misunderstood him. Other shamanistic tricks are of the same order; various objects are swallowed, others are made to appear and vanish, or the stuffed animal skins in the lodge are brought to life. The tricks of plunging the bare arm into hot wa.ter and of walking on fire are unknown. During these performances no one walks about or talks except as he is directed by the shaman. These are all carried out in semi-darkness. For the wood-swallowing trick, five, ten, or twenty small flat wooden splints, called wlokda'lls (see above), shaped like arrowheads, 1930 ] 115 116 University of California Publication in Am. Arch. and Ethn. [Vol. 30 are hung from as many strings. A circle of old men dance about the shaman each dangling one of these. The shaman goes to each in turn, swallowing the splint held above his mouth. Another night he brings the splints back as in the arrowhead trick, using a different song. He calls to them to build up the fire; they see the splints stuck into the mat with which he is wrapped. Long sticks are swallowed with the aid of a sea gull (l!ot'ia'was) spirit. This is one of the regular winter performance tricks, but the details I have are those of a Fourth of July celebration. Poketdjo'tk, a shaman, was hired by some whites to do this. He had a flat stick, six or seven inches long, which was lowered into his gullet by, another man. One old white man seized him by the throat, thinking he would choke. Pokletdjo"tk asked the crowd, "Do you want it to come back?" and at their response it jumped out of his mouth.167 Shamans also have a trick of drinking huge quantities of water, four or five gallons, without stopping. Corresponding to the series of swallowing tricks are several in which various things, fish, seeds, frogs, and blood are made to appear in a basket. For the fish trick the shaman has a man fill a small basket with water. They sing three songs of a big snake, those for male, female, and the young of the snake. The first song is that of the young snake. He then tells someone to look into the water. This man holds a light over it but there is nothing there. He sings another song, that of the male or female indifferently. When the same man looks again the water contains tiny fish (a'mpwan ye'l'as). After the third song, they see five minnows or several varieties of fish. He then tells this man to throw the fish into the river. Throughout the affair the shaman remains on his bed; no one goes near the basket except the man who inspects it each time. It may be that this trick is an inci- dental magical means of increasing the fish supply. A shaman is called on to bring the fish if they are late in running in the early spring, but the informant who told me this had never heard of the trick just described. The pond-lily seed trick is similar. The songs are those of the big frog spirit (djEdje'llstci). The shaman has an old woman half-fill a basket with water and cover it with a smaller one. He lies well back toward the wall during the performance, rising only to smoke. His speaker lights the pipe but is warned not to inhale the smoke. The shaman first sings the frog's child's (daughter's ?) song. One man 167 See also the snake-swallowing episode above. Spier: Klamath Ethnography after another talks to the various spirits that fill the house. The shaman tells the old woman to hold a light over the basket; nothing is seen. Two (?) songs of the frog are next sung and when she looks again the basket is filled with pond-lily seeds. All the old men look at it. When he sings the frog's final song, she finds the seeds gone and a tiny frog in their place. No one touches the basket. This con- junction of seeds and frog is rather interesting because it suggests that there is an associa,tion of the two. It will be recalled that pond- lily seeds brought to the feast cannot be touched until the spirit frog has partaken. It may also be significant that the attendant here is an old woman, for the frog spirit, is conceived as an old woman and a woman with spirit power is necessarily old. The spirit for the blood trick is nameless. It resembles a wild man, but was not described further. The shaman tells them to fetch a basket, part of the paraphernalia of another shaman. He sings this spirit's song. One man told to investigate finds the basket full of blood. When he sings again, the blood disappears. All the while the shaman lies covered up far from the basket. Only the most powerful shamans can perform this trick. A trick of quite another sort is that of making the stuffed skins of animals which adorn the shaman's lodge mo-ve about as in life. A powerful Modoc shaman, Djo'klonos (black sage-brush head), had little cottontail rabbits hanging in his earth-lodge. Sometimes they would run around as though alive; they were spirits. He had a stuffed butterball duck skin hanging on the south side. Sometimes this spirit made blood flow from the duck's bill and at, other times it flew about in the lodge. His song was gle'gotk sne'en dead one flying about nane'mb0'mba'ktisnise'6n butterball duck flying around. Even though I am dead, I am butterball duck flying around. Pat Kane described a performance which he saw when a little boy at Pelican bay. The shaman held a fire dance during the day. Then he told the people to place various stuffed skins on the floor of the lodge: mink, weasel, a shitepoke (tuwa'), and owl. He sang songs for each of the animals. He first sang one of the shitepoke's songs and told them to watch whether it moved. It did not. He sang its second song and the bird walked around pecking at the ground. It had been a dried skin for ten years or more. He sang the mink's song three times. The mink moved about holding a little minnow in its teeth. 1930] 117 118 University of California Pubtications in Am. Arch. and Ethn. [Vol. 30 He sang a song of the weasel spirit. The animal rose and danced, holding up his fore paws. That evening Pat's parents took him 'Into the lodge again. The shaman sang the weasel song again. It danced as before; then it disappeared. No one moved. It was dark in the lodge; the shaman had two men standing on opposite sides with torches. He sang a weasel song, telling them to watch for the weasel's return. Suddenly its tail appeared from the ground, shaking. He sang a second weasel song for some time and its hind quarters appeared. A third song brought the trunk, and with the fourth the weasel lay there motionless. Everyone saw it. The shaman asked, "Did you find that the weasel came back?" "Yes." "Did everyone see it?" "Yes." After this they had spirit singing again, so Pat was sent out by his mother. There were bad spirits coming, which might catch and kill him for no one could cure him.'68 WEATHER CONTROL AND CLAIRVOYANCE The shaman's art is not confined to these winter performances, but is invoked during the year, not only for curing, but whenever his ability to manipulate conditions and to prophesy is needed. The weather is controlled, fish are made to run, lost articles and thefts dis- covered, the fate of a war party is foreseen, and the community pro- tected against malignant spirits. It is easy to see why the shaman transcends the chief in the popular mind, for there is not only the recurring winter ceremony in which the shaman alone has a stellar part, but in every emergency they turn to him, not to the chief. Even in war, the one occasion for a layman to star, the shaman steps in to dim his prominence. There are no special weather shamans and the practice is by both sexes. There is the limitation, however, that weather control is exercised only through certain spirits, which the novice does not have. 168 There are three stone figurines in the Museum of Anthropology of the University of California (catalogue nos. 1-14181, 1-14182, 1-14212) which were said to have been collected on the Klamath reservation and to have been used ceremonially. Two roughly resemble animals, one with a pack, and the third a human bust. Repeated inquiries brought only denials from my informants of any knowledge of their use. Nor have any figurines been found in their country. They may nevertheless have been the private medicine of some shaman. Somewhat similar stone figures from the Wasco are in the collection of the Field Museum of Natural History. In the Field Museum collection is also a pair of eye rings (no. 61533), resem- bling our spectacle frames, which were "worn by men with blackened faces to frighten children at certain ceremonies." I learned nothing of such articles. The rings and the cords to fasten them about the head are made of strands wrapped with bark. 1 ]Spier: Klamath Ethnography The songs of these spirits bring southerly winds and storm or the cold north wind to clear the sky. There is the suggestion here that weather control is a winter practice; hence it falls in the period most suitable for shamanistic performances. Rain is brought with the songs of the flicker, a black bird (nao'lins), the horned owl, red shitepoke, and a big water snail (k!awi's). Wlle'akak, Dwarf Old Woman, lying on Mount Pitt, controls the west wind; she can cause the heaviest down- pour when the snow lies too deep. The prairie chicken makes the loudest thunder. Rattlesnake makes rain and thunder; the thunder is his rattling. To make it snow they use the songs of various white things, such as the pelican, which causes the greatest fall, and the white grizzly and brown bears. Clear weather is brought by a strong north wind; the songs for this are those of a hawk (tcl'ktu), magpie, crow, and Goga'nE, a dwarf, who is most potent. Crow is thought to have cross-eyes; when he rolls them, it stops snowing. The songs of the winds are also used; there are seven of these, one for each of the compass points except northeast. Winds raised by shamans cause sickness and death. There are other spirits which control the weather but these are the principal ones. The shaman lies alone singing in his earth-lodge or sweat-lodge. If he does not succeed in a single night or if he is only partly success- ful, he calls the people to help him with the regular performance (wa'hla), dancing and singing to invoke the aid of more spirits. Nancy's father could make the weather change. He had two spirits, crow to stop the falling snow and the water snail to bring rain when it grew too cold. While other shamans could make the wind blow, the snow drift, and cause it to thunder, Nancy does not know that her father could. His crow song was kak ko6mE ge'o crow mine b0lis galo' holi'ta cross-eyed sky cleared. His second song made it rain immediately, k!awikl!a ya'nu' water snail am I. He sang these in his earth-lodge. An exhibition of weather control of quite different character is described in the following incident. The shaman kiu'ksmag (Shaman Man) went to Ashland in the autumn to trade but he could not return because of deep snow on the Caseade range. He was challenged to make it rain. He melted a strip of lead in a frying pan and had somne one pour it down his gullet. They could hear it hiss. In the middle of the night it rained so hard that the snow melted away. 1930] 119 120 University of California Publicatios in An. Arch. and Ethn. [Vol. 30 The shaman is called on to exercise this art when the weather is unpropitious and the food supply in danger. If the ice stays late in the spring so that they cannot fish, they go to the shaman to get him to make it rain. If it rains but no fish come, they ask him again to bring the fish. The first catch is always divided so that everyone has something to relieve his hunger. Should the fish disappear from the mouth of Williamson river, for example, an old man will ask a shaman to discover the cause. He invites them to assist in the singing-dancing performance in his lodge to call the spirits, which in this case is expedited to two nights instead of the usual five. It appears that a shaman above them on the river, growing jealous of their success in fishing, has sent a. spirit mink under the water to drive the fish away or take them to some other people. The spirit that makes the fish run is a big sluggish snake. The shaman causes two female snakes to lie in the mouth of the river. Then he tells the fishermen to try their luck and to bring him a fish, male (ko'ktsials) or female ('ulu'ksals), from each net. This perform- ance is called kie'rmalo'E, to make fish. Rivalry among the shamans is glaringly apparent. It not only appears in the shamanistic tricks where the results are tangible and visible but extends for obvious reasons to superior claims of weather control and to the detection and conquest of other shamans' sickness- bringing spirits, with the dispatch of malignant spirits in turn. Con- tests in manipulating the weather are not a special form of shamanistic activity among the Klamath, and I doubt that they are among other American tribes where the practice has been described. A Modoc put a spirit on Mount Shasta to make it snow. The spirit caused sickness as well. A Klamath shaman performed the usual dance-song ceremony and discovered that the I?odoe shaman was at the bottom of the mischief. He sent one of his own spirits to catch the spirit on the mountain and enslave it. When it was caught the Modoc died. A very powerful Modoc shaman, Black Sage-brush Head, had a stone pipe. Whenever he wanted to smoke, the pipe lit itself, or rather a spirit did it. He visited Otai'is, a Klamath shaman living at the mouth of Williamson river, and exchanged pipes with him. After his return home, Ctai'is found the pipe missing. Later they had word from the Modoc that the pipe had returned to him.169 There is no question that the rivalry cuts deeper at times. A shaman accused of causing sickness stands in danger of his life, and the matter may turn on the relative personalities of the rivals. It is patent then why foreign shamans are accused, Modoc and Paiute for 169 This appealed to the humor of my informants enormously. Spier: Klamath Ethnography example, rather than local men whose secure position in the com- munity may prove a boomerang to the accuser. No man would be called a witch (loi'tsE) save one who is thought to possess scant ability to cure. Yet at least one instance was cited of the killing of a shaman of standing: he had repeatedly performed the fire-swallowing trick. Only one among the women shamans was credited with power to discover bewitching.170 That they believe that only a shaman of great power can discover this or can send a spirit on a similar errand is a sure indication of the seriousness of the accuser's position. There is a well defined procedure for discovering whether a spirit has been sent into the mountains roundabout by a malevolent shaman. Following the customary five nights' performance, the shaman leaves his lodge in the forenoon. He takes a position in front of a group of women who are to sing his song. He rushes about in an effort to catch the spirit, followed at a distance by five or ten men whom he has chosen. He runs all day; it is very difficult to catch. Wihen he has it, he falls, blood streaming from his mouth. The runners seize him, a man holding each arm while he holds the spirit between his cupped palms. He plunges it into a basket of water, holding it submerged for a long time. Then he exhibits it, a tiny lizard, for example. Dur- ing all this time blood gushes from his mouth; he can neither sing nor speak. Soon after this some powerful foreign shaman will die, for they have caught his spirit. Similarly a powerful Klamath shaman may send his spirit among the Modoc, the Pit river tribes, or into another local Klamath group. Warm Springs, my informant thought, is too far away. This is done during the winter performance. No minor shaman can do this, only those who have powerful spirits, animals that kill: for example, snakes, wolf, coyote, weasel, lizard, golden eagle, big black frogs (ko'wE), smaller frogs (wa`katas), and tiny ones with striped faces (le"eambi6'- tkls). The spirits of the sun, moon, stars, and the winds are of no use in this connection, although I was told that a magically produced wind causes sickness. Spirit poisoning (bewitching) is called tawi'.171 Disease is nepaks, according to Gatschet. 170 The case is described by G-atschet (1: 64-68). 171 There appears to be no word for poison, a noxious agent. Ste'tnus is the term applied to "poisoned" arrowheads, but these are obsidians from a certain magical source, and carry no venom. Spawu'ta is bait, such as fish bait and the bait used in traps, hence by extension it has come to include arsenic, for example, now used in poisoning bait. (Gatschet gives Modoc shpauitush, 1:18.) One cannot say he has been poisoned by the wild parsnip (ska'w6o7ks), only that he has eaten it: ska'w6oqks ot p6'pga. 1930] 121 122 University of California Publications in Am. Arch. and Ethn. [Vol. 30 Clairvoyance, at least the ability to see distant objects and events, lies within the power of any shaman. This is called dode'tiks, to dream. My informant Nancy, who in her youth was on the way to becoming a shaman, had at least this power before she was deprived of her spirits (see above). The spirits talk to the clairvoyant; among her spirits was a bear. Lost articles are found through the dog spirit. The people are lined up in two opposing rows-an excellent psychological expedient for discovering guilt-while the shaman passes along searching for the suspect. He sees the stolen object shine in the thief 's upraised palm. This takes place in the open. My informant had seen this practiced once when some beads were stolen in a gambling group. The sha.man's prevision of what will happen in battle is decidedly more important. This ability is acquired through the eagle and weasel spirits, and perhaps others. During the war dance preparatory to setting out or a.t camp on the march, the shaman dances in front .of the line of warriors, looking "through" them. He predicts who will be killed or wounded, and how the enemy will suffer. He sees them bleeding. One informant stated that those for whom disaster was predicted would not go to war, but others said this is not so and they are probably right. For example, while a mixed group of Klamath and Modoc were encamped on the way to Pit river, a Modoc shaman predicted, "If you are t.o be shot, then your bowstrings will snap," and the strings of two did snap.172 Shamans accompany war parties, not only to fight, but to watch over them and cure the wounded. CURING Curing practice lacks the mise en scene of the winter performance held in the figure-decked shaman's lodge, for it always takes place in the patient's home.. Nevertheless, it must be quite an impressive affair with its accumulating intensity as spirit after spirit is called to the assemblage, as one after the other they are charged with responsibility for the illness. Sickness in the Klamath view is due to spirit activity, the presence in the patient of an intrusive object rather than possession, and a cure (heswo'mbli or tco'ta) is effected by remov- ing and destroying it. There is no suggestion here of soul loss as a cause of illness. 172Gatschet, 1:21. Spier: Klamath Ethnography Curing is possible only through certain spirits, including the fish hawk, falcon, golden eagle, buzzard, magpie, crow, weasel, fisher (?wa'lk6tcka), coyote, wolf, lizard, rattlesnake, a big sluggish snake (wuimEniuks), and a woman spirit. Frog and her daughter, and the winds are necessary to the ceremony, but I do not know that they actually effect the cure. Other spirits were mentioned as non-curing: bear, bald eagle, clouds, thunder, lightning, sun, moon, and stars. Shamans of standing are invited from a distance. When the services of a shaman are needed, it is likely that he is to be found at home. Shamans are expected to maintain a certain amount of reserve, to lie quietly at home, especially in winter. The summoner does not enter the shaman's earth-lodge but calls down from outside, asking if the shaman is there.'73 Some woman of the household answers in the affirmative, never the shaman himself. The messenger announces his errand. If the shaman is willing to cure, the messenger stands on the roof shouting to the spirits while the shaman is getting ready. This is a long harangue calling on the spirits in the distant mountains: ya'moks dake'log's tcomgi'as north earth step forward. He calls to all the directions, every region from which a wind blows: mowa'ksda miulwa'nki south get ready. When the shaman feels that his spirits have come to him, that he is potent to cure, he is ready. Then he decks himself with his medicine outfit (miu'lwas) ,174 flicker feathers or grizzly claws, for example, and sets out. The shaman may refuse to go, "Do not ask me now. I have no spirit; it has gone back to the mountains. " When a man loses a child, his spirit quits him when he grieves. "After a month I will talk to my spirit and get it back. I mourn now; I cry. I cannot sing and dance; my spirit has gone. Return in a month." When the mes- senger returns-presumably the patient holds his peace!-he asks again, "Is the shaman there? Then I come to ask you to cure as you promised." Now if the spirit does not answer the shaman, another (?) man, called saswa'kia, mounts on the lodge before sunrise and again after sunset to call to all the spirits to give their songs to the 173 Also described by Gatschet (1:68-69, 71). 174 If Gatschet 's derivation is correct the word means "what is prepared" (2:221). 1930] 123 124 University of California Publicatiois in Am. Aroh. and Ethn. [Vol. 30 shaman that he may again be powerful. The shaman tells the mes- senger to return that night, perhaps then the spirits will answer. At night the shaman sings several songs. "All my spirits have now returned. Go back; tell all the good singers and my speaker, for now I have strong power to cure." The man returns to say that the shaman is coming. They hang the doctor's fee above the recumbent patient; a rope to represent a horse, or a blanket and beads. The shaman does not name his fee, but he doubtless gauges his efforts by what he sees offered. Before he begins he says that he will cure in two nights, or that they had better fetch another shaman for he is powerless to cure this patient, or that the patient has been poisoned by another shaman and he is not sure of a cure. He brings his wife to sing, for a shaman's wife is generally a good, loud singer. If they want one or two more to help him besides such relatives as know the songs, they pay them blankets, etc. These are individuals who can sing and talk loudly to the spirits. There is one in particular, called hamto'mgis, who prompts the spirits, shouting to them to excite them to work for the shaman. If the shaman 's speaker is not available, he will take anyone who is willing to serve. The patient always lies with his head toward the east, the shaman sits or kneels at his feet, the speaker near his shoulders. The singers sit back of the shaman, the prompter near-by. There is no dancing during a cure. Children are excluded. Apparently the curing nor- mally takes place at night, but it may continue into the following day. If it takes place in a dome-shaped mat-lodge during the day, the smokehole is covered to shut out the light. Baskets of parched and ground pond-lily seed in the form called cnops are provided.'75 The general plan of the performance is first a series of songs to call the spirits to work through the shaman's person; these spirit helpers then tax other spirits with causing the sickness in a second set of songs; finally the shaman tells what he has seen and sucks out the sickness. The prompter first commands the spirits to come to the shaman. The first he calls is Frog, for this is the principal spirit,'76 and Frog makes the shaman call all the other spirits. a'tik'esllke'w6op wasi e'itcomgi's Now you should stand there inside [the house]. '75 Curing is briefly described by Gatschet (1:71-72). 176 I am not altogether satisfied that other Klamath, besides my informants Nancy and Tom, would have said that Frog is the principal spirit. Spier: Klamath Ethnography Then the frog in the person of the shaman sings about the moving of the pond-lily seed upon the surface of the water as it boils.'77 te'ne'ko,gi'so t'en'eko-q moving (?) Frog's daughter is next called, singing: stai'las aino'liota'mpga pond-lily seed mush looking at it. ko'lamwe'as liota'mpga Frog s child looking at it. These two spirits then step out ready; other spirits follow standing there behind them. They are thought of as standing with one foot forward ready for action. The shaman sings all night. From time to time the prompter shouts to the spirits to help him. The shaman halloos to his distant spirits, throwing his head back and muffling his shouts with the back of his clenched hand. Whatever the shaman sings or says, the speaker repeats. The singers (wlno'tnis) follow the spirit songs. There is no. drumming or rattling with these songs. Now a spirit talks through the shaman from far away: tcokna'slsk nasiskwasi gale'lks ala'hi Now tell me tell me inside; dying show me. Various spirits are named to find the offender. tcaka'ni ige'ngi dako'newis i'adako'newis Now you are the one, this earth. you, this earth [that gave the sickness]. As he talks the shaman covers his eyes with his hands, or rather the wrists cover the eyes, the hands are extended prone. This is called smetd6'nwa. Again the spirit challenges: i'ag'engi lakti'w'ltcis You are the one, chief of the winds. And again, tca' i'angengi ke'lalomwe'as Now you are the one Earth's child, by which he means the spirit snake. So he runs through the gamut of suspected spirits. If it is this one, Earth's child, it will have to give up the patient so that he can recover. The shaman's spirit says: ita'lgengl'pka ke'lalomwe'as You are the one who gave the sickness, Earth 's child. 177 This is an esoteric reference; it is not cooked during the ceremony. 1930] 125 126 University of California.Publications in Am. Arch. and Ethn. [Vol. 30 If it is the shaman 's fish hawk or buzzard spirit who discovers this, it threatens the snake. It tells him that he will be picked up from the ground and destroyed, unless he lets the patient go. The snake is frightened and gives him up. The shaman begins to suck when he sees with spirit eyes where the sickness is located. He never asks for symptoms; he can see. As a matter of fact, he probably knows all about the case, for an illness sufficiently serious to call in a shaman is undoubtedly a standard subject of gossip. Before he sucks, he sticks his finger into water and sucks it. Then he bites the body hard and sucks, crying "wann.... while another responds "haa.... " as long as he can hold his breath. He really sinks his teeth in, drawing blood as he sucks. (My inform- ant Nancy's head was once bitten.) This is called ha'nsna. It is the spirit who bites with the shaman's mouth and only certain spirits can do this: fish hawk, buzzard, falcon, magpie, crow, and weasel. Magpie and crow in particular draw bad blood from the patient's body. On the other hand, the shaman does not suck when using a nameless female spirit, but cures by stepping over the patient. He interrupts the sucking from time to time to let the spirit talk. He utters his words rapidly and incoherently, while the speaker repeats and interprets. When the shaman finally draws the sickness out, his head goes back, his eyes close, and blood streams from his mouth. If they want to know what it is, he lets someone feel it on his tongue or spits it into his own palm. The fish hawk spirit is stingy and sometimes refuses to let them see it. This is always something tiny: a bug, the nail of a frog, or the frog itself, a snake, a rattlesnake fang, or a red-shafted flicker feather. When they have felt it on his tongue, the shaman, representing the spirit, eats it.178 I was told that after showing it, the shaman (especially a woman shaman) has someone strike him five times on the back with a basket, but I do not know what the point is. Instead of showing anything the shaman may simply tell what is the trouble. If a very young child is sick, the shaman may tell the parents that it is because they did not take care of themselves but dreamed of something evil. Gatschet records a shaman's statement, for example: "Therefore this patient was hurt because his mother ate too soon after dreaming. Then this patient turned his face to the spirit land."'179 178 Gatschet says the shaman's speaker swallows it (1:68-69). 179 op. Cit., 1:158. Spier: Klamath Ethnography One method for forecasting the recovery of the patient is to tie thin nettle cords around his head, each thigh, and u-pper arm, the free ends being held by as many men. The shaman strikes each cord with a basket. Unless they are all broken the patient is lost. - A patient who has suffered a relapse is cured with the spider song, according to Gatschet's informant. To paraphrase his teext slightly: Then the spider treats him; a piece of deer skin is the spider's curing tool. Then by means of that deer skin he treats him; he cuts out a piece of the deer skin just the size of the spot where the suffering is located. Then the spider song is started while applying that piece of skin. Then he stretches a blanket over it and they strike it with shaman's arrows (haina'sish) so that it enters the body. First a particle enters, then it becomes part of the body and now that piece of skin looks dark. Then after so many days that black thing at last looks like flesh.180 In the interludes of curing the shaman tells everyone to leave. He remains sitting in the lodge with the patient, silent, eyes closed, and smoking. His pipe has an unusually long wooden stem wrapped with a rattlesnake skin; the bowl is the usual cla.y or stone affair. (Possibly some shamans use tubular pipes.) It is lit by the speaker who is warned not to inhale the smoke else a spirit will hurt him. The shaman may also go into a sweat-lodge for communion with his spirits, probably only when curing in the daytime. The shaman and his speaker are alone in the lodge; others wait outside to sing with him. When he comes out, he sucks the patient. How long the shaman continues his exertions is indeterminate. If the patient is very sick, he sings until midnight or morning. Some- times the shaman announces that his spirit has told him that the patient must be "doctored" for two nights; otherwise the sick man's relatives must decide. Payment is made only if the cure proves a success; should he die within a month or two the shaman must return the whole- amount. While I received no hint that a shaman is merce- nary, the size of the proffered fee' must figure in his decision. Fifty years ago the usual payment was twenty or twenty-five dollars in cash or an equiva.lent of clothing, blankets, beads, or horses. Ten or twenty fathoms of beads is the amount. Twenty dollars or one good horse secured his services for one or two nights. -A man who has none of these, gives a canoe, a suit of elk-skin armor, or even food, and if his only riches are marriageable'daughters, he presents one of.these to the shaman. 180 Gatschet (1:72-73) assumes that the buckskin is actually inserted under the patient's skin but I do not understand this to be the case from the text. 1930] 127 128 University of California Publictiotns in Am. Arch. and Ethn. [Vol. 30 Shamans never practice in concert. If one fails another is called. If the first discovers that another shaman is responsible for the illness, the relatives will insist on the accused making a cure, if they do not kill him forthwith, which is much more likely.18' I was not told that a consistently unsuccessful shaman is killed, but it is very probable that he would be accused of witchcraft unless he retired from the field confessing his lack of power. A shaman of du'kwa named LA" 1' wis was thought by the people of iu' lalo-Jn to have bewitched one of their people who died. Later when another was sick they sent for him pretending their need. They slew him with an axe.'82 Coville observes that "the dried root stocks [of the blue flag or iris (ghii'-gum lak'-o, Iris missouriensis Nutt.)] are sometimes used by medicine men as a smoking material, mixed with white camas [wild parsnip?], Zygadenus venenosus, and a little tobacco, to give a person a severe nausea, in order to secure a heavy fee for making him well again. '183 It is distinctly unusual to find that among the Klamath the shaman also figures in curing practice of a non-shamanistic, matter-of-fact variety. Blood is sucked from wounds and ostensibly from fractures; this is one reason for his accompanying a war party. The songs for this are those of buzzard and coyote, who eat carrion, of the wolf, mosquito, fly, horsefly, or any animal or insect that feeds on human blood and flesh. The patient is then usually turned over to a layman for further care. A fracture is treated by applying a cautery all around the spot, smearing the limb with pine pitch and wrapping it with wide strips of buckskin. Wooden splints are not used, nor sweat- ing as a cure. Wounds are dressed with poultices of plants known to some of the older people. One informant said that only wounds in the limbs were dressed, but this is improbable. Endj!a'yukts (the father of my informant Nancy) received an arrow wound in an arm during a feud. Although he was himself a shaman, he did not attempt to treat himself, but called on tkts'as, the shaman who was serving as the spiritual guardian of this war party. The latter pulled out the arrow, applied a poultice made of the leaves of an unidentified plant, and bound the arm. The wounded man eventually died of this wound. Wounds made by a grizzly bear are treated by a shaman who has the songs of a huge spirit grizzly. The shaman prances about in imi- tation of the bear, palms raised in front of his shoulders to represent 181 Gatschet cites a case (1:68-69). 182 The revenge taken by his son K6'kdlnks is described above. 183 Coville, Notes, 93. Spier: Klamath Ethnography the fore paws. He wears neither a bear skin nor claws. This is the information of Pat Kane, but Nancy Phillips when asked about this scouted the idea. The only one wounded by a bear so far as she knew was a shaman, Poketdjo'tk, who cured himself of a black bear bite. It is of course possible that he had the song of the black bear unknown to her. There are no special rattlesnake shamans, but a shamanistic cure for the bite was described. While Jim Jackson was in the Modoc country a rattler struck him in the finger leaving a fang in it. He neither killed the snake nor cut open the wound as is customary. The fang traveled up into his arm, which swelled and began to be spotted like the snake. Poketdj6otk sang one or two songs and sucked out the tooth in a short time. He showed it to everyone after spitting it into his palm. It was minute and fine as a hair; one could not see it well. Next morning Jim, having recovered, paid a horse. This took place by day. This shaman also sucked a fang from the foot of another injured man. So far as Nancy knew, he was the only one who ever cured these bites and what spirit he used she did not know. The rattlesnake is the spirit of many shamans but they do not know how to cure snake bites. There is a quasi matter-of-fact procedure for the bites, however. The flesh is cut away around the wound or gashed to the bone. It is then wrapped with a fresh snake skin. Lo'lo6klls (Pat Kane's father) saved his sister in this fashion, a rattler having struck her instep. She recovered immediately. No tourniquet is used. Men tie the rat- tles of snakes they kill to their hair, women do not. There is a flat-topped mountain called seol'pgis, probably that southwest of Wocus bay on the western side of Klamath lake, on which is a pond where persons ailing from skin trouble and other illnesses are brought. They are laid in the pond so that the water bugs may eat the disease. A shaman's treatment for sore eyes described to Gatschet is put- ting a mixture of charcoal into the eyes, together with a louse "to make it eat up the protruding portion of the sore eye.' p184 It is desirable to record other medical practices here although no shaman is involved. The cautery is in frequent use; all my inform- ants have deep burns on the temples or arms. Two or more sage- brush pellets (siklo'kluts) are lighted and placed on the spot that pains. For rheumatism, for example, they are placed beside the 184 op. Cit., 1:71-72. 1930] 129 130 University of California Publications in Am. Arch. and Ethn. [Vol. 30 patella. One is allowed to burn completely; when the other is half- consumed the sufferer destroys it with a pinch and throws it away, telling it to carry off the pain. The dry bark of the black sage is used for this, chewed and shaped into conical pellets. One informant showed me some prepared from the sage-brush bands used by mourn- ing women and deposited in the trees, which he had gathered. Another informant denied that such bands alone are used. Rheumatism is also treated with skunk grease rubbed into the affected part and the flesh of the animal eaten. Earache, deafness, and hemorrhage of the ear are treated by the application at night of poultices of cooked wild parsnip root. Toothache is kima'dj p'an, ant eats, or literally diut ansma'sE. It is so called because on breaking an extracted tooth they find in it a little worm with a black head, believed to be the larva of the ant, which has been eating the tooth. The sufferer rests his aching face on a stout stick, a paddle for instance, while he sleeps. Some one strikes the paddle a sharp rap; the sleeper is frightened and the aching gone. Pat Kane ascribed this to the Klamath marsh-William- son river division (At'ukckni) alone. The river-mouth people's (du'kwakni) method is to chew a bit of rotten fish, holding it on the aching tooth. The cure of the Pelican bay group (gu'mbotkni) is a cautery over the spot "breaking the tooth." Nancy denied the last two methods. Teeth are never extracted. The Klamath used dog flesh in order to avoid smallpox (gotk!6oks, erupting) on at least one occasion. This was during a winter before the birth of my informants, now in their seventies. Many died because they had so few dogs. Ordinarily dog flesh is avoided, but on this occasion the flesh was eaten and the grease rubbed on the body. Con- sidering the date'85 it may well be that this was a cure learned from French half-breeds on the Columbia. For mumps, the root of a plant called sko'sskoltkgla'ks is rubbed on the swelling. (This waist-high plant grows near the river, is blue topped, with large green leaves.) Mumps has no native name because of its recency. For colds the root of the yayo'klaJ plant, found in the mountains, is chewed and swallowed. A poultice of this is also applied to a swell- ing, then covered with pitch. 185 Powers states that about 1847 smallpox killed 150 Modocs (Tribes of Cali- fornia, 254). This is the widespread epidemie of that date which led to the Cayuse uprising against the Whitman mission in southeastern Washington (Hodge, Handbook of American Indians, 1:224). Taylor mentions its presence on the Humboldt and Bear rivers, Nevada, prior to 1853 (Taylor, Oregon Bound, 147, 155). Spier: Klamath Ethnography Coville noted the use of lodge-pole pine pitch as a remedy for sore eyes, a very small fragment being placed inside the lid, and the leaves of a shrub (ba-bAk'-bak-lha'-nim, Eriogonum stellatumn Benth.) placed on burns to protect the surface from the air. He also observed that the roots of buck-brush (chiik'-lo, Kunzia tridentata), steeped in water, is frequently drunk as a remedy for coughs and other lung and bron- chial troubles. The dry ripe bitter fruits of the same, mashed in cold water, are sometimes drunk as an emetic. Cascara (siir'-um-biik-ish, Rhamnus purshiana DC.) is made into a tea for an emetic; foliage, twigs, bark, and berries are all used. Wild parsnip roots (p6d'-cho, Heracleum lanatum Michx.) are used medicinally. "In medicine a decoction of the herbage [of sage-brush (ghat or bol'-whe, Artemisia tridentata Nutt.)] is used internally to check diarrhea, externally as an eyewash, while the mashed herbage is used as a substitute for lini- ment." The roots of stii'-miik (Wyethia mollis Gray) are mashed and used as a poultice for swellings. The mashed herbage of wiil'-wal (Chondrophora nauseosa (Pursh) Britton), aromatic and resinous, is used to raise blisters. A poultice of the mashed leaves of wai'-mi (Chrysothamnus bloomeri (Gray) Greene and related shrubs), similar in principle to the preceding, is used for the same purpose.'86 THE SONGS With the aid of the large body of texts recorded by Gatschet it is possible to point out some general features of the shamanistic songs. The form of the song is as fixed as its subject. In the first place, the burden of a song is always brief and always invariably consists of words with meaning, not syllables inserted for euphony's sake. If Gatschet's translations are to be trusted, the great preponderance of the songs are in the first person, the spirit speaking. I doubt it. My knowledge of Klamath is very limited but I suspect that such a phrase as Ke-utchish ai nfu shul'sh gi which he renders "I am the gray wolf magic song" is as likely to mean "The wolf is my spirit." A very large number of songs mention the spirit by name and are otherwise not especially esoteric but easily intelligible to one with only a slight knowledge of Klamath beliefs. Certain song patterns recur with some frequency in Gatschet's collection. Four of these have simple forms. 186 Ooville, Notes, 89, 95, 98, 100, 102, 105, 106. 1930] 131 132 University of California Publication-s in Am. Arch. and Ethn. [Vol. 30 1. ai nu shui'sh gi_ I am the (spirit's) magic song [or the is my spirit]. 2. ge-u hu mu'luash The is my medicine object. 3. ge-u nepk is the disease I bring. 4. ge-u ni'paks is my disease. In these the appropriate name of the spirit, bit of paraphernalia, or disease is mentioned. There is a slight variability in the grammatical form, but ignoring this we find that these four patterns occur in a total of 145 songs, fifteen, seven, six, and three times respectively.'87 Another common form has a second phrase practically identical with the first, and, if not a mere repetition, referring to an associated idea or characeteristic. This scheme occurs twenty times in the col- lection. An example is tunepni ge-u welwash gi paltko ge-u welwash gi I have five springs and (all) my springs are dry (p. 157, no. 46). These five patterns make up a third of the whole collection. The number of songs seems well-nigh boundless. The possibilities for inventing new ones are limited only by the number of spirits and all their parts and characteristics. Not only does one get this impres- sion from Gatschet's array, but my own informants gave enough to indicate that it would require great effort to exhaust the supply. At the same time it is clear that once a song had been invented its form is fixed; it is known as so-and-so's song. It is not only that such a shaman has a. song about the weasel, for example, but it is a particular song, and my informants would not hazard it unless they knew it. The subjects of the songs are not only the anthropomorphic beings, spirits of animals, birds, etc., listed above, and others of the same sort, but Gatschet's informant listed others: gust of wind, sleet, mock sun (parhelion), cliff, rocks of various sorts and positions, springs, floating ice, smallpox and various pains, hunger, pond-lily seed and various plants, canoes and other objects, parts of the house, dress, and paint. This is a wide array of subjects, but the same omissions occur that I have noted above: the cats, deer, and most fish, while rodents and song birds hardly appear in Gatschet's collection. Not only do the songs refer directly to the animals but to their parts, 187 It should be noted that these occur randomly throughout the songs which Gatschet lists in the haphazard order he obtained them. Four informants were involved, so we may be sure that this recurrence of the same forms is not due to repetition on the part of a single informant. Spier: Klamath Ethnography "limbs and organs of certain animals which were supposed to exer- cise supernatural powers, and therefore were made the subject of a shul'sh, shuinotkish, or incantation. They are as follows: of the black bear, the head, snout, paws, fur, and heart; of the dog, the head, hair, fur, ears, tail, and paws; of the weasel (tchashgai), the head, eyes, snout, fur, tail, and heart; of the mink, the paws, snout, fur, tail, and heart; of the shne-ish-duzck [black-jack duck], the head and legs; of the salmon, the head and fins; of the fly, the wings (las, black or white) a.nd legs.""" The ears of the young antelope also form the subject of a song. In many cases reference is also made to the female and the offspring of the animal and even of the earth. The references to the spirits mentioned in Gatschet's song collec- tion are not random but center in some particular charact.eristic, habit, or association of the animal. Thus, the swimming of the mink, the underground habits of the weasel, woodchuck, and sna.ke, the soaring of certain birds, the standing shitepoke, the position or move- ment, of the reptiles, are signaled for song. The ducks figure as disease bringers. (These may be curing songs in which the birds are taxed with causing the sickness.) A number of birds control the storm and wind. (These may be songs used in changing the weather.) Some of Gatschet 's songs may have mythical reference. For instance, the dog straying in the north wind, Weasel as a trickster, the skunk dancing in the north wind, and the frog woman. The griz- zly and his springs may refer to a ta.le paralleling that of the brown bear and the spring. There may be others which I do not recognize. One of the songs I recorded refers to the mythical character 'Wawiwo'k !os. It also seems significant that the shaman 's songs referring to himself are not, int.rospective, do not refer to his bodily state or expe- riences, but to his paraphernalia, or objectively to the disease. The norm for Klamath songs seems to be a brief statement, explic- .itly na.ming some spirit or object and its characteristics or associations, rather devoid of esoteric reference, and couched in one of half a dozen simple forms. The following spirit songs are taken from Gatschet's collection. I do not. know which of these are curing songs and which pertain to the performances. I have grouped together those that seem to refer to the same practice or group of spirits, and I have taken consider- able liberty in retranslating those that stood in obvious need thereof. 188 Gatschet, 1:178, 179. 1930] 133 134 University of California PubUeetions in Am. Arch. and Ethn. [Vol. 30 I feel that many of his translations are far from the point,, but I do not know enough to correct them. The numbers in parentheses refer to the page and number in Gatschet's work. Fire dance (wa'hla) songs? In flames I am enveloped (154, 8). I am now wrapped in flames (166, 26). Songs referring to various items of the shaman's outfit (mii'lwas). This scoop is my instrument, that tool used on the lakel89 (167, 34). This is my medicine arrow (164, 3). The pipe is my instrument, that tool used for tobaccol90 (167, 33). I the shaman's pole [wa"llas] am shaking to and fro (170, 66). Planks [covering the lodge?] are rattling (155, 18). This is my song, that of the sweat-lodge stick-holel91 (168, 45). So looks the yellowhammer medicine-decoration; The otter is my equipment; the otter skin neckband is my equipment (167, 30). I the neckband am medicine-equipment; I the neckband am a spirit [or song] (168, 40). Thus I walk when I tie up the hair (165, 9). I will pull a rope from my entrailsl92 (165, 11). Songs referring to anthropomorphic spirits. I the shai'kish [tcakkia'] tramp'93 (169, 52). Tcakia'k is my song [or spirit]; thunder is my song [or spirit] (169, 53). I the ghost lean on a staff (168, 38). Bones only I rattle (157, 44). I the woman am painted black on the body'94 (158, 51). My dew claws are rattling (166, 17). Dew claws are my medicine-outfit'95 (166, 21). In quill-fringed buck-skin dressed (154, 6). I search the ground with my hands, find there the feathers of the yellow- hammer and devour them (154, 10). Quick! make ye eyes for me! (154, 11). In the fog I am straying blind; all over the earth I am wandering'96 (157, 40). 189 Or the lake is my instrument. 190 Or tobacco is my instrument. 191 I do not know what Gatschet means by the "stick-hole." 192 "Feigning to draw a rope or string from their posteriors is a trick some- times resorted to by doctoring practitioners to make a disease disappear " (Gatschet, 1:171). I was not told of this and it sounds more like a shamanistic trick than a curing practice. 193 Gatschet says "shai'kish is another name given to the tch'akiuks" and this is a " song of a gray aquatic fowl, called tch'akiuks. ' The following song, wherein he translates ts'akiag as little boy clearly indicates that the reference here is to Tcakia'k, the little boy spirit which lives in water. 194 Gatschet implies that this refers to a woman dancer, but it is as likely that it is a woman spirit. 195 Gatschet says of these two songs "Conjurer 's rattles are made of deer's claws" and "this song of a female conjurer or 'doetress' . . . ." (1: 172). But deer-claw rattles are used only in a girl's puberty danee; shamans use none at all. I interpret these songs to be those of a woman spirit, like the one which precedes them and perhaps the one to follow. 196 The first two of these three songs are those of "the blind medicine-girl" according to Gatschet and the third that of " the blind girl. " I assume a spirit is referred to. Spier: Klamath Ethnography Songs of oosmic phenomena. I the earth resoundl97 (158, 48). My lake is glittering in azure colors (164, 1). Who, I wonder, is blowing out of my mouth? The disease is emanating from my mouth'98 (153, 3). I am shivering! the wind blows down on me! (156, 27). I am the north wind who stalked (164, 4). The north wind blows around me from the distance (155, 20 and note). Southeast, southeast, southeast, southeasti99 (165, 7). I am indeed the south wind spirit [or song]; I sweep over the earth (167, 29). The storm gust dashes right on me (155, 25). I the storm wind I wind around (168, 42). My disease has indeed arrived; I am the sleet spirit (164, 2). Heavy hailstones I possess200 (156, 26). Fog followed drifting after me (158, 57). Songs referring to animal spirits. I have five springs and my springs are dry201 (157, 46). I am scratching up the ground (157, 42). On the mountain top I am peeping out; of the grizzly bear I am the child (156, 36). I am the wolf song [or spirit] (165, 12). I, the she-wolf, am rolling against (a tree?) (157, 37). Long and slim I am (154, 7). Crazed I am wandering2O2 (155, 22). The red fox's child I follow up203 (156, 30). I am blowing air from my flanks (156, 32). Everything I can devour (158, 53). I the dog am straying; in the north wind I am straying (155, 24). I [the weasel] am squealing, I am squalling (155, 23). From under the ground I [the weasel] am singing (162, 5). Fooling, fooling I [the weasel] run around (158, 52). While walking I [the weasel] shine in multiple colors (169, 55). Mine is this ground, the weasel's; muddy is my ground, the weasel's (169, 56). I [the marten) go up and stick fast to the tree (168, 37). I [the mink] am swimming out while the south wind blows (156, 29). Ripples in the water sheet I [the mink] am spreading far and wide (162, 6). Shaman: The smallpox brought by me, the otter, is upon ye 197 "This is sung when water is poured over the patient." 198 "Song of the wind." 199 Ye'was is southeast and the southeast wind, not east as Gatschet has it. 200 "Probably refers to one of those birds to whom the power is attributed to bring about storms, fog, snow, or any change of the weather. " 201 This and the following are songs of the grizzly, according to Gatschet. 202 These two are songs of the "washpalaks-fox" according to Gatschet. This is a fox, he says. But the translation elsewhere (2:476) "the emaciated prairie-wolf" rather indicates that these refer to the coyote. 203 This song and the two that follow refer to one or both species of fox. Of the first Gatschet says, "This is probably a song of the wind, not of the young silver-fox (as I was told), and I have translated it as such." 1930] 135 136 University of CaUfornia Puboawtions in Am. Arch. and Ethn. [Vol. 30 Chorus: The otter's tread has whirled up the dust (166, 24). I the marmot, I travel around this land2O4 (154, 15). I [the woodchuck] am descending into the ground (154, 5). In the north wind I [the skunk] dance gaily around (155, 16). With shortened steps I [the skunk] am dancing (162, 7). Through what do I [the black mouse] pass with my pawst My paws glide over the hair of the disease (155, 21). Down in the dark ground I [the black ground-mouse] am singing (165, 10). Songs referring to bird spirits. Blown off! I blew off the feather crest (153, 2 and note). As a head only, I roll around (154, 12). I flutter along the ground (while walking) (158, 49). I [bald eagle] am screaming high up in the skies (162, 4). High up in the skies I [bald eagle] circle (165, 5). I am the eagle feather, I am going down, hot (163, 11). High up in the skies I [fish hawk] circle; through the clear skies I carry my prey (169, 50). I [fish hawk] am pinching hard (162, 3). North wind is my [fish hawk's] song [or spirit] (170, 58). This is my head crest; the crest of the wl'tg6tgis hawk (171, 71). I [sparrow hawk] carry my offspring with me (166, 18). I the buzzard circle in the air (170, 62). I [the shkii' bird]205 am a song [or spirit] and circle high above the earth (167, 36). I [the female shka bird] am snapping (or scolding) at the ground (168, 44). This is my sickness, the wa'-u'htuash-duck (170, 67). Belly-ache is the disease I [the butterball duck] am bring on (167, 28). Belly-ache is the disease which I206 carry along with me (166, 27). The nAta-duck is now singing about itself (167, 31). I the fish duck would like to fly over the country (169, 51). I the mallard float on the water's bosom (170, 68). In my lake ripples I [the kilidshiksh-duck] am spreading (165, 15). I [the shitepoke] stand upon the rim of my nest (154, 9). I [the shitepoke] crouch on the water's edge (170, 63). The disease brought on comes from me, the young shitepoke (170, 64). Noisily I [the pelican] am blowing around (166, 19). By me, by the swan, this storm has been produced (166, 20). This tempest is my207 work (170, 69). I am the loon and my waves follow me (168, 46). I the kingfisher am eating up the salmon (169, 54). High-crested I (the bluejay) sing my song (or am a spirit) (170, 61). I the young woodpecker have brought on sickness (168, 39). I, the woodpecker, am holding fast the tree stem (170, 65). I [the woodpecker] am picking hard at the bark of a pine tree (162, 2). The disease brought on by me, the lark, spreads everywhere (168, 43). 204 Gatschet gives wilxatchaka as the long-tailed black marten, but this is the marmot. 205 If its name is in imitation of its cry, this would seem to be a hawk. 206 The ma'maktsu duck: by my information ma'mimkls means waterfowl in general. 207 The wei wash goose, the snow goose; by my information the white brant. Spier: Klamath Ethnography This is the k'als-bird 's fog (166, 22). This is the female kals-bird's fog (166, 23). I [the tchikass-bird] am wafted off from the mountain (157, 43). I am the p6p-tchikas-bird 's song [or spirit] (167, 35). Fearfully the wind blows underneath here2S8 (157, 39). The snow made by me, the tshutsi wash-bird is ready to arrive (170, 60). A flint-headed arrow I [the g-awi-bird] am ready to dispatch (163, 10). White chalk is my [the nfi'sh-tilansn4 ash-bird's] medicine-tool (167, 32). I the little black female bird209 am lost and strayed (163, 16). I [the tsisxixi-bird] am singing about myself (170, 59). I [the wipeli'wash forest-bird] am singing my song [or am a spirit] (171, 70). Of the sxib-bird I am the song [or spirit] (168, 41). Songs referring to frogs, lizards, fish, and bugs. An old frog woman I sit down at the spring2l0 (163, 9). I the slow speckled snake [wi'mzniks] am hanging here (157, 47). This is mine, the black snake's, gait (165, 8). Lo! thus I the lizard stick my head out (155, 19). When I the [male] lizard am walking, my body is resplendent with colors (165, 14). The land on which I, the female lizard, am treading, belongs to the lark (165, 13). Which game did you play with me [the turtle]? (159, 58). Now the wind gust sings about me, the little sucker fish (165, 6). I the tsawas-fish am singing my own song [or am a spirit] (169, 57). Here I [the yellow jacket] am buzzing around (165, 16). I the bug,211 I bite and suck (156, 28). Other songs. I the song [or spirit] am walking here (156, 31). I am the song of evil (ko'idsi) (166, 25). Songs referring to sickness and curing. I the disease am meandering through the skies (162, 1 and note). The pangs of hunger I carry about (169, 48). I feel too bad for dancing (163, 12). I carry pains about (168, 47). What thing do I blow around? The disease I am blowing around in the air (156, 35). Breath I am emitting2l2 (157, 45). What do I remove from my mouth; the disease I extract from my mouth. What is the thing I take out; it is the disease I am taking out (153, 4). 208 "Song of the tchiwititikaga-bird. " 209 Shaixish-bird; elsewhere (169, 52) shaikish is given as a name for Tcakia'k, the little boy spirit. 210 " Songs by a companion of the old frog,'" but the frog spirit is an old woman herself. 211 Shaixish; but elsewhere Gatschet translates this as a little black female bird (163, 16) and as a bird called tch'akiuks (169, 52) which is Teakialk, the little boy spirit (see note above). 212 "Sung by the disease, found to live in water." 1930] 137 138 University of California Publucations in Am. Arch. and Ethn. [Vol. 30 What is coming out of my mouth? Black substance is hanging down from my mouth (158, 56). What do I suck [bite] out? The disease I am sucking out213 (155, 17). I have recorded several songs the use of which I do not know. A shaman's song is about a mythical hero, Wawiwo'k!os, who is rep- resented as singing. ga'no wawi'aw6k I am Wawiw6'k!os. Nancy dreamed a song about the pole and pendant (wa"lasdiu'pka) which stands before a shaman's earth-lodge. wa'"li'sasa nos nos shaman 's pole I I nisowat lis tu'ntani shaman 's pole hanging on. I am the pole and the cord hangs from me. The rainbow is the spirit of Wile'akak, Dwarf Old Woman. In a song she says that she swallows it like a sword. wlle'akakni wlle'akakni stW'nri wi"tc !i6k Dwarf Old Woman something long swallowed rainbow MISCELLANEOUS BELIEFS In addition to omens and miscellaneous beliefs that have no proper place elsewhere in this account I have resumed below all cases of beliefs which have their setting in the social and economic life of these people. I have not repeated here beliefs touched on in other parts of this section on shamanism. The omens of death or disaster are varied. The thrush tells of an approaching enemy: wl'mpis Ege'pkwa, they are coming. The horned owl also warns; mm, mm, mm, kokdElo6n skiu'klu, who sent me? Or he predicts: ma'kIlaks gElek1 gEle1k gEle'k, someone will die. Gatschet recorded similar constructions of the call of a crow, the howling of coyote or dog, a cat's mewing, or a hen's clamor just after sunset.214 One may suggest that these associations are derived from the association of the sunset with the land of the dead which lies in the west. On the other hand the call of the loon at early dawn is a sign that someone will kill another. Yet this may really be a case in point, for then the general association would be with any night cry, since night is the time when the dead arise and roam. 213 Chorus while the shaman bites and sucks the patient. 214 1:88, 133. Spier: Klamath Ethnography "When the sun is environed by lamb-clouds, or a mottled sky, this is figuratively expressed by: 'K'muikamtch [KEMEmu'ku-mps] has taken the beaded garments of Aishish [his son] and dressed himself in them' A peculiar red smoke or haze appearing in the northwest- ern or western sky, shnuiish, announces his a.rrival; he is also recog- nizable by his bulky posteriors, or as the Modoes say of him: 'K'mu- kamtch muinish kutuilish gitko.' By this they evidently refer to the white and heavy mountain-shaped summer clouds. "215 Frogs come out before the rain, hence this is a sign that a storm is due. When the storm rises, old people cry to it, "You go back to the mountains. We do not want you here." The only precaution against lightning is a burden basket or sieve hung on a high pole outside the house.216 When a man. sneezes or his ears ring, someone is talking about him, but not necessarily a woman, as elsewhere. If a door opens, someone is coming from a distance.217 To end the long winter months and hasten the coming of spring, the ring and pin game is played. A proper catch " breaks the month.'" Or little spears of the hard interior of the tule are thrown into the ground. On the contrary, playing cat's cradles in winter makes the months longer. These beliefs are not seriously held; they a-re more of the nature of manners of speech. Gambler's luck comes not only through spirit power but by the possession of parts of dea.d moles, frogs, et.c., or obsidian blades. These must have been found, and they are carried to the play, and are kept, like the gaming sticks, at crematories, or at least away from the house. A series of beliefs surround birth and childhood. A storm always blows at a woman's first parturition. The child will cry if the umbili- cal cord is not tied with the mother's hair. If an animal eats the afterbirth the woman will become barren. She may not eat meat or fish after the birth, must use a head-scratcher and wear a bark belt. The baby is massaged so that it will grow well formed. A girl's cradle is wide so that she will grow broad. The navel cord is saved else the child will die. Sterility is cured by drinking a crushed shell. A turtle head is hung about the neck to ward off rattlesnakes. It wa.s said that this is the general custom but turtles are found only in 215 Gatschet, 1:lxxiv. 216 Also recorded by Barrett, Material Culture, 257. 217 This is said to be a pre-white belief, but the only doors of those days were mats over the earth-lodge hatchways and the flaps of the mat-lodges. 1930] 139 140 University of California Pubtication8s in Am. Arch. and Ethn. [Vol. 30 one part of Klamath territory, on the eastern shore of Klamath lake and southward. A bracelet made of the shed skin of a water snake (wi'slqk) will make a little child swim well. Girl's hair is singed so that it will grow long. Menstrual regula.tions prescribe that at first menstruation girls may not eat meat or fish else they will get sick. They must use a head-scratcher and run toward the dawn. Their clothing and bands must be burned. Older menstruants a-re permitted dried fish and may do without the scratcher. Women must not care for the roots they gather while in this condition, else they will fail to find them in the future. Shamans must avoid menstruants else they lose their power. The sa.me food restrictions are imposed on mourners, who also use the head-scratcher. If they touch hands to face it will wrinkle, to hair it will turn gray. They must spit into a hole. Mourners must use special sweat-lodges. A woman's bands are hung on the trees near-by. Men wear bark belts so that their backs will not grow stooped. The house in which a death takes place must be burned with its contents. Mourners must abandon their first kill or harvest, or their gains at gambling, and give the second away. Similarly novice hunters and root-gatherers give away their first fruits, a gambler his first gains after acquiring power. Mourners must not fish nor cross a river else the fish will disappear. These mourners' observances have largely' a purificatory intent. A storm rages when a shaman dies, when he is cremated, and again when his lodge is burned. Klamath marsh, which is now almost. wholly dry, is known to have been in that condition on earlier occasions, for instance, about 1840 and again about 1860. On the former occasion it was so dry that its southern arm could be crossed on horseback between the settlements Du''ilkiut and I'wal. They did not know what caused it. As a remedy a man would be dispa.tched to one of the spirit pools in the Cascade mounta.ins (as Crater and Diamond lakes) to bring back water in a squirrel or chipmunk skin to pour into the marsh. The first fish rite of these people is held at one place alone, Koma.'- eksi on Sprague river. Why at this point, is anything but obvious. The first sucker caught is burned to ashes, others must be roasted on the spot, otherwise no more will follow. The gall of fish caught with difficulty, as when speared through the ice, must be thrown back into the water else the fish will avoid the fisher. Spier: Klamath Ethnography Grace is said before eating; bits of food are cast toward the earth, the mountains, old house pits, and cremation places so as to feed them first. I am not sure that the accompanying formula is a prayer. If this is neglected, the earth would harm the little children. An earthquake (ke'la t!ill',1gE, earth shakes) is caused by the hero KEmiu'kiumps who moves the earth. It has no special significance. The only quake recalled by my informant occurred during the Ghost Dance of 1870; it was interpreted as a not unfavorable sign. An eclipse is called lok slo'ki, grizzly bear eats. They say that Grizzly Bear eats the sun or moon at this time. People gather to make a noise, crying ko, ko, Frog; they hire the Frog (ko'we') in the moon to use her powers on the bear, because Frog is the moon's wife as the following tale explains. The frog can be seen in the moon. The sun is included in this explanation. Although sun and moon are called by the same word, sa'bas, they are known to be separate entities. Moon had two sisters. When Frog went to visit, she went into Moon 's house. Moon told his sisters, "You had better let her sleep with you." But they said, "No, you had better take her." Frog was angry at this and jump- ing on Moon, clung fast. Moon tried to cut Frog loose with a knife but he could not. So she is stuck fast, part of him, today. He said, "Well, I will keep her as my wife, so that whenever Bear eats me Frog can exercise her shamanistic power to drive him away. " So they are living together today. There are several magical ways of making the wind blow, but they turn on the same notion. There is a rock in the water south of Bare island which is a mythical hero, and another on Eagle Ridge near-by is his dog. If the latter is struck the wind will blow a gale from the south. My father told me [Pat Kane], "Do not hit this else a very strong wind will blow. Do not throw stones in the lake there or play there." I wondered what would happen. I was a big boy. I took a club and struck the rock. Next day a strong wind blew, the strongest wind I have ever seen. It thun- dered, snowed, and hailed for two days. My father asked, "Why did you hit it?" I could not answer. He said, "I told you not to hit it." Similarly, poking in an eddy (sui'ls&uks) near the mouth of William- son river with a pole, pumping it up and down, will cause a strong wind to blow. Both localities are connected by the same tale; the eddy is where the hero disappeared under the water, the rock his resting place. There is another spot where the wind can be made to blow, the rocks at Squaw point. Pounding in the holes on the south side, for instance, will bring a south wind. 1930] .141 142 University of Califoria Publcatiotns in Am. Arch. and Ethn. [Vol. 30 Some beliefs concern the nature of wild animals. The " cat " (da'slats, Bassariscits sp.) kills deer by wrapping his long tail around their necks. The mountain lion is very dangerous; it is afraid only of fire. One cannot escape it by climbing into a small tree, because the lion will dig it up. When bears fight, the female is always victor. She always buries the male. The grizzly bear during his hibernation sleeps a month on each side alternately. "I do not know who watched them," my informant remarked. The grizzly is a notorious thief, breaking into the winter stores buried near the villages. But to call him that is dangerous, for he then pursues you. The chipmunk or ground squirrel (ts!Il'as) hibernates like a bear from November until the spring, they believe. When boys shoot at this animal it changes into a frog. Fig. 9. Pictographs on the eastern side of Klamath lake. Elder is supposed to grow on cremation piles; perhaps it does. The Klamath do not make pictographs. There are however a few in their country, said to have been made by KEmiu'kuimps, the culture hero. They refer to them as shaman's mu'lwas, paraphernalia or, better, objects pertaining to a shaman. They are repainted from time to time by old men, "who work for a shaman," by which my inform- ant may have meant shaman's interpreters. They are all of simple form. Mly informants knew of only two pictographs (su'malo'ta) in the whole Klamath country. Those on a rock slide on the eastern shore of Klamath lake south of Modoc point are shown in figure 9. This is near the former fishing village Iul'a'u (47). There are a dozen circles in white paint, the largest four or five inches in diameter. Spier: Klamath Ethnography Another set of graphs is on the rocks at the southern end of Buck island. A tale concerning one of these figures tells of a man shot in the eye with a straw so that the blood streamed down his face. This suggests the form of the figure, which I did not see. One informant said that the figure of a lizard was also sometimes drawn. Klamath territory abounds in rocks conceived as petrified mythical beings. Where the tales provide no other explanation, it is held that the transformation was effected by Crow laughing at them. Back of the village Du''llk-ut on the southern side of Klamath marsh is a group of small boulders, called ne'knuk., "doctor" stones. Should a little child play among these he will become crazy. A similar group lies north of D-uno'ksi on Pelican bay. Another series of rocks is some twenty miles above Chiloquin on Sprague river. The western- most is the house of KEmui'kuimps; to the north is a sharp mountain, blai'was, the golden eagle; the third is wlkwlks, magpie; the fourth, maiti'ktok, a person, the fifth k!adj!'tewatewe'es, a person; sixth, gao'djis, wolf; and the last, djl'ktu, mouse hawk. A cave at Koma eksi (13) on Sprague river is another house of KEmiu'kuimps. It is a stl'nas, a dome-shaped lodge. As he covered it with second-grade mats (steholas) it now leaks. Inside is a pile of rocks said to be his tunic armor. KEmui'kuimps' net spread out to dry is an outcrop of white rocks on the hillside a few miles north of Klamath Falls. 1930] 143 144 University of California Publication in Am. Arch. and Ethn. [Vol. 30 III. ECONOMIC LIFE AND MATERIAI CULTURE DIVISION OF LABOR At this late date it is difficult to define the sexual division of labor with precision. Some impression derives, however, from random observations let drop during the course of these inquiries. One of my informants summarized the division of domestic activi- ties in this wise. Women fetch wood and gather food. Old men as well as women grind and pound seeds. Men lie about all the time, working but once in a while. They hunt, fish, and in the winter build houses. Undoubtedly he is justified in his view that man's work is spas- modic, woman's work in preparation of food and the care of the home a more continuous round of labor. Such in fact is the picture of native life in perhaps all parts of the continent. Yet a fairer view of their economic participation shows a more equal division of activities. These may be summarized as follows: Men, but women too, are the fishers. Inasmuch as their primary dependence is on fish, the men's activities in this direction constitute well-nigh a year-round pursuit. Men are hunters, but hunting is desultory. Even here women participate, for it is noted at least that they occasionally lie in wait in canoes to kill floundering deer driven into the water. Root and seed-gathering is woman's work; I am not certain that men assist even in harvesting fruit. Seeds, especially pond-lily seeds, form the second staple in Klamath life. Preparation of food is wholly woman 's work, save as they are assisted by old men, perhaps only those who have no women members of their families to undertake the task. Storing food and drying it may be wholly feminine occupations as well. At least where foods are concerned the man 's work is sufficiently time-filling, but the woman's much more so. With manufactures, the division is somewhat more even. House- building, one of the main activities, is a function of both sexes, and is no small part of the domestic routine. Timbers are prepared and erected by men, grass and dirt spread over the structure by women, and the house pit dug by both. Woodworking and the manufacture of weapons are masculine activities. Yet the most laborious wood- Spier: Klamath Ethnography working task, the making of canoes, is sometimes undertaken by women. I do not know who tans skins, but since the manufacture of all garments and moccasins is done by women, this suggests that they prepare the skins as well. Mats and baskets are made by women and in considerable quantity. This is a sedentary pursuit which dovetails nicely into leisure moments of household duties. Mortars, metates, and their adjuncts are fashioned by women, though I suspect that men may lend a hand. Nets are manufactured by men, but by a few women also. Ropes and cords are made by women. All these suggest a somewhat heavier share borne by women. There is a further factor which must be kept in mind. In winter, movement is impeded by intense cold and heavy snowfall, so that all members of the family must stay close at home. Hence the men might be thought to have enforced inactivity thrust upon them during this season. But the facts are probably otherwise. This is a period of great scarcity, even famine, especially at the end of a long winter, and it behooves the men to act and act quickly in search of food. The ardors and perils then forced on them surely make demands upon whatever leisure they may have at this time. FOOD AND ITS AOQUISITION Klamath life is not that of a lake people, but a river and marsh culture, for the simple reason that their primary food is fish. The large lakes of this region are solidly frozen in winter, and fish, which are not abundant there at the best, can hardly be had in winter. It is otherwise with the swift rivers and creeks; these are always open and fish are always present in abundance.218 Fish is the staple; after this ranks water-lily seed (wokas) and to a lesser extent other roots; deer and other game are only of minor importance. It is difficult at this late date to know whether fish or lily-seed was eaten in greater quantity, but there is no doubt that, of necessity, more attention was given the year round to fishing, since the seed is harvested only from the middle of August to the end of September. Winter villages, the permanent settlements, are located where fish may be had at that time, beside running streams and near springs 218 In view of the excellent descriptions of Klamath environment given by Gatschet, Barrett (pp. 240-243), and Coville (Wokas, A Primitive Food, 727- 728) it is unnecessary to add another here. (See also pp. 8-10 of the present work.) 1930] 145 146 University of California Publicationm in Am. Arch. an,d Ethn. [Vol. 30 at the lake's edge, large and warm enough to prevent ice forming over a large area. Winter is severe at this altitude (4000 feet), snowdrifts are deep, and Klamath lake freezes over to a depth of some ten inches in late December. Such sheltered valleys as that of the Williamson river, while but a few miles from the open lake, are distinctly warmer. The annual cycle of activities is dictated largely by the food habits. It is customary to leave the winter villages in April or May, when plant growth is well under way and the flat valleys have dried, not to return until the snow falls again and houses need refurbishing. When fish begin to run in the spring, the Klamath congregate at the fishing places to stay for two months or more. For example, some join the Modoc on lIost river for several weeks to catch and dry fish. The spring and early summer are spent on the marshes and prairies, gathering roots such as camas and ipos. When the camas has been dried and stored, they move to the marshes, especially to Klamath marsh, to harvest pond-lily seeds from mid-August to the end of September. While the women are busy with this harvest, men hunt mule deer, antelope, and other animals whose habitat is in eastern Klamath territory. Late summer and autumn, seeds, berries, and nuts are gathered, the Indians congregatinig where these are plentiful. Many of those at Klamath marsh, for example, move directly to Huckleberry mountain, southwest of Crater lake, to garner these berries. Summer and autumn are the seasons for journeying to visit and trade, for raiding, and especially for laying up a store of fish against the coming winter. It must be understood that the winter locations are not completely abandoned during this season, for many of them are favorably situated for these very activities. But it is customary for such a group as the aukckni division, whose winter residences cluster thickly along the middle Williamson river, to scat- ter and meet again at Klamath marsh for the pond-lily harvest in August and September. (This is the beginning of the Klamath year.) They begin to return to the winter sites in the latter month, to start rebuilding the earth-lodges for the coming winter, a task completed in December, by which time full winter is at hand.219 The move to Klamath marsh of the people at the mouth of William- son river, the du'kwakni, can be made by canoe only as far as ya'ak on the middle river, where they join the aukckni in packing afoot the rest of the distance. When the water is high, canoes can go upstream as far as bEziukse'was at the mouth of Sprague river (see the maps, 219 See also Gatschet, 1: 74-77. Spier: Klamath. Ethnography figs. 1-3). When returning in the fall, they take canoe as far as Kirk, storing the vessels in the lake bottom there until the next season, and packing the rest of the distance to their craft on the middle river, some twenty miles distant. A similar seasonal migration of the Klamath falls group takes them to Barclay spring and Modoc point on the lake front for the spring fishing. Crossing the lake to d6'ktco6qks, they leave canoe and ma.rch by a well-defined trail to Wocus marsh for the lily-seed harvest. At the end of their stay, they sink the canoes used on the marsh in the water at woksa'lks, hide grinding slabs and similar articles in the juniper thickets, and retrace their way across the lake to Algoma. After gathering wild fruits there they return to the falls for the winter.220 When one considers that the countryside teems with a wide variety of animal life in summer, and the seeming abundance of seeds, roots, and fish, it is difficult to comprehend the change that winter brings. Yet outright starvation in winter must have been frequent. The constant refrain of Ogden, who visited the country in the winter of 1826, bears on the scarcity of game, the near approach to starvation, throughout the months his party was there.22' Yet it is not incredible in view of the deep snows that cover the land, the thick ice on lake and marsh, when travel is at a minimum and fish can no longer be taken. Snow falls early, in November, and lies on the ground in exceptional years well into May. The early spring, when winter stores have run low, is the critical period. When the lake remains frozen unduly long and snow still falls, starvation sets in. Then horses are killed, and fina.lly the hides or even dressed elk skins are roasted in their extremity. Nothing will induce them to turn to dog flesh.222 When the snow leaves the pra.iries, they hasten to glean whatever roots have escaped the previous season's harvest. Then too a few trout may be had to eke out their subsistence. Fishing Fish, the primary food stuff, can be taken almost anywhere in Klamath territory, but the supply is more plentiful in some sections than in others. Williamson river is one; fish can be caught there the year round, but in many other streams they run only in the spring. 220Compare Gatschet, 1:xxix. 221 Elliott, The Peter Skene Ogden Journals. 222 This has served as a specific against smallpox, however. 1930] 147 148 University of Califoria Publications in Am. Arch. and Ethn. [Vol. 30 For this reason, the greatest number of settlements cluster on that river. The runs of fish there begin in the early spring, are at their height in March and April, and continue, one variety following another, into the fall. According to Coley Ball seven kinds of fish run in the spring, followed in the fall by the larger varieties. Mid- September marks the end of the sucker run. The time of the salmon run is not clear. Gatschet 's statement is that salmon ascend the Klamath river twice a year, in June and again in the autumn.223 This is in agreement with my information, that the run comes in the middle- finger month, May-June, and that the large fish run in the fall. Pat Kane did not know whether there is more than one variety of salmon, which he called tcia'lEs.224 They ascend all the rivers leading from Klamath lake (save Wood river, according to Ball), going as far ul) Sprague river as Yainax, but are stopped by the falls below the outlet of Klamath marsh. Other fish live in the marsh, however. There are not many special attitudes toward fish nor restrictions on their use. The principal restriction is that one bereaved of a spouse or child may not fish nor even cross a river for fear the fish will flee. This must have been a serious curtailment of the mourner's activities, considering the high infant mortality which must have existed. At the expiration of a year, the mourner must use a special sweat-lodge a second time before he can resume his occupation. Throughout this period he may not eat fish for fear of sickness. Respecting any fish that is caught with difficulty, for instance those speared through the ice, its gall (blis) must be thrown back into the water else others will cease to come. The fish are thought to turn away if this rule is neglected as children turn aside from a morsel of fish they think too small. The practice is called notowa'ble a'mbotot, to throw back into the water. In place of the first salmon ceremony common among the North- west Coast tribes,225 the Klamath have an observance over the first sucker. The locale is wo'kstat on the bank of Sprague river near the settlement koma"eksi, south of Braymill. Above this spot is a cave styled the home of KEmu`'kunmps, the culture hero. The first sucker is 223 Gatschet, 2:436. 224 Information from the lower Klamath river confirms these data. Salmon run through Yurok territory at the mouth of the river in April and in early autumn; the latter being perhaps the greater run, for at this time the Kepel dam rites are held (Kroeber, Handbook, 58, 60, 84). 225 There is no first salmon ceremony, no.prayers for salmon, no salmon heart magic like that of the Yurok, no prohibitions against speaking of the salmon, as among the Wishram, and no special relations or taboos connecting twins with the salmon. Spier: Klamath Ethnography roasted and allowed to burn to ashes. Those that follow must not be taken home but roasted there, else no more will come. If the rite is observed, suckers will be plentiful. Wo'kstat is the only place where the rite is held, the only place "where KEm-u'kiumps made this law." KEmni'kuimps was living at komhV5ski. He made a dome-shaped mat-lodge [the cave] using stWholas mats to cover it [the poorest mats]. That is why it leaks. Right at his home he killed fish. The fish had great difficulty in swim- ming up there, so KEmui'kuimps killed the first he saw. He roasted it right at the river bank. He did not take it; let it burn to ashes. He said, "This is the way the pinsaodiwas [humans, in mythical terminology] will do." After he did this, fish came in great numbers. There is no individual ownership of fishing places, as with dams. Nor, for that matter, are there proprietary rights to hunting terri- tories, berry or seed patches. A chief has no control, no ownership of fishing rights. Even those whose permanent dwellings are near the dams have no particular claim to them. To be sure, one might ask those who live near-by to fish in the spot for him, but solely because they know best how to use the nets there. One reason for this is that the fish dams date from remote an- tiquity. They are, in fact, said to have been built by KEmiu'kumps.226 These are quite common in the rivers wherever a shelf of rock in the stream bed favors their construction. Their purpose is to create an eddy of still water in which the fish can be netted when they take refuge from the swift current. Most of these have been destroyed by the loggers who have cleared the channels of obstructions to float their logs. Gatschet seems to doubt that these are artificial. con- structions, but it is clear that only the foundation is a natural configuration. These are short dams (sa'mkauius) or wings of rocks extending out from one bank. One such in the low falls of middle Williamson river at takalma'keda takes advantage of a bend in the river bank to enclose a pool thirty-five feet across.227 Weirs are not used in connection with dams; in fact they are unknown to the Klamath. Fish are taken mostly with nets (wftco'lhus).228 As among pro- fessional fishermen everywhere, line fishing is hardly sufficiently profit- able to be trifled with. Yet lines are sometimes used, as well as the spear, and there is some sporadic use of traps. There are several kinds of dip net as well as a gill net. 226 See Gatschet, Mythologic Text, 161. 227 Other dams on Lost river are mentioned by Clarke (Rock Piles, 40). 228 The Klamath do not poison fish. Gatschet has a curious reference to kfilxamsh and slW'ds as "fish-killing-articles" put in the nets, but these are respec- tively flax and nettles used for net cordagel (1:150). 1930] 149 150 University of California Publications in Am. Arch. and Ethn. [Vol. 30 In late September there are eddies in Klamath lake where small fish, called k!o'ptu, go round and round. These are caught in dip nets which are slid under the fish. Chub, plentiful in Williamson river in spring, are driven into the rectangular pouch nets. Driven all night long, I was told, sufficient chubs were caught to fill two canoes. A triangular dip net is the principal device for use on the open lake. This was called de'owius by one informant, hlksu'la by another. Barrett has given an excellent deseription of it, which I quote in full.229 One of the most interesting devices connected with fishing is the large triangu- lar dip net used on the open water of the lakes or the larger deep streams. This net is used on a pair of poles held apart by a cross-bar near the vertex of the angle formed by them. The net itself is attached to the poles at the sides and to this cross-bar at the rear, the tips of the poles being connected by a strong string upon which the front of the net is fastened. . . . Usually two men go out to fish, one sitting in the stern of the canoe and paddling, the other in the prow and manipulating the net. While the net is being dipped, the fisherman in the stern paddles quite rapidly along and makes a great noise, swishing the water baek and forth in order to scare the fish near the stern toward the prow. He also has a couple of short sticks with which, just before the net is to be raised, he drums upon the sides of the canoe in order to frighten as many as possible of the fish toward the prow. The fisherman in the prow finally raises the poles and brings up the net, placing the angle at which the poles meet under the prow of the boat, which, as has been stated, has a long, flat, upward slope. The cross-bar of the net poles is slipped over the top of the prow so as to prevent the points of the poles from falling back into the water. In this manner the points of the poles project from the prow like two great horns. Their tips are several feet from the surface of the water, so that the fish are prevented from jumping over the sides of the net and into free water. The man in the prow then hauls in the net, which tapers to a long pyramidal point. As he hauls the net in, it is laid over on the poles to the sides of the prow until finally the point of the net is reached. Here the fish are at last gathered together. It is then a simple matter to take them from the net and throw them either into the bottom of the boat itself or into coarse tule baskets made for the purpose. These baskets of tule are of two forms: a long canoe-shaped basket and the ordinary eircular straight-walled basket ... To manipulate a large net of this type, a canoe not less than sixteen or eighteen feet in length is required. Another net of this same type, but of smaller dimensions, is used in a similar manner on the more shallow streams or in the shallow water of the lake. The poles are of fir, twelve feet long and ten feet apart at their outer ends. The net, which is quite large, is fashioned of five rec- tangular sections, linked together by their long sides, and placed transversely to the main axis of the triangle. The mesh is an inch square. As an index of the value, Coley Ball gave a good milk cow for a single net. In using these nets it is usual to have other canoes approach from the front, splashing to frighten the fish into the net. 229 Material Culture, 249. Spier: Klamath Ethnography Several nets may be carried in as many canoes paddled side by side. The fisherman in the prow, using the leverage of the poles over the gunwale, raises the net somewhat, but a heavy load will cause the prow to settle dangerously. The weight is relieved by those in the other canoes lifting out fish until he can manage the net alone. A favorite place for the use of this net by the southern lake people is above the riffles of Klamath falls. I probably misunderstood an informant who said that this net is used only in winter. A rectangular net (lu'di'has) is much used, especially in the rivers. The net is a pouch nine feet long, its mouth threaded on a vertical pole at each side. The mouth is a rectangular opening, six feet wide and two high; fastened at the bottom to the pole, the sides can slip down and close the opening. Two men wade holding the poles erect, with the net floating behind them. Others drive the fish into the net. When it is filled, they move the poles into a horizontal position, drag- ging the net after them into quiet water; the weight causes the mouth to slide along the poles and close. A circular scoop net (called lu'di'has like the last) is used from the bank. This is a hoop of service berry wood, three feet across, fastened at two points to a pole which crosses its diameter. The mouth of the net is fastened to the hoop; its bottom is closed in a line, not a point; it is three feet deep. The net is held with the hoop vertical while others scare fish into it. When full it is lifted out and dumped on the bank. There is no rope attached to the hoop to steady it such as is necessary in the swift waters of the Columbia. There are several excellent places for the use of this net among the rocks of Klamath falls. In early spring high water flows through a channel into a pothole (gau'wam), some twenty feet in diameter. Fish are carried into it by the current. As they swim back, they are scooped up with the net from the mouth of the hole. In this manner, little white fish (1!6'ptu), suckers (tc!uwam), and trout (me'h!as) are taken. On the eastern side of the falls opposite the pothole the current is very swift. It plunges into a hole in the bed, boiling up with great force at the downstream end. Little white fish are thrown up in the whirlpool. A man stationed on a rock, armed with a circular net, watches. He must jump into the water as it surges, scoop up the fish, and jump back before the current catches him. This is exceedingly dangerous, as he may be carried down into deep boiling water. This is called ge'tba'ksalks. The gill net, called a set net (dekwa, or wlitso'las de'kolga), may not be aboriginal, or rather, my informant may have meant that 1930 ] 151 152 University of California Publications in Am. Arch. and Ethn. [Vol. 30 formerly it was not set on stakes. He referred to mention in a myth of its use as a hand net. At any rate, of recent years it is set on stakes in the lake or rivers. It has a small mesh, is forty or more feet in length, three feet in width, and provided with sticks or dry tule stems in order to float the upper edge, and sinkers in order to weight down the lower. The sinker (ile'mdi'as) is a pebble three to four inches long, grooved about its middle through pecking it with a hard pointed stone. Nets are manufactured by men, although a few women also make them. They have meshes of several sizes. A net gauge (to6ks) of bone or wood is used with a slender shuttle (tco'mdlkius) wound end over end.230 Nettle bark cord is used, from either brown or gray nettles (sle"ets). Coville notes the use of Rocky mountain flax fibers.23' The fish traps used by the Kla'math are strictly speaking not traps at all, since there is no closing device to prevent the escape of the fish. They are very little used. I was told of only one place where the device is used, a little stream above Spring creek on Williamson river, where the water is so clear that the actions of the fish can be watched in detail. There it is used only in winter to take trout. The trap (spuwEm) is a cylindrical basket of willow branches, woven in open twine, six feet long and two in diameter, closed at one end. The opposite end is open and not provided with reversed barbs or other device to prevent the fish leaving the trap. Cords are tied to it so that it may be hauled up when the fish is seen to enter. A trap of this sort figured by Barrett,232 but of only half this length, has one side extending somewhat beyond the mouth. Meacham describes a trap of somewhat different shape: The Klamath mode of taking fish is peculiar to the Indians of this lake country. A canoe-shaped basket is made, with covering of willow work at each end, leaving a space of four feet in the middle top of the basket. This basket is carried out into the tules that adjoin the lakes, and sunk to the depth of two or three feet. The fishermen chew dried fish eggs and spit them in the water over the basket, until it is covered with the eggs, and then retire a short distance, waiting until the whitefish come in large numbers over the basket, when the fishermen cautiously approach the covered ends, and raise it suddenly, until the upper edge is above the water, and thus entrap hundreds of fish, that are about eight inches in length. These are transferred to the hands of the squaws, and by them are strung on ropes or sticks and placed over fires until cured, without salt, after which they are stored for winter use.233 230 Illustrated by Barrett, op. cit., pl. 22. 231 Coville, Notes, 99. 232 Op. Cit., pl. 19. 233 Meacham, Wigwam and Warpath, 283. Spier: Klamath Ethnography Even more simple is the taking of a variety of thick-skinned minnow (t!e'abE). When the marshes begin to freeze, these congre- gate among the tules, when they may be scooped up with a basket. Fish spears are not much used; even salmon are ordinarily taken in nets. For one thing, Williamson river and Klamath lake outside of Pelican bay are much too dark for spearing. Salmon are sometimes speared from the river bank, which is quite low, or from the rocks at Klamath falls. No staging is needed as on the swift Columbia or the lower Klamath. The salmon spear (k'lc) is the usual two-pronged Northwest Coast-Californian variety; a shaft bearing two divergent prongs provided with detachable points. The shaft is fir; the lashed- on prongs of mountain mahogany234 or service berry wood are of equal length. Each prong fits into a socket in the base of a bone point. The three-inch point is barbless. A cord tied through a hole in the middle of each point is fastened some distance back on the shaft.235 Thus the head driven through the fish is detached from the prong, but the salmon is held fast to the shaft by the cords. A special club is not manufactured for killing salmon. Fish are also speared through the ice of the lake with this instru- ment. A wall of snow or tules is built about a little hole cut through the ice and covered with a tule mat to exclude the light. The fisher- man lies at full length on a bed of grass or rushes with his head under this cover, protected from the cold by a blanketing of mats. A second type of spear, called ka"lEks, is used for suckers and other fish whose habit it is to swim near the lake bottom. This is a long pole having a conical bundle of hard wood prongs (eighteen inches long) bound to one end. Their points are spread to a circle of five inches by a wooden hoop thrust down and bound among them. There are said to be twenty such prongs in a bundle; Barrett however notes half a dozen to fifteen ;236 and a spear with iron points which I saw had nine. This spear is used in dark, still waters. Poised above the spot where bubbles rise and jabbed down into the mud at a ven- ture, it may pin a fish among the prongs. Another spear provided with a single barbed point is thrust into the fish to haul it up. The barb is made by binding a splinter of bone at an angle to the point of the foreshaft, which is of mountain mahogany. 234 Coville, Notes, 98. 235 Illustrated by Barrett, op. cit., pl. 22, fig. 4; also p. 251. 236 Op. cit., 251. 1930] 153 154 University of California Publcations in Am. Arch. and Ethn. [Vol. 30 They rarely fish with hook and line. These are either hand lines, no rod being used, or set lines. The hooks (o'tktls) are of several forms. The simplest is the tail bone of a sucker which is naturally provided with spines to serve as barbs (fig. 10). Baited with the liver of sucker or trout, this is used to catch minnows which in turn serve as bait, though they are also eaten. Two larger fish hooks are de- scribed by Barrett.237 " The smaller is a straight piece of bone pointed at both ends and attached to the line by means of sinew and pitch at the middle. ....A small fish or some fish eggs are used as bait, being placed so as to completely cover the bar of bone. The other form of hook is a bone shank with two bone points ... . The two points form angles of twenty-five or thirty degrees with the shank. The three pieces of bone are secured one to another by means of sinew Fig. 10. Tail bone of a sucker used as a minnow hook. and pitch. This hook is used chiefly in fishing for large fish such as salmon and salmon trout, and is baited with minnows .... . The main part of the line is of the gray nettle string, but the brown milk- weed string, which is said to be somewhat stronger and also less visible in the water, is used as a sort of leader." My information is that the line (not or swe'us?) is nettle bark; brown or gray nettles are equally strong. Set lines (k'nau'us) are arranged to take trout at night; they are visited in the morning, a procedure called soke'tcan. Some twenty long poles are set into the mud of the lalke, each bearing a thirty-foot line with a single hook. The hooked fish are clubbed or speared. Only five or ten are caught in a night in this fashion. For want of ability to identify them, no attempt was made to obtain a list of the food fishes. In addition to those mentioned in the preceding paragraphs, minnows (V!o't6ks) are eaten. There are several varieties: M!a'am, blue minnows, Endils, a larger variety, and the thick-skinned minnows mentioned above. Crawfish are taken from the streams. These are simply boiled and are soft enough to eat in 237 Op. cit., 250; pl. 22. 10Spier: Klamath Ethnography this fashion. To judge by the quantities of fresh-water clamshells at the old house sites, clams are also an article of diet. All varieties of fish are*iq4gip the sun, not smoked.238 The fish is slit down the back, entrailsqfjrbackbone removed, the head cut off, and the flanks opened. Of the d,rying of salmon, Coville observes "it was their custom, after a fish was split open, to lay in the body cavity a yarrow stem (liil-wiil'-siim, Achillea millefolium L.) with its leaves and flowers still attached. This treatment, by holding the fish open, hastens the drying process and prevents the decomposition that would be likely to follow if the walls were allowed to collapse. My informant knew of no special significance attached to the use of this particular plant and of no special adaptability it had for this pur- pose, except that it did not give the dried fish such a bad taste as some other plants. p239 A hole is cut through the fish so that they can be strung on poles, which are placed in rows on a high scaffold or set across branches of a tree. Sometimes the head is left on and the poles passed through a hole near the tail. The cache is covered with bits of board and bark. Dry fish are pounded up to make kamalsh.240 Parties returning from fishing excursions to Lost river in Modoc country sometimes cook their catch in a hot spring east of the railway station in the present town of Klamath Falls. Hunting While game is varied and plentiful in the Klamath country, the Klamath are not much given to hunting. As one informant phrased it, "We know very little about hunting deer." Their attitude is betrayed by the exaggerated value put on elk hides, although elk were plentiful. And while fish can be taken by anyone, success in hunting is assured only to one who has spirit power. In a word, the Klamath prefer the easier exploitation of stream, marsh, and prairie to invasion of the forest-clad mountains which invite only the solitary seeker after power. They hunt outside their territory westward' across the Cascades, but not so far as the valleys of the Rogue river system. That region is held by enemies, the Upland Takelma of the Ashland-Table Rock region. Antelope and mountain sheep are hunted in the lava beds south of Modoc territory. Mule deer are pursued in the desert to 238 Meacham refers to drying whitefish over fires (see above). 239 Notes, 105. 240 Gatschet, 1:76. 1930] 155 156 University of California Publwiations in Am. Arch. and Ethn. [Vol. 30 the east, which takes them to the border of hostile Northern Paiute country. In their own territory, the larger game animals are abundant. Elk are everywhere, perhaps more numerous to the north in the Cas- cades. Deer and antelope are plentiful. Deer, as well as ducks and other game, abound in the open country south of Klamath Falls toward the Lost River hills. A small blacktail deer is found in the Cascade mountains. Wolves, coyotes, black and grizzly bears, and other predatory beasts are common. Game birds, especially waterfowl, are abundant about the marshy lakes. The following list shows their discriminations among the game.241 Eaten Deer, generically, ll'lhuinks. Blacktailed deer, mu'smuis. Small red deer, swai'.243 (The meat is very good and the tallow sweet.) Mule deer, bargu'ls (lives in the east near the desert). Elk, wuin (excellent meat). Antelope, ted'u (taken in Klamath marsh). Mountain sheep, wi'6s. Black bear, wlth'm. Grizzly bear, lok (paws and flesh eaten). Porcupine, tce'lls (perhaps eaten only recently) .250 Beaver, pum. Ground hog, mu'i. Mink, kli'p'a. Otter, k!ult (eaten by some). Raccoon, wa'tck!nz. Not eaten Wolf, kaio'tcis.242 Coyote, was. Red fox, wan (found west of Klamath lake). Black fox, he'hai.244 Marmot (?fisher), wa'lk6teka.245 Bob-cat (or lynx?), slowa' (or slet).246 Ba8sarisew (?), da'slats.247. Cougar, loi'yaka!w&s.248 Marten, ph'zp.249 Weasel, tc!6egai. Skunk, tea'sis (eaten only as a cure for rheumatism). Badger, kols. Mountain beaver, pum (or gitekani pum, small beaver, or puma'k, beaver cub).251 Chipmunk, wa'sla. Larger chipmunk, ts!I'l'as. 241 All identifications were made by my informants, except Bassariscus. 242 The name s8ggests the English (< Aztec) word coyote, but repeated inquiry failed to reveal any other name for the animal. 243 Having a black (or white!) tail; long straight horns with a slight hook at the tip, pointed forward; eats roots and leaves, not grass. Nancy Phillips says this has horns like other deer, and is identical with the black-tailed deer, save that the tail is slightly longer. 244 My informant suggested that this is a Chinook jaxgon word. 245 Black; larger than the last; can climb trees; whistles. 246 Spotted; short tail; pendant cheeks; eighteen inches to two feet tall; seen but not often killed. 247 Striped; long striped tail. This is not often killed, because they fear it. 248 Larger than the last; screams like a woman and whistles; seen on the high- est mountains, but not killed for fear of it. 249 Red; like a mink but lives in the mountains. 250 The Modoc eat porcupine and chipmunk, according to my Klamath infor- mation. 251 Not killed. Spier: Klamath Ethnography Eaten Not eaten Jack rabbit, kai. Wood rat, kotca'. Cottontail rabbit, k !oik!ois (or Mouse, muik!o'km. tc!w6'gann). Small mouse, blai'nzhiitcnA's (or Ground squirrel, mEsa's (occasionally n6nw' 6'kfis).258 eaten). Shrew, cu'lsi.255 Tree squirrel, giu'was.252 Small gull, l!A'6tc.256 Larger tree squirrel, kW'nk6n.254 Waterfowl, generically, mh'm-6kl&. Pelican, ku'mal. Goose, los. Swan, kos. Brant, la"'l6k.257 China brant, l6kl6k.257, 258 White brant, wai'was. Sandhill crane, k!llti's.259 Blue crane, so.o'ks. Loon, d6plal. Small loon, ko'kiuks (not eaten by children for fear of sickness). ,ngu'k.257, 260 Gull, ku'l'a.261 Smnallest gull, lo6tlia'was.257 262 Mudhen, toho's. Teal, tc!hks.257 Mallard duck, w'Aks.257 Pintail (or Sprague) duck, golks.263 Canvas-back duck, go'la. Butterball duck, b6mba'ktis. Wood duck, djikdjlks. Small duck, koka'swa.264 Blackjack duck, sne'is. Pishduck, tc!W'knnos.265 Blackbird, djuiks (occasionally eaten). Redwing blackbird, k6k6'klaufis.257 252 Calls giu, giu. 253 Very small; white; long tail; lives in the ground. 254A foot long, with long bushy tail; blue; lives in the mountains. 255 Three inches long; long pointed black nose "like a grizzly bear 's"; very scarce. 256 Has swallow tail. 257 Named in imitation of its call. 258 Like a goose. 259 Scarce. 260 Black; long neck sticks out of water. 261 Cf. English gull. 262 Smaller than k.a'5tc; white; forked tail. 263 Ljong white neck. 264 A little duck with big head and neck. 265 Red head. 1930] 157 158 University of California Publications in Am. Arch. and Ethn. [Vol. 30 The Klamath are evidently catholic in their tastes, excluding among larger animals only the out-and-out carnivores. Most land animals are taken with the bow. Snares are not used, even for birds, and the deadfall trap finds only occasional use. Snares (knes) for deer, an important hunting device of northwestern Cali- fornia,266 are said to be used by the Molala. Several hunting methods which might have been expected here are unknown: driving deer or antelope into an enclosure (chute-and-pound) or over a cliff, and the use of nets for taking rabbits.267 The first is a Great Basin practice; the latter used both there and in California. Nor are pits dug on deer trails. Deer driving of an alternative nature is practiced in at least one locality. At Pelican bay tongues of land extend into the marsh from the toe of the timbered mountains. The deer are driven onto these, where they can be readily shot, or pursued into the water where women wait in canoes to slay them. The humming sound of a buzzer made of a deer astragalus bone is a "sign of this." The Klamath do not stalk the deer in disguise, but my informant ascribed this to the people at Yainax, i.e., Modoc or Paiute. They stuff the head skin of a deer to wear over their own heads, but add no horns; the body is covered by the attached deer hide.268 Dogs are used in hunting, but not to the same extent as with the Molala, who employ them to track deer or drive game toward the hunter. The dogs are described as small, slim, with short legs, the hind legs crooked. They were short-haired, had erect ears, short faces, with sometimes black marks on the upper eyelids, called k!au'li. A rock south of Barclay spring is famous as a place to train hunting dogs. The puppy is thrown into a near-by spring while the master hides behind the rock. A dog's name is given for individual markings or happenings, as wano'mgoks, marten bites him. One received the name of a man, loi'yas, because the dog was brown( ?). Dogs are called by name, or tc!isk, tc!isk. They are chased away, Cccc ge'ka, get out. To encourage them to fight, one says, kuic, kic, kic. 266 Kroeber, Handbook, 817. Barrett (Material Culture, 246) states that Klamath and Modoc took deer in snares, but I am certain he is mistaken. 267 Despite a myth referring to this, my informant did not know whether the Northern Paiute use it. 268 There is, however, an antelope headdress in the Field Museum of Natural History (no. 61678) which is ascribed to the Klamath. It is formed of the whole head skin of the animal, but is hardly more than a cap. Pendant from the rear is a bunch of hawk(t) feathers. Spier: Klamath Ethnography A deadfall trap (sna'ndotsloks) is set in the Cascades or along their foot to take such animals as marten, fisher, fox, and coyote. This is constructed of logs, with the entrance blocked in such a fashion that the animal entering must spring the trap. I was unable to get a description of the trigger. Waterfowl are shot, caught in nets, or hunted with a jack-light. The arrows used for these birds are of cane with foreshafts of wood sufficiently light to float. Near the tip, to serve as a barb, the fore- shaft has a wad or ring of pitch, or of pitch and sinew, or carved from the shaft, as described by Barrett.269 The long nets (nut) for waterfowl are set up in the marshy margins of the lakes where ducks commonly congregate. These are fifty or sixty feet long by three wide. They are set under water, reaching quite to the bottom and hardly appearing above the surface, for the purpose is to catch the birds as they dive. The net is watched day anid night. People may even sleep and cook in a canoe to be near it. While my information is that flying birds are not caught in these, Barrett is probably correct that they have such use. The net is stretched upright above the surface and dropped on the birds as they fly or swim into it. The birds are also caught with the aid of a jack-light and net as they rest on the water at night. This method (sliu'mta) is used with success where they flock in warm, unfrozen spots such as the mouth of Williamson river and Pelican bay. A fire is built in the prow of a canoe; a triangular scoop net, commonly used for fishing, is held upright behind it.270 Pitchy wood is used to provide a bright light; the canoe is protected by a thick layer of dirt. The ducks flying toward the light are caught in the net or fall into the canoe where they are clubbed. Four or five men armed with clubs and bows some- times go together in a large canoe. Diving birds are sometimes taken with the many-pointed fish spear, according to Barrett. But a curious variant of it has been described by Meacham. The young fowls are captured in nets. The arrangement is quite cunning, and, although primitive in construction, evinces some inventive genius. A circular net is made three feet in diameter, and to the outer edge are attached eight or ten small rods of half-inch diameter, and about fifteen inches in length; three inches 269 Material Culture, 247. My informant denied Barrett's statement that the primary purpose of the ring is to cause the arrow to skip over the surface and strike in the flock, though this undoubtedly happens. 270 Barrett (loc. cit.) has the fire behind the net. 1930] 159 160 University of California Publicatioms in Am. Arch. and Ethn. [Vol. 30 from the lower end, which is sharpened to a point, the net is attached. The upper end of the rods are beveled on one side, and inserted into a rude socket, in the end of a shaft ten feet long. Armed with this trap, the hunter crawls on the ground until he is within safe distance of the mother-bird and her little flock, when, suddenly springing up, the old birds, geese or ducks, as thg case may be, fly away, while the little ones flee toward the water. The Indian launches the shaft with the net attached in such a way that the net spreads to its utmost size, the sharpened points of the rods pierce the ground, and, the upper end having left the socket on the shaft, stand in a circular row, holding the net and contents to the ground.271 A few data were gathered on the manner of preparing game for food. The flesh of such animals as porcupine, beaver, badger, and raccoon is boiled. The viscera are not eaten. The skin is peeled from a beaver's tail after it is roasted. Mink and otter are also pit- roasted, but not boiled. The paws of the grizzly are baked in ashes and then skinned. Salt was not known until the advent of the whites. Insects are probably no regular article of diet, but G-atschet records that women gather moth chrysalids in the month spe'lwis (forefinger, i.e., late August and September). The ground is scraped up with a paddle to gather the chrysalids (puf'lXuantch); they are pit-roasted between layers of grass, with a covering of bits of bark and earth.272 Seed and Root Gathering In summer the Klamath congregate in several districts to gather pond-lily seeds, roots, and wild fruits: the great swamps and prairies of Klamath marsh, about northern Klamath lake (Agency lake) and Pelican bay, and Wocus marsh west of the lake. The prairie south- west of Fort Klamath is a meeting point for those from middle Wil- liamson river, from its mouth, and from Pelican bay. The huckle- berry patch some fifteen miles southwest of Crater lake (called iwumk-a'ni, huckleberry place; iwum, huckleberry) is a favorite camp- ing place. Here Molala from west of the Cascade range join them. Williamson river people (aukckni) on Klamath marsh cross directly westward to this spot afoQt. Root and seed gathering is woman 's work. They tell of old women digging roots alone who were killed by grizzlies. The principal seed supply of the Klamath is from the pods of the water lily (wo'kas), which grows extensively in the marshes and 271 Meacham, Wigwam and Warpath, 282. 272 1:76, 148. Spier: Klamath Ethnography along the lake borders. It was estimated that Klamath marsh alone contained 10,000 acres of the plant in 1902. This seems an excep- tionally abundant food supply, but the difficulty is that a day's harvest is relatively small. A woman may pick from four to six bushels of pods in a day, but the seed content is only one-fourth as great. Gatschet's informant implies that the season's activities netted an individual seven to ten sacks of ground seed, each of fifty pounds content.273 The season is from the middle of August to the end of September. So complete an account of the wokas industry has been given by Coville274 that only a summary is needed here. The preference is for the fully ripened, open seed pods, but as these constitute only a fraction of the day's harvest, the partly ripened pods are subjected to a special process to force maturity, with a variety of further treatments depending on the stage they reach. The various grades of seeds, in the order of preference and maturity, are known, according to Coville, as spokwas, stontablaks, lowak, nokapk, and chiniakum. This is truly the order of maturity, but the native preference as foods, as stated by Nancy Phillips, differs: spo'kwas, the fully matured seed, loll'nc, cooked and ground seed, snops, ground and parched, and si'wullnc, the cleaned seed of rotted pods. It must be borne in mind that the hard-shelled kernels are found in a mucilaginous mass within the pod. The fully ripened pods, called spo'kwas, are fermented in a pit for one to five weeks, the mucilaginous matter washed out, and the seeds drained. When the process must be hurried, the pods, mature or otherwise, are steamed in a fire. These steamed pods are called a'wal; separated into two grades, the better is known as no'g'tk, the inferior, tc!ini'Ekum. The seeds of these pods are freed of their sticky coating -by mixing them with absorbent pulverized rotten wood, charcoal, or ashes. Like spo'kwas, these two grades of seed are prepared by parching in a tray basket with live coals, grinding them lightly to remove the shells, which are then winnowed out, leaving the free, fresh kernels called loll'nc. This is either dried and stored, or parched to make snops. The latter is eaten dry, or mixed with water, whole kernels or ground, to make a mush called shlotish. The greater part of the harvest consists of 273 1:74, 76. 274 Wokas, a Primitive Food of the Klamath Indians. I have enquired con- cerning his description, but have nothing substantial to add. 1930] 161 162 University of California Publication in Ain. Arch. and Ethn. [Vol. 30 immature pods, which are spread in piles to dry. The fully dried pods are removed from the surface of the pile, pounded to break them open, and the dry seeds, known as lowa'k, collected and stored. The bits of pod with seeds adhering are also stored against famine. The pods in the interior of the pile rot; these are also pounded, dried, and winnowed. These seeds are sno'ntablo'ks. The dry lowa'k and sno'ntablo6ks seeds must be treated differently from the fresh moist seeds described above. These are parched, the shells cracked by pounding, and the kernels boiled to a mush called siwulinc or ta'lEwa (the latter meaning stone-boiled). The shells may be winnowed before boiling, or they may be skimmed off. What remains in the vessel is called sti'llns. During the time when a pause is made in the gathering-process the conjurer carefully watches the ripening of the pods not yet harvested and arranges public dances. When the sun has done its work, he solemnly announces it to the women, and they go to work agaan in their canoes.275 A liquor derived from prepared wokas seed of any sort is called kau'wutcE. This is imbibed cold. Boiling (ta'lEwa) is brought about by dropping hot stones into baskets of water. These are picked up with two green sticks, or the bare hands if not too warm. Split or bent sticks are not used for tongs. Exceptional years have seen Klamath marsh so dry as to seriously affect this supply of food. It was dry when my elderly informant was quite small (about 1860), and even earlier (1840), in his grand- mother's day, it had shrunk so that the pods could be picked from the ground. To restore the water little sacks of chipmunk and squirrel skin were carried to spirit places in the mountains, such as Diamond and Crater lakes, whence water was brought to pour into the marsh. The Klamath make use of a wide variety of other seeds and roots. The seeds of goosefoot and rye grass, which mature in late summer, are beaten out with sticks over mats. The seeds of the first are ground on the metate and mixed with cold water to make a mush which is eaten with the fingers. Rye-grass seeds are parched in a tray basket, wherein they are tumbled ab-out with live coals, then pounded in a mortar set on a large mat. Dry pinches of this are eaten. Coville lists many other seed plants.276 The seeds of the abundant Balsamorrhiza are gathered, roasted, and ground as a farinaceous 275 Ga.tschet, 1:76. I was not told of this and am doubtful about it. 276 Coville, Notes on the Plants Used by the Klanath Indians. See also Gat- schet, 1:146-150. 1]Spier: Klamath Ethnography food. Gatschet adds that these are beaten from the plants with pad- dles early in autumn and pounded in a mortar. Those of Polygonum are gathered, the hulls rubbed off by hand, and the seeds parched and often ground. Gatschet says the seeds are beaten off into baskets at the end of the summer. This meal is either eaten dry or mixed with water and boiled, a process which turns the material red. The seeds of a marsh grass, Panicularia, are a favorite food. Tarweed seeds are often used and those of Mentzelia. The minute seeds of a low grass, Agrostis, and those of slew grass are eaten. The seeds of the former at least are beaten from the plants by women in the familiar manner. Other seeds are used less frequently: those of the sugar pine, the common cane, tule (mii'-em lii'-wiils), amaranth, and dock. There are three species of dock used for food, each with a native name. The seeds of Sisymbrium are parched, ground on a metate, and eaten as mush. This plant is more plentiful in the Modoc country. The seed- like carpels of Triglochin are parched and eaten, sometimes in a decoction. The leaves and stems of Rumex are eaten fresh, and the seeds are eaten when ripe. The youmg shoots of the wild parsnip are used and the aromatic leaves of Sium are eaten as a relish.277 To these I can add that the leaves of koko'dom, a bush used for making flutes (not elder), are eaten green. Roots constitute an important element of diet. Those most com- monly gathered are those of marsh plants or of plants that grow in the open prairies: the roots of the water lily, tule, cat-tail, ipos, camas, arrowhead, etc. Wild potatoes or arrowhead, a small sweet tuber, are plentiful about the mouth of Williamson river, for example, and are procured by the canoe load. Coville observes that the plant (tcuwa') is so common on Chewaucan marsh (between Summer and Abert lakes) as to give its name to the locality. Fremont saw there that "large patches of ground had been torn up by the squaws in digging for roots, as if a farmer had been preparing the land for grain." Gat- schet 's information was that the plant was scarce in the neighborhood of the agency.278 They are roasted in the ashes and eaten. When the intention is to store them, they are pit-roasted and set away in large tule sacks. For winter use these roasted roots are boiled long in water. 277 The last is confirmed by my own information. 278 Coville, Notes, 90; Gatschet, 1:149. 1930] 163 164 University of Califoria Publiatione in Am. Arch. and Ethn. [Vol. 30 Camas is abundant; it is sometimes eaten raw at the time it is gathered, but is usually baked and stored.279 Coville states that it is gathered about April 1. It will keep indefinitely. Ipos is one of the earliest spring plants gathered for food, accord- ing to Coville, "the roots being dug about the 1st of May, at which time the contents are soft and milky. The root is commonly dried and eaten raw." A closely related tuber (the yampa of the Utes) is an equally common article of food.280 A tuber called yantch (Calo- chortus) is eaten, which resembles camas and is described by Gat- schet as cylindrical and in size as large as the thumb.28' While the onion-like roots of Peucedanum. are a staple of the Modoc, they are occasionally used by the Klamath. The roots are strung on cords or kept in sacks. The mealy interior is soft enough to be eaten when dry or they may be mashed and.boiled. Young tules are pulled up from canoes, the lower white portion being retained. These are palatable only when fresh.282 The tuberous base, called khii-iils', of a sedge is eaten. The short rootstocks of the cat-tail are eaten late in the season, when full of stored food, under the name ktoks. Similarly the young rootstocks of the common bur- reed late in summer develop at their ends tubers which have a sweet- ish taste and are used for food. The bulbous expansion at the base of the stem, similar in its qualities, is likewise eaten, and is called klop'-ii. Undoubtedly other roots are used, the only' poisonous one accord- ing to the Indian's account being the wild parsnip. Coville lists at least another, "poison camas," scii'-6 or scou (Zygadenus venosus Wats) 283 The nutlets of k!ol (Valeriana) are pit-roasted. Coville gives the following account of the procedure: "In cooking, a hole perhaps a meter or more (about 3 or 4 feet) in diameter and half as deep is dug in the ground and lined with stones. A fire is then built in the hole, and after burning for a sufficient time is cleared out. Fresh grass is next laid over the hot stones, then k-o1', then more grass, and the whole covered with earth. The mass is then allowed to cook and steam the rest of the day and over night, when the pit is opened and the cooked roots are ready for eating." Gatschet adds that they are 279 Gatschet, op. cit., 76. 280 Coville, Notes, 101. 281 Gatschet, 1:146; Coville, Notes, 93. 282 Gatschet, 1:148. 283 Notes, 93. Spier: Klamath Ethnography then pounded.284 "The odor of the root is very disagreeable to white people, to such a degree, indeed, that the use of the plant about the agency was formerly forbidden. "285 Because of its odor, it is placed in storage pits to keep animals away. Acorns are absent from the Klamath diet. The oak grows only about Klamath falls but the nuts are not used. The nuts of the so-called chinquapin bush, growing in the higher mountains, are sometimes gathered. The hazelnut, which grows along the Klamath river north to Pelican bay, is occasionally used for food. The principal fruit foods are huckleberries, service berries, choke- cherries, currants, and wild plums. Huckleberries are gathered in enormous quantities, especially at Huckleberry mountain, southwest of Crater lake, where the Klamath congregate in the third week in August. These are sometimes boiled so that the liquor may be drunk. Besides the common huckleberry, a low variety of the mountain slopes is eaten fresh or dry. Service berries grow everywhere. The fruit is gathered in large quantities in August, spread out on mats to dry, and kept for winter use, a practice followed with all other berries where practicable. The fruit of the chokecherry is gathered in large quantities in September and dried. Yellow-flowered currant and com- mon currant berries are frequently gathered. The gooseberry and the raspberry, which grow at Afodoc point, are eaten fresh or dried. Blackberries, the green berries of manzanita, and elder-berries are articles of food. The berries of a bush honeysuckle or "cranberry" are always eaten fresh, but never to any extent. No attempt is made to dry strawberries, which are not abundant. Wild plums are abund- ant about the eastern shore of Klamath lake from Modoc point south- ward and in the vicinity of Summer lake. These are eaten fresh or dried, sometimes after being pitted. Wild rose haws, which ripen in September, are dried and pounded to grind the seeds. Bearded moss (k!al), which hangs from the boughs, is pit-roasted; the product is bo'ka. Coville states that "black moss," a lichen, prob- ably the same, is a famine food.286 Pine gum is also eaten. In the spring, usually in May, a broad strip of the bark of a young yellow pine is removed, and the sweet mucilaginous layer of newly formed tissue (stop'-iilch) between the bark and the sap wood is scraped off and eaten. This is a frequent 284 1:147. 285 Coville, Notes, 105. 286 Notes, 87. 1930] 165 166 University of California Pubications in Am. Arch. and Ethn. [Vol. 30 practice. Gatschet notes the use of bone implements to peal off the bark.287 In April the cambium layer of the lodgepole pine is similarly scraped off and eaten. A few additional food plants are mentioned by Gatschet, which seem not to be included in Coville 's extensive list. Klapa is evidently a cylindric root; klu, a rounded root growing in the Modoc country. A large bulb of a low plant (tsuni'ka) is eaten. Watksam is pre- sumably a root. Pawash is a plant said to resemble k!ol (Valericna), hence it may be that the nutlets are used. It grows chiefly at Yainax. They are baked, pounded, and dried. Kela'tch is a small blue berry, gathered after the wokas season; it is dried and boiled. Another is washlalam iwam, squirrel's huckleberries. The white buds of ktfu'ks, growing on a straight stalk, are eaten, as well as those of wi'wi, a similar plant. No description is given of the edible portions of other plants: kfu'ktiu, kiis, ta'ksish, tok, tsi'kal (Phragmites ?), and tsuak.258 The native and scientific names of these plants follow, mainly as given in Coville's paper. These are listed in the order of mention above. Seeds Goosefoot, kotca'nlks, Chenopodium fremonti Wats. Rye-grass, glfbihi, Elymus oonden.atus Presl. Lbia, BaEsamorrhiza sagittata (Pursh) Nutt. and B. deltoidea Nutt. Kiip'-i-inks, Polygonum dougLwsii Greene. Marsh grass, kim'-ch6-da"-lis, Panioukaria fluitans (L.) Kuntze. Tarweed, go'-e-whai, Madia glomerate Hook. Lo'-lis, Mentzelia albicaulis Dougl. A low grass, no'-tak, Agrostis perennans (Walt.) Tuckerm. Slew grass, chap'-to, Beckmannia erucaeformis (L.) Host. Sugar pine, kta'-lo, Pin.ua lambertiana Dougl. Common cane, Phragmites phragmites (LA) Karst. Tule, ma'-i, Scirpus lacustris ocoidentals Wats. A small amaranth, Amarantu,s blitoides Wats. Dock, go'-klaks, Bumex salioifolius Weinm. Western tansy mustard, te!i'pas, Siqymbrium incirum Engelm. A rush-like perennial, gil-len'-ai, Triglooh-in maritima L. Herbage Ken-ii'-wat, Bumex geyeri (Meisn.) Trelease. Wild parsnip, pod'-cho, Heracleum loanatum Michx. Wa'-kam, Samum cicutaefolium Gmel. Mint, mach-as'-sam, Mentha oanadensis L. 287 Gatschet, 1:148. See also Barrett, Material Culture, pl. 22, fig. 5. 288 Gatschet, 1:146-150, 75. Spier: Klamath Ethnography Roots Wild potatoes or arrowhead, tcuwa', Sagittara arifolia Nutt. Camas, p6'ks, Quamasia quamash (Pursh) Coville. Ipos, Caruim oregonum Wats. "Yampa," ndiilk, Carum gairdneri (Hook. and Arn.) Gray. Yantch, Calochortus ma?rocarpus Dougl (?). An onion-like root, Peucedanumn canbyi. A tall sedge, wich'-pli, Carex sp. Cat-tail, p'pfis, Typha latifolia L. Bur-reed, pod'chak, Sparganium euryearpum Engelm. Nuts and Fruits K!ol, Valeriana edulus Nutt. Chinquapin bush, Castanopsis chrysophylla minor (Hook.) A.DM. Hazelnut, Corylsw california (DO.) Rose. Huckleberry, iurum, Vacoinium mem.branaceum Dougl. Low huckleberry, Vacoinium sooparium Leiberg. Service berry, chiik'-iim (tclai'o-ks), Amelarwhier alnifo.lia Pursh. Ohokecherry, de-wich'-kiish, Prunus detmissa (Nutt.) Walp. Yellow-flowered currant, ch6ms'-kaim, Ribes aureum Pursh. Common currant, chmar-lak, Bibes ceroum Dougl. Gooseberry, lh6-l6'-6l6--sam, Ribes oxyacanthoides saxosum (Hook.) Coville. Raspberry, miis'-li, Rubus leucodermis Dougl. Blackberry, tO-tiink'-skm, Bubus vitifolius Cham. and Schlect. Manzanita, shle-shliLp-shiim, Arctostaphylos patula Greene. Elder, s15'-li-sam, Sambucus glauca Nutt. Bush honeysuckle or "cranberry," o'-tam, Lonicera conjugialis Kell. Strawberry, jb'-i-jiks, Fragaria tvirginiana Duschene. Wild plums, to-mo-lo, Prunus subcordata Benth. Wild rose, tewi'di, Rosa fendleri Orepin. Lichens Black moss, Alectoria fremontii Tuckerm. Gu Young yellow pine, kiip'-ka, Pinus ponderosa Dougl. Lodgepole pine, wii'-ko, Pinus murrayana Balf. Food is stored in the ground. The only exception is fish placed to dry on scaffolds. The food is put into long tule sacks which are placed at the outer margin of the earth-lodge between the timbers, where they are buried in the earth covering of the roof. Large communal storage pits (pEno'`ks) are dug near the houses. Some of the Wil- liamson river sites are surrounded by wide expanses of such pits. These are about fifteen feet in diameter, about three feet deep. A group of neighbors combine to dig and use such a pit. Tule mats are heaped over the sacks before the dirt covering. Caves are occa- 1930] 167 168 University of California Publioations in Am. Arch. and Ethn. [Vol. 30 sionally used, the entrance blocked with stones. In all these cases, an evil-smelling plant, called k.!ol (Valeriana edulis Nutt.) is buried with food in order to keep animals from digging it up. Grizzly bears are particularly destructive of caches, but even they avoid this plant. Gatschet's information was that they would even flee from one who carried the plant about him.289 Human thieves are also known to ransack these hoards. A family commonly buries its extra stores in this fashion when temporarily away from home. Such storage is called p'nana. The first fruits of a young hunter or root gatherer are looked upon as something altogether special. They must be abandoned if future success is to be assured. While I received conflicting versions of the young deerslayer's procedure, they agree that he cannot eat his kill. One informant had it that the first deer is abandoned by the hunter and never eaten. If he has companions, he does not even show it to them and must present the second one slain to one of them. Another informant agreed that he cannot partake of the flesh of the first deer, but his parents make a feast of it, giving away articles in honor of the occasion. Similarly a young woman destroys by burning the first roots she gathers alone, and does not exhibit her success to anyone. The second basketful is distributed among her companions, and only after this may she keep what she gathers. Analogously a man leaves the first fish or game he takes after the loss of a close relative, and a gambler gives away his first gains after reacquiring power following the death of a relative. The practice of abandoning such fruits is called sapu'tsa. Unlike the comparable case of the novice deer hunter, no celebration is held for a little girl who gathers roots under her elders' tutelage. Meals are not eaten at regular intervals, I was told, but that they eat all day long "like horses." Of this I am doubtful: what may have been implied was that food was available at any time during the day. I was also told that a meal was eaten in the morning and again in the evening. Travelers at least made no regular stops for meals and if necessary cooked in their canoes over little fires built on protecting layers of dirt. A sort of grace is said before eating. Morsels of food are cast all about, to the land, the mountains, the old house pits, cremation piles, stones that are transformed people, etc., so as to feed them before one feeds himself. They say bui'pwi'amEni genige'lE genidelo'tclipkE, to 289 Gatschet 1:147. 1930]Spier: Klamath Ethnography throw to the ground and the stones. I am not sure that this formula is a prayer. If they neglect this grace, the ground (ge'1E) will harm the little children.290 It is customary to take a mouthful of water to wash out the mouth before eating. CANOES The Klamath canoe is a clumsy-looking, trough-like affair, with blunt ends, heavy and long. But I have seen an old woman guide a fully laden twenty-foot canoe upstream through a log jam in a manner to awaken considerable respect for their riding qualities. They look unwieldy, but with the load and paddler set well aft so that the prow rides out of the water they can be spun about on the ster as a pivot and driven through the still waters of much of Klamath territory with considerable ease.29' The shape of the canoe (wu-ntc) is exactly that of the peeled log, hollowed to a thin shell, and the ends beveled upward from the water- line quite sharply. They vary much in length but are remarkably uniform in width, eighteen to twenty-four inches, which is the diameter'of the majority of the firs used for canoes in this country. Most of the canoes are made relatively short, ten to twenty feet, to wind about in the narrow crooked channels of the marshes. One canoe measured eighteen feet long, twenty inches in greatest width, and sixteen or eighteen inches from gunwale to keel. Another old example was fifteen feet long, seventeen inches wide, eight inches deep at the waist rising to twelve inches at the ends, which were somewhat pointed, with the under surface of the nose and stern rising at an angle of about thirty-five degrees. Larger canoes for transport on the open lake measure up to thirty-five feet in length, three feet wide, and thigh-deep inside, I was told. Fir is the most suitable wood, pine makes a canoe that is too heavy to handle, and the cedars in this region are too small for canoes.292 Besides this, cedar readliy cracks when it is dry,'especially in such thin shells as these. One of my informants even said that cedar canoes are too light to handle properly! Fir also checks and cracks but not 290 For further notes on utensils and the etiquette of meals, see below. 291 For illustrations, see Barrett, Material Culturej pl. 10; Coville, Wokas, pls. 3, 4, 5. 292 My informants probably meant by "fir" the yellow pine (Pinus ponderosa Dougl.) and by "pine" the sugar pine (Pinus lambertiana Dougl.). Coville does not list fir in this region (cf. Notes, especially p. 89). 1930] 169 170 University of California Publications in Am. Arch. and Ethn. [Vol. 30 badly because of its somewhat twisted grain. The worst cracking occurs of course in the cross-grain at the ends of the canoe. Minor cracks are filled with pitch alone. A good fir canoe will last from fifteen to twenty years or more. A canoe that is properly dried before it is put into water will remain white indefinitely, but a green canoe soon blackens. Fire is the active agent in hollowing the dugout. It takes about a week to complete a fair-sized canoe by this method. The fire is, started in a long notch cut along the upper face of the log. Stones are placed at each end to confine the flame. The ends are cut off on a, flat bevrel at an upward angle of thirty to forty degrees with its axis. Barrett states that the angle at the prow is considerably more acute than that at the stern.293 The gunwales are quite parallel for nearly the full length of the canoe and less than its full width, so that the sides curve inward at the top, giving a crescentic cross-section to the craft. The fire is allowed to cut the gunwales slightly wider and lower at the center so that the canoe curves upward a little at each end. And depending on how far the gunwales are cut. down into the log at the ends, these are either squared across or, as is more usual, somewhat rounded. The shell is cut quite thin, a thickness of about half an inch at the gunwale and not much more in the bottom and ends. The exterior is hardly trimmed after the bark is removed, so that the whole canoe is quite cylindrical. Nevertheless the canoe has not much roll and rarely capsizes, especially if well laden, for then the flat under- ends steady it. The finish is quite rough, especially when compared with canoes of the lower Klamath river and those of the Northwest coast. There are no thwarts; they are unnecessary to the paddler who sits or kneels in the bottom. But crosspieces (hehO'vkwis) are used to spread the gunwales while the canoe is still green or to keep them from spreading after the wood has dried.294 The paddle (kEts'ek) to be light is made of cedar, four and a half to five feet long. Half its length is blade, widest near the tip which is pointed, tapering to the long thin handle which is without any sort of grip at the upper end.295 The paddler, who sits near the stern in order to lift the prow, paddles for a time on one. side, then 293 Material Culture, 247. 294Gatschet has an obscure reference to "ti'lihash as a 'swimming-sucker'- wood on canoe they stick up" (1:150), by which he may have meant a piece of wood carved to represent a swimming fish attached to the canoe. 295 Kroeber illustrates an excellent exmple (Handbook, pl. 67h). 1]Spier: Klamath Ethnography on the other. The blade is turned parallel to the canoe at the end of each stroke. The size and shape of the blade are well adapted to his position so low in the water, for the stroke is effective and splashless. At intervals he dips his hands to prevent blistering them on the paddle handle. A punt pole (lEko'k) is also used, especially in the shallow marsh, This is eight to ten feet long, with the lower end split and held open by an inserted wooden block. The punt pole is a young lodgepole pine (wa'-ko), stripped of bark, according to Coville.296 Sails are not used on canoes. Anyone may make a canoe. Even a widow may fashion it for herself. One such had seen her father make them many times when she was a girl. Naturally skill varies very much. Rafts or balsas, formed of large bundles of tules lashed together, were used, according to Barrett,297 particularly by war parties. These were paddled with the hands. Alternative methods for winter transportation may be noted here. It is customary to pack on the back through the snow, not to fashion a sled. But something of the sort is used over ice, a bundle of willows, called spi'klos. VARIOUS MANUFACTURES AND TOOLS Woodworking is undeveloped among the Klamath, whereas it might have been expected that the expertness of the Yurok or the Wishram would have made itself felt here. The only noteworthy uses of wood are for dug-out canoes, rough planks for the earth-lodge roof, and carved figures set up by shamans. The planks are only rough segments of hollow trees; the canoes are without fineness of line and crude in craftsmanship. Northwest Coast social organization rather than craftsmanship has left its mark on the Klamath and their neighbors. Fire is more rapid and effective in trimming large masses of wood than the stone and bone tools at their command. Hence it is gen- erally used for felling trees, cutting logs to the desired length, and shaping canoes, but it is customarily used for smaller objects as well, such as pointing a digging-stick. When they wish to fell a tree, a deep notch is cut into it at the base with elk-horn pick and maul. A 296 Notes, 89. 297 Material Culture, 256. 1930] 171 172 University of California Publicatio' in Am. Arch. and Ethn. [Vol. 30 fire built in the notch quickly eats into the dry heart wood but will not spread in the green wood. Sometimes a lazy man will attempt to cut off the dry top of a standing tree by cutting all around with the horn pick. This is attended with considerable danger, since the top may drop straight down on the-workman perched in the branches. Stone axes and adzes are unknown, although Barrett speaks of adzing the canoe.298 In their place, every man had several elk-horn picks (to'ke) and hardwood mauls. The picks are the unworked tines of elk antlers, not made wedge shape. These are cut off by laying a fire at the appropriate spot on the antler. Yew wedges are not known. The maul is only a knotty club of oak or mountain mahogany; a knot with the limb attached. Barrett describes a conical stone maul, but I was not told of this.299 The fire drill is the familiar stick rotated between the palms on a wooden hearth, so that the flowing wood dust falls on the tinder. The wood selected for the drill (sll'kwls) is preferably the dry tip (ha'bE;) of the bull pine; alternatively a willow or sage-brush branch is used. It is two feet long, a half inch diameter. Coville was told that "for a twirling stick a dry, dead twig of yellow pine, about 6 mm. (one- fourth inch) in diameter, seasoned and somewhat softened by the weather, is often employed; but the best stick is made of sage-brush, Artemisia tridentata.'"00 The hearth (k'ts!lk) is of indifferent wood, commonly a paddle blade,'but the best wood30' for the purpose is cedar (wo'lwuims). The size and shape of the hearth is not standard- ized, and curiously it lacks pits for the drill to rotate in302 and notches to carry the wood dust to the tinder. The tinder is black sage-brush bark. Although one of my informants denied the use of the com- pound drill it was described by another and a specimen is figured by Hough.303 Such a drill has a willow-root point, three inches long, spliced to a shaft of any sort. While this suggests the possible use of a cane arrow and foreshaft as a makeshift drill, I was told this was not used. Men carry the drill in the quiver, women in the cradle. They coax the drill while twirling it, "I want to kill all kinds of animals, to put them in the fire to cook, when I get flame. "304 298 P. 247. 299 P. 252. 300 Notes, 88, 105. 301 Confirmed by Coville, who states that the bark is not used for this purpose. 302 Barrett says otherwise (op. cit., 257). 303 Fire Making Apparatus, 536, 538. 304 For the method of laying a fire, see below. 1930] Spier: Klamath Ethnography A bit of rotten wood is ignited to carry fire about the village but for an overnight journey a woman carries a spark in a long fuse. This is of black sage-brush bark, twisted into a rope thirty inches long and bound about at intervals.305 Knives are fashioned from stone blades or shells. These river shells (endj!'etc) are not ground to an edge but used as they are. Forms ranging in size from arrowheads to large lance blades are chipped from obsidian, flint, and similar stones. So far as I am aware these are never ground into shape. Those intended for knives are sometimes wrapped around the butt to provide a handle.306 Stone knives are intended for butchering rather than woodworking. Blades never figured in the display of wealth, as among the Yurok. Barrett figures a bone knife used for separating the inner and outer bark of pine.307 Sewing is done with awl and sinew. The awl (so'ktE) is a frag- ment of the fore-leg tibia of a deer. Although the upper end of a tibia is nicely rounded to fit the hand, and is accordingly used else- where, the Klamath make no use of it. Instead a splinter sharpened at each end is inserted in a wooden handle. Old men carry these at the chest tied on a buckskin thong. The sinew used is that which lies along a deer's backbone. (Sinew is called Emboi'itc; this particular sinew bi'lhop.) Skin garments for either sex are sewn by women. Women also make moccasins whether of skin or tule. The smoking process is used when tanning. The skin is stretched over a tipi-like frame of little willows set over a hole as deep as one's forearm, The fire is made of rotten wood. Coville states that "the bark [of the white fir (bai, Abies concolor (Gord.) Lindl.)] is some- times used to dye and tan buckskin, giving it somewhat the appear- ance of ordinary tan-colored leather. "308 Neither bags of skin (k!o'wanE) nor parfleches are made by the Klamath. These are brought from the Dalles, I was told, and prob- ably from Warm Springs, for ceptok, parfleche, is clearly a loan word from Sahaptin (cf. Yakima cata'kai). Mats are of four kinds, each designated by a name. Sewn mats (st!ops) are used solely as the outer covering of mat-lodges; when 805 Illustrated by Barrett, pl. 22, fig. 1. 306 Pitch was not used to fasten these. One type of knife, at least, found as far apart as Mexico and the Arctic, and consisting of stone fragments set into the sides of the handle, is unknown here. 307 P1. 22, fig. 5; cf. Gatschet, 1:148. 308 Notes, 88. 1930] 173 174 University of Califomia Publicatiot in Am. Arch. and Ethn. [Vol. 30 the tules are placed vertically the mat sheds water admirably. These are made of tules (Scirpus lacustris occidentalis Wats.) placed side by side, through which cords of swamp grass are sewn at intervals. These are sometimes as much as four and a half feet wide, i.e., as long as the tules, and fifteen or more feet in length. The needle (st!opo'tks) is of hard wood or willow, seven or eight inches long, and provided with an eye. No wooden creaser is used as on the Northwest Coast, but the fingers are run along each side of the needle while it is still in place, so as to leave a crease when it is withdrawn. Two other varieties of mats are twined. That in most common use is called slai'is, the best-made mat short of those with decorative effects. These serve primarily for bedding, and are made solely of flattened or split tules. The tule warps are twined at intervals of about three inches with fine nettle or flax309 cords. A border or sup- plementary warp is placed on two edges, lying parallel to the tule warps. This is either a two-ply twisted rope of tules or a three-ply braid. It is caught in the twining cords, which form a continuous strand, i.e., on reaching the edge of the mat the cords are carried out- side the mat to the next point where twining is to commence and there twined across the warps in the opposite direction. Warps are commonly twisted together where they are free of the mat. A coarse mat (ste'holos) is used to shingle the earth-lodge under its covering of dirt. This is usually made of bundles of swamp grass (bateni, Carex sp.) twined at intervals of three to four inches with the same material, or with Rocky Mountain flax (Linum lewisii), according to Coville.3'10 Tules, cat-tails, rushes(tsin-a'-o, Juncus balti- cus Willd.) 311 etc., are used for warp and weft. In order to hold such swamp grass warps in place before they are twined, one end of each is caught in a twined border weft, the short end of the warp is turned back into the twining and trimmed off. This twining weft is then carried continuously through the fabric as described above. Tule mats made in this fashion can of course be twined without the border weft. These coarse mats also have the extra marginal warps on two sides.312 A decorative mat called soliumtai'is, used for bedding, is infre- quently made. This is a checkerboard weave; the warps are 309 Coville, Notes, 99. 310 Coville, Notes, 92. 311 Coville, toc. cit. 312 Mats illustrated by Barrett (Material Culture) are of the st!6ps variety (pl. 24), sla-i'is (pl. 23), and ste'hol6s (pl. 25). 1930]Spier: Klamath Ethnography alternately tule and a small cat-tail (watc6'k), the wefts are cords made of the rolled outer layer of the tule (called k!o'sgi). These wefts are placed in pairs in simple tabby weave at intervals of one to three inches; they are not twined. The wefts are so thin and so tightly drawn that they barely show in the finished mat, but they produce a checkerboard effect (solumtai') which gives this variety of mat its name. Possibly there is a fourth variety called ak!a'l1ks, an open twined mat, but of this I am not certain. All these mats are provided with a border of twisted strands of the same material caught in the weft cords. Blankets woven of strips of rabbit skin are made by Modoc, not by Klamath. Coville states that "a long time ago the bark [of the cottonwood, ko-osh', Populus balsamifera L.], when peeled and split, was used in the manufacture of an Indian cloth.'"313 What this fabric can be I do not know, unless bags are meant. Ropes are made of three braided strands. The material is grass or the outer husk of the tule for such rough work as carrying wood. Strips of hide are sometimes used, but little elk hide is available for pack ropes. Finer cords, such as those for fish lines and nets, are made of nettle (slets) bark fibers, of which there are three or four varieties (or perhaps phases). Other plant fibers, such as the Rocky Mountain flax (kol'-ii-kams, Linu.m lewisii Pursh. ) 314 make stronger cords. Brown and gray nettles are most commonly used because of their strength.315 Swamp-grass cords are used for sewing mats. Bar- rett notes that string made of brown milkweed is but little used and that all string is formed of two strands twisted together.316 The latter point is confirmed by my own observations. Cords are made by plac- ing two strands, or a doubled strand, side by. side on the bare thigh; rolling them separately under the palm by a movement up the thigh, and by a quick reverse movement down the thigh, without raising the palm, rolling one around the other. Cord-making is a woman's art. A rope is called pek!os and cord gEniu'ks; braiding is k!Fb6'k!a and rolling p'tc!'l-. 313 Notes, 94. 314 Coville, Notes, 99. 315 Goville (Notes, 95) lists only one species, "the native species," Urtica breweri Wats. 316 op. cit., 250. Coville indicates that milkweed (not, Apocynum cannzbinum II.) does not occur in Klamath territory, but on Lost river where it is used by the Modoc (Notes, 103). 1930] 175 176 University of California Publications in Am. Arch. and Ethn. [Vol. 30 The root digging stick (a'mda) is made of a very hard wood, mountain mahogany (Cercocarpus ledifolius)317 or service berry, three feet long and an inch and a half in diameter. One end is burned and rubbed on a stone to a point. Since modern examples made of iron are slightly bent at the lower end, it may be presumed that wooden examples were also. A cross-handle of the same wood, nine inches long, is set on the upper end, wedged and tied in place, the joint covered with pitch. Wooden spoons (mitc!o') shaped like little paddles are used. Spoons are also cut from the breast bone of the swan, which is con- a b c Fig. 11. Pestles and mortars. a, c, d, Klamath; b, Upland Takelma. Length of a, 9 inches. veniently shaped for the purpose; these are called sh'o-kobh'.318 Others are river shells, which are not provided with handles. But mushes are frequently scooped up by the handful.319 Buckets or receptacles of the bark of lodgepole pine (Pinus mur- rayana Balf.) are mentioned by Coville. "Sections of bark from trunks of the proper size are often used to make buckets for gathering berries, particularly huckleberries. The cylinder of bark is sewed together on the slitted side and at one end, the bottom therefore being wedge-shaped. Huckleberries when placed in such receptacles and 317 Coville, Notes, 93, 98. 818 Coville, Wokas, 733. 319 In this connection we mav note that turtle shells, which would have provided excellent utensils, are not used. Nor are they used for rattles. Turtles (nng6'k) are found only in the southern part of Klamath territory, on the western shore of Klamath lake from Modoc point to the falls. Spier: Klamath Ethnography properly covered with large leaves retain their freshness for a long time. ' 320 Both metates and mortars are used for grinding seeds, mashing roots, and similar operations. The metate or grinding slab is how- ever the favored utensil. The metate is a flat slab containing a little depression. The muller in use with it is unique in that it is commonly provided with two horns or knobs to serve as handholds and which project away from the user. This form is however not invariable, for any shape is used so long as it provides a flat grinding face and a narrower upper part that can be grasped with both hands. For light work, a hemispherical or conical stone is used. Barrett de- scribes the grinding process: "the muller is held so that the horns or ears point from the operator and the grinding is done on the stroke of the muller from the operator, the stroke toward the operator being very light indeed."'32' A flat tray basket is tucked under the far end of the metate to catch the meal. The mortar (ga'm6ks) is a lava block bearing a depression which deepens with use and whose form is thus largely determined by the manner in which the pestle is used. The exterior is frequently nicely finished. The stone pestle (sk!a) is elongate, and on the whole nicely reduced to form by pecking. Several mortars and pestles seen and sketched are shown in figure 11. The more nicely finished mortar (fig. llc) has seen use over three generations. Figure llb is a lava pestle obtained from Upland Takelma territory; its form is not that of Klamath pestles.322 BASKETRY While no special study of Klamath basketry was undertaken. Barrett's notes on the subject can be amplified.323 As with the Cali- fornian -tribes, basketry provided most of the domestic utensils. Klamath forms are few, the technique simple, and decoration used but moderately. The vast majority of the baskets are of twine weave: there is but one type, the water basket, made by coiling. The types recognized and named by the Klamath are the follow- ing:324 hats (k!Emui', women's; k!ane'o, men's), burden baskets (ya"gi 320 Notes. 321 Material Culture, 252, pls. 12, 21. 322I venture that the stone maul figured by Barrett (op. cit., pl. 21, fig. 8) is such a Takelma pestle set to a new use, for I was told of wooden, not stone, mauls. 323 Material Culture, 253-257. 324 See also Coville, Wokas, 738. 1930] 177 178 University of California Publication in Am. Arch. and Ethn. [Vol. 30 and sklotgiuls), boiling baskets (lka'la), small twined bowls (m6o'lsE), parching or gambling trays (p'a`la), triangular fans used with the parching trays (wlliu'ks), seed beaters (we'ks, small; ti'a, larger, also used as a sifter), triangular plates (sa'plius), storage bags (wa"klo and t'a'yis), storage-baskets (tliuks), boat,shaped baskets used in har- vesting pond-lily seed,325 cradles for newborn infants and sifting baskets (both called kwatls), water baskets (name unknown), and flat bowls (k!we'lu) traded from the Pit river peoples but not made by Klamath. The basket hopper attached to a mortar is used by Modoc, not by Klamath. Despite the large number of named types, the actual number of technically different forms is small. Close twine or open twine, flexible or rigid, essentially the same technical features are used throughout. The methods of beginning the fabric, the kind of twin- ing, addition of warps, and finishing the rim are practically identical in all. Minor variations give Klamath basketry a specious appearance of technical diversity. There are really only four or five types of basket from a structural standpoint. Twined bowls, hats, gambling or sifting trays, and circular baskets, all in close twine weave; burden baskets of all forms and storage baskets, in open twine, are simply variations of one struc- tural procedure. The second type, comprising seed beaters, ladles, and winnowing baskets, are open twine construction based on rigid warps bent into a loop. Triangular fans and trays, two-handled bas- kets, canoe-shaped receptacles, fish traps, and cradles, all of open twine, constitute a class differing but little from the last. The fourth group is the twined flat bag. Each of these groups merges structurally into the others, largely as variations in form demand. But the fifth type, the water basket, is built on an entirely different principle, namely, coiling. Practically all baskets are made by plain twining, that is, of two weft elements crossing one warp at a time. Some use is made of diagonal twine, which differs from plain twine only in that the wefts pass over two, three, or four warps at a time. Three-strand twining is also used to add rigidity, as around the base or rim of a basket, and for decorative effect. As most twined wares are made with warps as well as wefts somewhat flexible, the fabric as a whole is not rigid. The direction of weaving, at least for twined wares, is from left to right on the forward edge of the basket, held mouth up; that is, counterclockwise from the viewpoint of one looking into the basket. 325 Mentioned by Barrett, 255. 1Spier: Klamath Ethnography Bowls, gambling or sifting trays, hats, and all other solid twined wares are manufactured in the same way. The small twined bowls mo'ksE) have warps of two-ply tule cords (Scirpus lacustrtis occi- dentalis) and cat-tail (po'psam, Typha latifolia)326 cords for wefts. Such cords are rolled on the thigh. The two-ply cords are made by doubling a length of material, rolling the strands side .by side down the thigh under the palm, then, without raising the hand, rolling one strand around the.other by a quick movement up the thigh. a b' Pig. 12. Mode of commencing twined basket. The method of beginning weaving is to lay an indeterminate number of warps side by side. A weft is doubled and the free ends twined through these warps, which are divided into convenient bundles for this purpose. Having caught the last of such bundles the weft is twined back across the bundles to the starting point (fig. 12a). Again the weft is carried across the warps, but this time in diagonal twine (passing over two single warps at. a time). At the completion of this row, the ends of the warp at the margin are brought together so as to be caught in the next twine stitch (fig. 12b). At the appro- priate point the opposite marginal warp is treated similarly. This has the effect of spreading the warps radially. From this point the weav- ing continues as regular diagonal twine, crossing the wa.rps in pairs. This is continued for a short space, or quite until the edge of the base is rea.ched, when it gives way to simple twining which is carried up the sides of the basket to the rim. Additional warps, necessary to give- a flare to the sides of the basket, are added as needed. One end of the additional warp is simply caught in a twine stitch. 326 Merrill, Plants Used in Basketry, 236, 240. 1930] 179 180 University of Calfornia PubUication in Am. Arch. and Ethn. [Vol.,30 The base of the basket does not always pass into two-warp diag- onal twining so early as just stated. The weaving of some baskets, perhaps the majority, proceeds by continuously halving the bundles of warps as successive rounds of twining are placed, until.the warps Fig. 13. Base of a twined basket.. The points aa- lie on the same warp (a, beginning of diagonal twine crossing-four warps;. b, beginning of two-warp diagonal twine; c, begininng of three-strand twine; d, first stitch of design). The termination of the basket rim is on the same warp (not shown in the illustration). may be treated singly in simple twine. One example, for instance (fig. 13), having a basal diameter' of five inches, has the warps treated as large bundles for two inches, in fours with a diagonal stitch cross- ing four warps for a half inch, in pairs with two-warp diagonal stitch to beyond the edge of the base, thence simple twine for the remainder of the sides. 19]Spier: Klamath Ethnography At the juncture of base and side a ridge is produced by substitut- ing three-strand twining for simple twine for two or more rounds. This is not only decorative but adds materially to the rigidity of these otherwise flexible baskets. Three-strand twining consists in the use of three wefts, each of which passes back of a warp and crosses the face of the two adjoining. The three wefts are so inserted that each in turn passes back of successive warps.327 On the face of the basket the effect is that of a raised diagonal twine crossing two warps; on the interior, of simple twining. Similar bands of three-strand twine may be introduced on the sides or near the rim for decorative effect and strength. Mason figures two Klamath baskets woven entirely in three- strand twine.328 Designs are introduced in contrasting color. This is done either by substituting two colored wefts for the neutral colored (white to buff) cat-tail strands over the design area, or using the peculiar twine stitch described below. Both methods may appear on the same basket, even in the same area of decoration, without affecting the surface appearance in any way. In placing the decoration only the stitches of the designs are counted, not those of the background space between them. Care is thus taken to have each design symmetrical on itself; the spacing between them is merely estimated. The common material used for designs is tule root (called mai'omlo'k, the tule's hair), reddish brown in color. To finish the edge a round of diagonal twine crossing warps in pairs is woven; the ends of the warps are caught in this in the follow- ing way. Alternate warps are left standing, ultimately to be trimmed off close under the rim. The other warps are each carried to the right and forward over the edge of the rim where they are caught under the next succeeding twine stitch. As they pass forward they are twisted around the succeeding warp but one, passing back into the interior of the basket. The ends of these are then trimmed short. (Thus in my diagrammatic sketch, figure 14, warps 2, 4, 6, etc., are ultimately cut short. Warp 1 is brought forward, passed around warp 3, and carried back into the web between warps 3 and 4, being caught under the twine stitch marked A. For the sake of simplicity, the warps are shown as single strands; they are however each comprised of two strands twisted together.) This imparts stiffening to the rim.329 327 For a further explanation of this and other stitches, see O. T. Mason, Aboriginal American Basketry, 239 passim. 328 op. cit., 462, pl. 174. 329 This is somewhat in the manner of the turned-down warp edging described by 0. T. Mason, op. cit., 270. 1930] 181 182 University of California Publications in Am. Arch. and Ethn. [Vol. 30 Merrill notes that the rims of fine cooking bowls and coarse burden baskets are bound with gray nettle-bark string (Urtica breweri).330 One peculiar and almost invariable feature of Klamath baskets is that all changes in rounds of stitch begin and end at the same warp. That is, at one particular warp occurs the point, of change from diagonal to simple twine on the base, of beginning the first stitch of the design figures on the sides, of beginning and ending any bands of design on the sides, and the end of the rim finish (fig. 13). Where there are a series of independent figures on the sides, only the first stitch of one figure begins at this warp. I was told of no esoteric reason for this; it seems merely a technical habit. This trait is usual, for example, in the large collection of Kilamath baskets in the Field Museum of Natural History, but it is not invariable. On a typical tray basket, bearing a- series of concentric bands, those nearest the center begin at the same radial warp, but nearer the margin they begin only in the same general sector of the surface. This trait also appears in certain rosettes of beadwork used as dress or blanket ornaments. I 2 t34 5 6 ENDS OFWARPS . _ DIAGONALTWINE-r 3-3TRANDTWINE-- SIMPLE TWINE Fig. 14. Method of finishing the edge of a twined basket. Hats are made in this manner but differ in their appea.rance because of the different materials used. The shape of the woman's hat (k!Emu-) is subconical, its sides la-cking curvature, and rounding rather abruptly into a flattened top.33' The man's hat (k!ane'o) is similar, but has a brim. This, I was told, existed prior to contact with whites. Hats have warps of twisted tule strands.. Twining com- mences however with wefts of finely. rolled nettle cords.332 The sides of the hat are made in simple twine technique with a strip of cane (Phragmites phragrnites (L.) Karst.)333 and a tule (? or cat-tail) 330 Merrill, 222. 331 Illustrated by Barrett, Material Culture, pl. 18. Modoe hats of similar shape are shown by 0. T. Mason, op. cit., pl. 167. 332 Probably flax cords as well (Coville, Notes, 99). 333 Coville, Notes, 91. Spier: Klamath Ethnography strand as wefts.334 Designs, at least, are formed with wefts of cane (t'k!op) and dyed tule (mo'.kwas). Only the outer layer of the cane is used; this is yellowish-white. This strip is flat, and being stiff, cannot be twisted; the tule weft being fully flexible, wraps a-round both warp and cane weft. The effect is unique; at. alternate warps the cane weft appears passing horizontally, a.t the others the tule weft makes a wrapped stitch over warp and cane weft (fig. 15). Due to the technique the tule weft lies flat like a ribbon. Women's hats bear designs. Men's hats have none, but when the hat is completed the part covered by nettle weft (the central portion of the top) is painted. The brim of a man's hat is ent.irely woven with wefts of dyed tule (and cane?). Fig. 15. Modified simple twine stitch of basket hats. The several structural parts of such twined baskets are named. The area of weaving at the center of the base, with which the basket is begun, is called dEle'lluks; the base of the basket, tse'lalwls; the round- ing margin of the base as it merges into the side, sto'li; and t.he turned- back warps of the rim, wo'lg'tk. Twining in general is letca'; spe- cifically this is simple twine. A twine stitch over two warps is la'pwiomnotk (lap, two); over three, ski'tgaltantk; over four, wuni'- banttka letc!a (wuni'p, four). It is possible that ski'tgaltantk means three-strand twining ra.ther than three-warp diagonal twine. There is no name for open twine construction, so far as I know. It may be called g1ga'vka, many holes (from gi'nka, hole). Cat-tail cord for wefting is solo6tcpas; such cord is wrapped around the hand to make a roll for storage. Warps (of the tule only?) are called twiich.335 834 Coville (Notes, 94) states that in former times the bark of the aspen (vo'-liil, Populus tremuloides Michx.) was used to make hats. 335 Coville (Notes, 92) also notes the use of the rush (tsin-ii-6, Juncus baitious Willd.) in the weaving of light baskets. 1930] 1fS3 184 University of California Publicatins in Am. Arch. and Ethn. [VoL 30 The flat flexible tray (p'a"'la) differs from the twine bowl only in that it is made quite flat. This is used indiscriminately as a basket under which to hide the gaming sticks, or as a parching or winnowing tray. Those which are decorated are used in gambling; as they grow shabby they are put to other purposes. Barrett observes that the gambling tray is made of white material, the others of brown.336 This is doubtless true, but the Klamath do not look on these as baskets of different type. Specimens in the Field Museum collections have the crossed warps at the center twined with nettle (?) cord wefts. The edge finish is identical with that of twined bowls and hats; in other words, this is the general edge finish. Carrying baskets (ya"gi and sklogiuls) are made of rigid materials in open twine, with the warps at wide intervals. Possibly the name skJotgiuls indicates conical forms; certainly the somewhat globular type is ya"gi. One globose basket stands sixteen inches high, its flat base is twelve inches in diameter, its well-rounded sides curving inward to a mouth two feet in diameter. The general construction of this basket is similar to the flexible twined bowls described above; the difference lies in its rigid warps and wefts and in its open construc- tion. Both warp and weft are peeled willow twigs (yii'-yiik, &alix Sp.) ;337 on the sides of the basket the warps are comprised of two or three willows placed together; the wefts are always single rods. The base is begun by crossing three warps at right angles with four others. These are caught in a plain twine stitch for several rounds, changing to diagonal twine crossing two warps at a time for some distance, and completing the base with four-warp diagonal twine. The base is thus fairly solid compared to the sides. New warps are inserted in the base as needed; near the edge of the base extra warps are added so that all warps on the sides of the basket lie in pairs. The edge of the base, like that of the twined bowl, has a round or two of three-strand twine. In this case, this gives considerable rigidity to the base as these willow wefts are quite stiff. On the sides of the basket the warps, each con- sisting of willows in pairs or threes, are spaced a half inch apart; the weft rounds are three-quarters of an inch apart. The edge of the basket is identical with that of the twined bowl, save that here every warp is turned down and caught in the last row of twining. In this case the turned-down end is not caught in the next stitch beyond this warp but in the second succeeding. To stiffen the rim, a hoop of service berry wood (swlt6'ka) is bound inside the mouth by a wrap- 836 Op. cit., 254, pl. 11, figs. 2, 3; 0. T. Mason, op. oit., pL 48. 1Spier: Klamath Ethnography ping of willow bark. The carrying band of deer hide is carried over the forehead for heavy loads, over the breast for lighter ones. A conical wicker burden basket of this general type in the Field Museum (no. 61665) has a somewhat different edge finish. This is generically of the same "turn down and catch under" type, but the effect is that of a raised welt Fig. 16. Methods of adding warps in a conical carrying basket. of diagonal twine both inside and outside of the rim. From the top, the edge has a braided appearance. (I was not able to analyze the stitch because of the stiffness of the material.) This also appears on bowls made in the same fashion of willow sticks (e.g., no. 61666). In the conical forms the base and sides are not separated, hence there is no necessity for the three-strand twining near the bottom. 337 Coville, Notes, 94. 1930] 185 186 University of California Publicatioqw in Am. Arch. and Ethn. [Vol. 30 The specimen figured by Barrett, however, shows three-strand twine at the rim where it serves as stiffening.838 These baskets have but single willow warps, peeled and unsplit.339 In one example, new warps are added as usual by inserting the butt end in a twine stitch so that the butt protrudes on the inside (fig. 16a). This is however rare in this basket. Ordinarily one end of a weft element is carried up to the rim as an additional warp, a new willow twig replacing it as a weft element, the latter having its butt end caught in the twine stitch (fig. 16b). Or the butt end of an additional warp does service as a weft (warp 1 at fig. 16c). Mason describes an old conical carrying basket made of rushes.340 It is made of coarse stems of rushes. The warp begins with a few stems brought together to a point at the bottom and as the specimen widens out fresh warp stems are added. These are securely joined together by a con- tinuous coil of weft, which is a three-strand braid. At the beginning these turns of the coil touch one another, but as the work progresses and the basket widens the distance from one row to the next increases until they are nearly an inch apart at the top. The braiding is done from the outside, two of the stems showing always there and only one on the inside, resembling common twined weaving. This is the only specimen in the [National] Museum in which the whole surface is braided. In many twined baskets of the Pomo an inch or so at the bottom is thus woven. The top is finished off in the following manner: Three warp ends are braided together for at least 2 inches, turned down and cut off. The hook-shaped ends are held in place by a row of common twined weaving at the top. Just below this and close to the ends is a row of three- strand braid. Another row of the same kind is made halfway between the upper edge of the solid weaving and the border. A hoop of wood is held in place on the inside by a wrapping of coarse twine. The appearance of three- strand braid in the drawing on the inside of the basket is given by the strands of twined weaving and the ends of the warp bent over. The basket is strength- ened on the outside by five vertical rods [bound to it], and the carrying string is in three-strand braid, precisely as in the body and inargin of the basket. Height 22 inches, diameter 23 inches. The vertical rods are bound to the basket. A number of conical burden baskets in the Field Museum collections have three or four supporting sticks (e.g., no. 58493), of one-half to three-quarters of an inch in diameter. These extend considerably below the apex, where they cross and are bound with a two-strand tule cord. They extend beyond the rim, where they are tied to the rim hoop by tule cords. All of these have the open space described for an inch or two below the rim. Those made with twisted tule warps have the warps gathered by twos and threes and then braided for this distance; those having coarse untwisted warps have their warps at this point twisted about each other, but not braided. 338 Material Culture, pl. 16, no. 3. 339 Coville (Notes, 88) observed a basket apparently of split cedar. 340 Aboriginal American Basketry, 241, fig. 33. Spier: Klamath Ethnography Storage baskets (tlui ks) are used to hold plums and other edibles, or as containers for smaller twined baskets in which small articles a-re placed. These are made in much the sa.me manner as the globose burden basket previously described. The material for warp and weft is the large tule, and the weave is technically open twine; but the tule elements are so broad that there is little interstitial space. Barrett describes them as "more or less globose or flatly cylindrical rigid baskets . . . . used for general storage purposes. s341 One such I bought is nearly conical. Another I saw is roughly cylindrical but bulges two-thirds of the way to the top. It has a height of fifteen inches, both base and mouth diameters of that dimension, and a maxi- mum diameter of eighteen inches a third of the distance from the rim. This is woven in simple twine throughout; on the sides the twine rounds are about one inch apart. The margin of the base has the familiar three-strand twine round to stiffen the structure. (Some ba.skets of this type lack this marginal reinforcement.) Additional warps are inserted in the manner described for the conical burden basket above. The border treatment is similar to that described by 0. T. Mason for the conical burden basket. Like that, the turned-down warps are caught in a row of simple twine. It differs however in that this basket. made in simple twine has its last round in the body of the basket of three-strand twine. Further, the upper ends of the warps a-re neither braided nor twisted together, but are gathered in pairs (or threes ?) ; each pair is then crossed with the adjacent pair, and the ends then turned back and caught in the simple twine row as noted. In the basket which I collected each warp consists of two tule strands. Its border finish is similar to that just described in that there is an open space immedia.tely below the rim where each double warp crosses the a.djacent double warp. However the ends are treated in the manner of the solid twined baskets; one element of a warp is cut short at the rim, the other turned down to be caught in the second twine stitch succeeding. A bail or handle provided for this basket is braided of three strands of tule. The seed beater (we'ks), the winnowing or sifting basket (ti'a), and the basketry ladle (niip or se'-ot ako'-olks),342 although identical in method of manufacture, differ in size and funct.ion. These are shallow, oval baskets of rough open twine construction. The seed beater is used to.beat seeds, etc., into a burden basket carried in the free hand. The 341 op. cit., 257. 342 Coville, Wokas, 729. Coville also notes the use of rush (Junwus) in the manufacture of small spoons for temporary use (Notes, 92). 1930] 187 188 University of California Publications in Am. Arch. and Ethn. [Vol. 30 sifting basket is used to screen seeds such as wokas, "as a grater to remove the skins of roots and tubers, "33 or as a larger beater. The ladle serves to scoop floating wokas seeds from the surface of the water. The seed beater and sifting baskets are made of willows, or of split. juniper roots; the ladle'of tules. Each of these is made of a group of warps bent into oval shape, the central portion filled in with addi- tional warps, and the whole woven in open twine by a continuous. weft passing back and forth across them. One example examined is eighteen inches long with a. maximum breadth of nine. The three fundamental warps, which are bent in a loop, are each comprised of two willow rods. The first few additional warps are fastened in by doubling them and twining them over the proxima.l end of the three fundament.a.ls. The weft, a doubled willow twig, is then twined across the warps now in place, beginning at one side of the proximal end of the basket. Reaching the edge warp it is brought back across the structure for a second row, and so on continuously. A second set of additional warps is inserted by doubling them around the second row of wefting. Where further warps are needed to fill a space, wefts are transformed into warps as described for the conical burden baskets. The loose ends of the warps at the distal (handle) end of the basket a-re bound into two bundles, which are further bound to each other, to provide a grip. Essentially this construction is in a basket figured by Barrett in which the two bundles of warp ends are bound together to form a loop handle.344 When fine seeds are to be beaten, a pad of twined tules (later horse hide) was inserted in the bowl of the seed beater. An example in the Field Museum collections (no. 61785) is peculiar in that the warps are not kept parallel, as in other specimens, but are bent in zigzag fashion from one weft round to the next, so that the final appearance of the fabric resembles diagonal wicker work. Recently the Klamath living at Yainax on the eastern part of the reserva.tion have learned to make sifting baskets in imitation of their neighbors, the Northern Paiute. An example I collected is of the same general nature as the Klamath sifters, but broadly elliptical rather than ova.te. This is woven in compact diagonal twine, regularly crossing the warps in pairs but occasionally carelessly slipping into three warp stitches. The warps are willow twigs set close together, the wefts split willow twigs. The underlying construction is the 343 Barrett, Material Culture, 274. 344 Material C'ulture, pl. 16, fig. 2. Spier: Klamath Ethnography same as that described above; several fundamental warps bent into elliptical form, a few additional warps turned back over the proximal edge of these, with the remaining warps doubled over the first few rows of twining. The wefting is done with continuous strands. The butt ends of the warps are merely held in place by the last rows of wefting. A willow rod is bound at intervals to the.rim of the basket to stiffen it; the ends of the rod do not meet at the distal end. Triangular fans (wliliu'ks) used in conjunction with-the parching trays are also described by Barrett as plat.ters "used for serving food, p.articularly broiled or otherwise cooked fish and roasted meat. "345 As plates they may be called sa'plius. No details of their construction were obtained, but from Barrett's illustrations it appears that these are made by doubling a bundle of tules so that their free ends diverge, serving as warps. The bundle is roughly tied about at the joint to hold it together. The wefting is of tules in open simple twine. This is a continuous strand, a.pparently commencing at the joint or apex; where the wefts leave the web to enter for another row the weft elements a-re twisted around each other. The loose ends of the warps of one of the three baskets he figures protrude beyond the la.st twine row; in the other two the warps may have been turned back and caught in this row of stitches. Circular baskets made with coarse tule warps and wefts in close simple twine are more rarely found, a.ccording to this author. " Occa- siona.lly they are bound a-round the rim with a hoop. " A vessel in the Field Museum (no. 61758), made in this fashiQn, is peculiar in that it is elliptical. It is about seven inches long, five wide, and three deep. A round of three-strand twine appears midway of the base, and again at the juncture of the base and side. Stiffening is also provided by a doubled strand of tule sewed lengthwise to the inside of the base. Several other forms are found, among them a two-handled tray of tule. This is in open twine; a series of pa.rallel warps bound together at each end, and wefted in simple twine stitch with a continuous strand. Similar to the last mentioned is a "large canoe-shaped basket of tule, in openwork plain twining, used as a receptacle in gathering wokas pods," which closely agrees in construction. A pouch of plain twined tule is of similar manufacture.346 A fish trap, with willow rod warps, and quivers of tule, a-re made in open twine, according to Barrett. The cradle for a newborn infant is also manufactured in open twine construction.347 345 Op. cit., 256, pl. 13, figs. 1, 4, 6. 34f Barrett, op. cit., pl. 13, figs. 3, 5; pl. 19, fig. 3; pl. 14, fig. 1. 347 Op. cit., pl. 19, fig. 1; pl. 20, fig. 1; pl. 14, fig. 3. 1930] 189 190 University of California Publications in Ain. Arch. and Ethn. [Vol. 30 Woven bags for storage purposes, such as holding wokas and other seeds, are called wa'klo and t'a'yis. The former are of the size and shape of our pillow cases and made with warp and weft of nettle cords; the latter are of tule. Barrett records that the latter were so large that, filled with wokas seed, one alone was sufficient load.348 Gatschet names two sizes of sack, wi'llishik (the larger sacks generi- cally) and w'axoks. The former holds fifty pounds of seed, the latter twice that amount.349 Perhaps these are made in the same fashion; my information applies to the former. These are flat bags. The warps are doubled at their midpoints, so that one end will form the warp on the nearer side of the bag with the other half as warp of the other side. The weft is a simple twine stitch woven continuously around the bag, i.e., first on the nearer side, then passing to the other, and so on. Weaving begins at the points of doubling of the warps so that the bottom of the bag is closed, a mere fold in the fabric. Modern examples of this type of bag are woven in diagonal twine, but this is in imitation of fabrics of the whites. One such specimen alternates rows of diagonal twine with wefts inserted in tabby-weave. Certain flat bowls called k!we'lu are not made by the Klamath, but originate among the Pit river peoples who trade them to the Modoe. These are of solid twine weave and highly decorated. Coiled basketry is rare among the Klamath. In fact, I was told it was never made, yet the same woman informant gave'the following description of the water basket (name unknown to her). This is a bucket-shaped affair, possibly twelve inches high, with a mouth diameter of ten inches. It lacks a bail. The coil has a two-rod foun- dation of split tamarack twigs (wa'ku), sewn with tamarack roots (wa'ku Emwa'wu-k). Stitching is on the nearer face, proceeding from left to right, working from the outside. That is, the coil is counter- clockwise to one looking into the mouth of the basket. This bears no designs, nor is it painted. It receives a coating of pitch on the outside to render it watertight. I suggest that this type of basket is of recent introduction in imitation of Northern Paiute technique, although their baskets have constricted necks, I was told. Coiling is called skle'nsis, sewing. The Northern Paiute also make coiled bowls on a two-rod foundation, although they normally use a single rod. These rods are round willow twigs; split willow is used for the sewing. The foundation rods are so placed that the 348 op. Cit., 256. 349 1:76. 193]Spier: Klamath Ethnography two lie in the same vertical plane.350 The split willow strands are flat and are so placed as to break joints, leaving uncovered sections of the foundation rods exposed between the coil stitches. In former days hats were the only baskets decorated by the intro- duction of designs. Today the majority of baskets are twined bowls, intended largely for sale to the whites, to which these decorations have been transferred. Or, one might legitimately say that the modern bowls are degenerate examples of the old hats, since the techniques are practically identical. As to the materials used in decoration, Barrett observes that the "designs are usually worked out in the reddish brown roots of the tule, though the outer leaf skin of a certain jointed rush [Phragmites ] which provides a shiny, creamy white material is also used. For the finer baskets the quills of the porcupine, dyed yellow by means of a yellow moss, probably the widely used Evernica vulpina;, are used, .351 or more rarely, tule leaf, aged to yellow, according to Merrill. The latter also notes that rarely tule and porcupine quills are dyed a bright pink with the root of an unknown plant.852 Alder bark (wip'-lam, Alnus tenuifolia Nutt.) furnishes an orange dye, according to Coville,353 which may have been used for basket materials. The bark is peeled from the tree in spring or scraped at other seasons, and boiled in water. Coville also states that tule strands are dyed black, and called m6k'-wiis. "Ordinarily this is accomplished by immersing the tule stems in the black mud of sluggish springs containing iron. A superior color, however, is obtained by the addition of a quantity of wokas hulls, which contain a large amount of tannin. The same result is now frequently secured by these Indians by prolonged soak- ing of the tule stems in an iron kettle, in water containing a liberal amount of the hulls. 354 A number of designs placed on bowls and hats were recorded (fig. 17).3 These bear names, and according to my information, the habit of naming them antedates the whites. Decoration is generically 350 My notes not being altogether clear on this point, they have been corrected by Miss Isabel Kelly. 351 op. cit., 254. Coville (Notes, 88) confirms this identification (native name, shwa'-wi-sim) and adds that the quills are obtained from the Modoc, but I see no reason why they should not obtain their own. 352 P. 221. 353 Notes, 94. 354 Wokas, 737; Notes, 92. 355 These are arranged so that the rim of the basket must be understood to be above the design, that is, in the order of their manufacture. My notes are uncertain whether figure 17i is correctly placed. 1930] 191 192 University of California Publcations in Am. Arch. and Ethn. [Vol. 30 soma'lwas, to draw, to design. The Klamath gave me to understand that these figures are in no sense representative nor symbolical. The names are solely design names for convenience of reference. Decoration consists ordinarily of large unconnected units, leaving much background space. A basket rarely has more than one kind of design on it, although these larger units may be combined with narrow bands around the base or rim. Simple decoration sometimes appears on the bottom of the basket where it cannot be seen, but this is intelli- gible if we remember that in former times this was the top of the hat. 4 d 4~~~~~~~~*~~M k Fig. 17. Basketry designs. Beside a few isolated self-symmet-rical units, the majority of designs are arranged in bands extending obliquely upward to the left. Others are disposed as horizontal bands. Rings around the margin of the b-ase are called loga'kiomn'otk. The solid ring is sometimes interrupted at frequent points (fig. 17a, b). Common border decorations, at either- the lower margin of the side or the rim, are shown in figure 17c, d, e, and f. These are all called o'pk!o'tk. Several designs are formed by a series of triangles laid as horizontal or oblique bands. One such is called motcHiks, ocean clams (fig. 17g), another (h) sawa'lsalt'is, arrowheads (from sawa'ls, a.rrow- head). The reason for the first name, a.t least, is obscure; the latter is more intelligib-le. One basket which ca.rries the latter design has five triangular elements in each u.nit of decoration. Three such units 0Spier; Klamnath Ethnography are on the basket, so placed that the upper end of one overlaps the lower end of the next to the left. The units are of exactly the same length, which suggests that the. stitches were counted with some accuracy. But they are disposed somewhat irregularly around the basket. Figure 17i, which may be upside down, is Crow's knee, ga'kuimk!o'l6ntc (from gak, crow; k!o'lontc, knee). The knee is rep- resented by the angles in the median line. A design which does not occur in band form, but is isola.ted and symmetrical upon itself (fig. 17j) may be ca.lled sklotco'ta, legs (tc!oks is the proper word for leg). This is properly the name for the vertical lines; the horizontal spur is patc, foot. An old Klamath design is shown in figure 17k, but this is called mo'talta'ntk, because Pit river peoples (mo'Etuwas) used designs of this sort. The Pelican bay division of the Klamath call it. sto'kbunks a'ltantkc (sto'kbunks,. leech). On one basket ten rhomboids form each unit of design: the unit extends diagonally nearly around the basket, and overlaps with two other units. A similar design (fig. 171) has the units arranged as horizontal bands. There are four of these units disposed on the four quarters of the basket, so tha.t their ends overlap. This represents a mountain flower: the name of the design is unknown.356 It is interesting to note that the number of elements in one design unit (h) is five, in another (k) is ten, in view of the frequent use of the significant number of the Klamath, five. On the other hand, the number of elements and units in other cases is not a multiple of five. WEAPONS The warrior's armament consists of bow, club, sometimes a spear, with body armor and perhaps a shield. I have seen none of these; what follows is from descriptions. Dependence rests primarily on the bow both in warfare and hunt- ing. Hand to hand fighting was not mentioned and the use of snares and traps is slight. Waterfowl are ordinarily shot with a reed arrow so light that it will float. The Klamath never shot. fish although in many places the water is sufficiently clear. Arrow poisoning was not mentioned to me, but Coville was given the following description: The root [of the water hemlock (ski'-wiinks, Circuta mauulata I.)] mashed- and mixed with poison from a rattlesnake 's poison sacs or with the decomposed. liver of a deer or some other animal, which has been buried in the ground a 356 Other designs are illustrated by Barrett, op. cit., pls. 11, 15, 18; 0. T. Mason, op. oit., pls. 48, 174. These conform in style. 1930] 193 194 University of California Publications in Am. Arch. and Ethn. [Vol. 30 few days, was used to poison war arrows, the heads of the arrows being dipped in the moist mixture and dried over a special kind of fire with a certain cere- monial. Circuta was sometimes used among Indians to poison people in very much the same way as arsenic or other well-known poisons are used in civilized communities. It was also stated that pieces of dead fish dried in a certain way (my informant did not know exactly how) are a deadly poison when taken in food.357 The bow (ente'is) is normally three feet long or a little more, although longer bows are used by strong men. Bows are ordinarily made of juniper wood, which is very hard. Better bows, such as wealthy men might have, are of yew, which has only a moderate growth on the Cascades.358 Boys' bows are made of juniper, according to Coville, men preferring the yew as superior. The Molala make theirs of yew (tc!opi'nks). Juniper (l'ai'elo) grows on the deserts to the east and in the Modoc country, but may also be found on the high mesa above Modoc point on Klamath lake. War bows are backed with sinew: all are, according to another informant, who held that an unbacked bow will dry out and snap. His statement is borne out by Coville: "they are commonly about a meter in length, backed with sinew, the tips covered with fishskin and the string made of twisted sinew." On this sinew layer designs (me'eks) are drawn in green. Barrett figures an unbacked bow bearing a simple decoration on the inner, concave face.359 A purple color derived from the seed coat of buck-brush (Kunzia. tridentata) is sometimes used to give a temporary color to bows, arrows, and other objects.360 The bow is broad, flat, and with a constricted grip at the middle. The ends recurve toward the outer, convex face; these are notched to receive the string. The grip is wrapped with a strip of mink skin. The expanded part either side of the grip is three fingers wide, taper- ing to a single finger-width at the tips. How the bow length is measured was not known to my informant. Three bows figured by Barrett are forty to forty-three inches long. The bow string (na"'lis) is back sinew of a deer, rolled on the thigh. Bows taken from the Pit river peoples are reputed as of better manufacture; "they were very strict in the making of these." 857 Coville, Notes, 101. Gatschet (1:150) mentions this plant as poisonous, adding several others, among them the limbs of a young pine and wild parsnips. 358 "Abundant on the western slope of the Cascades, occurring on the east- ern slope in Union Creek and on Pelican Bay, Klamath Lake" (Coville, Notes, 89). 359 Material Culture, pl. 20, fig. 5. 360 Coville, Notes, 98. Spier: Klamath Ethnography In shooting, the bow is held not quite vertical, but with its upper tip somewhat to the bowman's left. The arrow rests on the right side of the bow. The arrow release is the simple primary, held between the bent forefinger and thumb only. The other fingers do not touch the string. The bow is carried in the quiver with the arrows. No bow guard is worn. Arrows are both wood and reed-shafted. The wood is service berry or lolk (glok?), a very tough wood growing in the southern Modoc country (probably currant, Ribes cereum). Coville also mentions Philadelphus lewisii Pursh., which grows at Wocus bay, and for light shafts wild rose wood.86' Reeds of two kinds are used, cal (Phragmites phrargmites) and wltci'v-E, which must be provided with foreshafts (la') of light wood, currant wood (lolk, Ribes cereubm),862 or mountain mahogany (? temo'lolkum). War arrows (l!e'ls) have wooden shafts thirty inches long, are feathered for seven or eight inches of their length, and are provided with obsidian heads. The arrowhead (sa'wals) is not set in a split in the end of the shaft but tied on one side of the tip loosely so that it will remain in the wound. Other wooden arrows (do6ldji) used in play or hunting small game are feathered but bear no heads. Arrowshafts are smoothed with a scour- ing rush (wii-chiik'-wis, Equisetumrn hyemale L.) .368 Reed arrows, used primarily for water fowl, are usually feathered. Barrett figures some, thirty-two to thirty-eight inches long. The end of the reed is wrapped with pitched sinew, so that it will not split when the foreshaft is driven in. The foreshaft is pointed by burning and often has a little wad of pitch near the point in lieu of a barb. My informants denied that the intention was to cause the arrow to skip on the surface of the water as recorded by Barrett.364 Shafts are not painted nor are the wooden ones grooved. Feathering (lalo'tk) is placed on most arrows. Wing plumes of eagles, mouse, and chicken hawks are used; said to be selected solely for their appearance. War arrows carry bald-eagle feathers. The quills are split and the narrower side of the feather discarded. Four vanes are mounted on the shaft; whether parallel to its axis or spiralled is immaterial. The vane is glued with pitch for its whole length and bound at each end by a sinew wrapping. At the butt end the terminal plumules are turned back and caught under the wrap- 361 Notes, 97, 99. 362 Ibid., 92, 97. 363 Ibid. 88. 864 Material Culture, 247. 1930] 195 196 University of California Publications in Am. Arch. and Ethn. [Vol. 30 ping. The free edge of the vane is then notched throughout its length to give a servated appearance. The butt of the arrow is cut wedge- shaped transversely to the nock. The quiver (do'konks) is slung under the left arm by a cord over the opposite shoulder so that the mouth is at the breast. A war leader, a rich man, might carry one under each arm, and a third at. his back. To put on a quiver is called sodo'k!e. Quivers are commonly made of tules, well dried, and twined at intervals with sage-brush bark strings. Barrett figures one thirty-two inches in length. Wealthy men, as Barrett notes,365 had quivers of skin: raccoon, marmot, or otter. The hide is stretched over a frame of sticks or stuffed with grass that it may dry in the proper shape. The club (de'kis, l%'6ples, or k1u'pg-us, perhaps several types) is a limb of a tree having a knot to serve as the ball-head. Others are simply long sticks. The spear (yo'kmom) is a short stabbing lance, four to six feet long, bearing a large obsidian blade. The spear is rather the weapon of a woman than a man. In the course of a raid on the Upland Takelma, the Klamath chief LI&le'ks stabbed their chief through the throat after he had been shot down. Both the long skin tunic and slat armor are in use. The tunic (kol!no'ls) of elk hide, alternatively of bear or deer skin, hangs to the calf, has holes for the arms, and is tied together down the sides. This is a single layer of skin, except that a small separate piece of buckskin (called sep6'kdilE) might be worn under it as additional pro- tection for the front of the trunk. Slat armor (da/lok) is the familiar waistcoat form; vertical wooden slats, two inches wide, split from any sort of tree, and twined with nettle cords. It is long front and back, but necessarily short under the arms, opens in the front, and ties over the shoulders with buckskin thongs. The round hand shield (snoto'wltc) is but little used, to judge from my informants' ignorance of it. This is strange for it may well be suspected as of recent introduction among them from Plains-like tribes to the northeast. It was described as eighteen inches in diameter and of unknown material, but undoubtedly skin. These informants knew nothing of the use of a skin as a shield hung like a curtain in front of the body. The arms are sometimes protected by mittens (wakakpks) of arm or elbow length. These are sock-like affairs of coyote or other fur. 365 Op. cit., pl. 20, fig. 1, 255. Spier: Klamath Ethnography HOUSES AND HOUSE LIFE Houses are for the most part so closely crowded in the settlements that their peripheries nearly touch. At least this is the evidence of the house pits, which my informants would have me believe were occupied at the same time. Some of them are small structures, used for cooking or to house slaves, crowded between the large earth-lodges. House-building seems to be one of the most considerable occupa- tions of the year. Earth-lodges are frequently torn down in the spring to be rebuilt when snow flies in the autumn. In the second month, September, everyone is busied with house-building, commenc- ing with the preparation of the tule mats that cover all the lodges. The houses are all completed in the fourth month (December) by the time full winter arrives, when the shamanistic performances begin. New timbers have to be prepared, as those from the razed structure are commonly used for firewood. Gatschet implies that occupancy of the winter house is always inaugurated by a feast.366 The mat-covered summer houses are built anew in the spring. It is quite likely that the covering mats are thoroughly rotted and dilapidated after stand- ing through a hard Klamath winter. The timbering of houses is man's work; women place the grass and dirt covering. The characteristic house is a circular, conical-roofed affair. It is essentially of the same shape and construction whether merely mat- covered or set over a pit and roofed with planks and earth. The Klamath definitely distinguish the two types however, the earth-lodge, lu" ltEmaloks, and the mat-covered summer house, wu'ke'plJoks. In addition a small dome-shaped mat-lodge (sti'na'c) is used in summer, when traveling, or for cooking.367 Poor families who cannot afford the larger earth-lodge, build dome-shaped lodges for winter use, set- ting them over pits for the sake of the added warmth. The earliest descriptions of these houses are by Ogden (1826) and Fremont (1843-46). Ogden's account of what are probably earth- 366 1:75. 367 Barrett describes a summer house possibly of different form. This is elliptical or rectangular in ground plan, with "sides which slope very abruptly to the groun.d and a comparatively flat top. The framework is of willow poles stuck into the ground and brought together along the ridgepole, to which they are bound securely." This suggests in part the structure of the dome-shaped lodge, but implies the presence of at least two posts bearing a ridgepole, a type of construction which my infornxants specifically denied. The photograph Barrett figures belies his description: the form seems to be the regular struc- ture having four posts with the side poles resting on the rafters joining these (Material Culture, 244, pl. 11, fig. 1). 197 1930] 198 University of California Publication in Am. Arch. and Ethn. [Vol. 30 lodges is curious. "It [a Klamath m.arsh village] was composed of 20 tents built on the water approachable only by canoes, the tents built of large logs shaped like block houses the foundation stone or gravel made solid by piles sunk 6 ft. deep .R.e.. . Reached Indian village of five huts, hut large size square made of earth flat on top the door at the top a defense against arrows but not balls."368 Fremont notes "they were large round huts, perhaps 20 feet in diameter, with rounded tops, on which was the door by which they descended into the interior. Within, they were supported by posts and beams." At another place "the huts.. ... were made of tall rushes and willows. ' 3f9 There is considerable variation in the size of houses, even those of one type. Earth-lodges range in diameter from twelve to thirty feet or more, the larger accommodating several families. The largest pit I saw measured thirty-five feet in diameter. The largest earth-lodges are invariably those of shama.ns, larger even that those of chiefs. This is not only the statement of my several informants, but on every site I visited the largest pits were pointed out to me as the former resi- dences of shamans. The shaman's lodge, to be sure, houses the winter performances, but this also implies that shamans are the wealthiest individuals, or were so at least before the ascendancy of the chiefs during the last century. Winter houses are commonly larger than the corresponding summer mat-covered forms. The semi-subterranean earth-lodge is erected over a shallow pit (sloko'ps). This is commonly but knee deep, but the thirty-five-foot pit mentioned above had been excavated waist deep or even more. All the clearly defined house pits that I saw were circular, but I got the impression from one informant that the house might be somewhat squared. This is probably the case only with the summer mat-lodge: wu'ke'plolks. The circle is not measured in any way. The pit is dug with digging sticks, the earth thrown back all around the excavation with storage baskets (tlJuks) or slabs of pine bark. The roof has the form of a truncated, wedge-shaped cone, resting as it does on the circular periphery of the pit but with its upper trun- cated surface an elonga.ted rectangle. The main frame is formed by four central posts, set to form the corners of this rectangle.370 The tops of these are connected by stringers. Poles extending obliquely 368 Elliott, The Ogden Journals, 210. 369 pr6mont, Narrative, 122; Memoirs, 494. 370Barrett implies a single center post ("climbs down .... by means of steps cut into the center pole itself," p. 244), but this was denied by my informants. Spier: Klamath Ethnography from this rectangular frame to the edge of the pit throughout its whole circumference form the conical roof. The whole is then covered with mats, grass, and dirt, leaving only a hatchway open in the center of the rectangle. The lodge is entered by walking up the slop- ing roof to the hatchway, through which a ladder from the interior protrudes. The rectangle formed by the central posts (A, fig. 18) measures about fifteen feet from front to rear in a large house, with a width -of five feet. These posts (studi'ls) are tree trunks, twelve inches in diameter, bearing forks nine or ten feet. above the floor level. String- ~~~~~~~~~~~~~~~~~~~~~c Fig 18. Plan and section of earth-lodge showing construction. *ers connect these in pairs; the first set (B) across the short ends with longer stringers (C) resting on these in the same forks. These are poles of ten inch diameter; the first pair is nameless, the second called .sEo"nhis. On this frame rest four oblique rafters (D, stc'au'ns) let into holes at the brink of the pit. These are split logs. A supplemen- tary rafter (E, swlkke'nis) is laid across each pa.ir of these near its upper end. Planks or poles (F, ctcau'uis) are then laid close together along these supplementary rafters, with their lower ends resting on the brink, but not in holes. Similar poles are set obliquely a-round the corners and along each end, where they rest on B. The hatchway is formed within the rectangular roof frame by a series of long poles or planks (G, also called sEo'nhis) laid close together and resting on B. Its ends are blocked in by similar, shorter crosspieces (H) resting *on these in turn. The hatchway is nearly as wide as the spacing of the long stringers (C), three to four feet; its forward end is quite close to the cross-member (B) at the front. of the house, but the roof is *decked over for several feet at the rear, leaving an opening some ten feet in length. Short sticks and bark are placed over any crevices still remaining in the superstructure. 1930] 199 200 Unniversity of California Publications in Am. Arch. and Ethn. [Vol. 30 The roof is shingled with long mats laid in horizontal courses. The variety used is a relatively poor twined mat, called ste'holos, made of tule, swamp grass, or cat-tail rushes. Dry grass (not brush) is heaped on the mats halfway up the sides and the whole roof then covered with a thick layer of dirt. The grass is omitted from the upper part for! fear sparks from the fire benea-th the hatchway may catch in it. The mats are three and a half to four feet wide and so long that they stretch more than halfway around the roof. Swamp- grass mats ma.ke a warmer structure than tule. No mats are placed on the inner side of the roof, nor around the sides of the pit. The floor however is liberally carpeted with them. Food is commonly stored around the outer margin of the roof, buried under the dirt cover. The dirt is banked so thoroughly there that the lodge needs no cir- cumscribing ditch to protect it from rain water. The hatchway in the roof (ginkl'a, hole) is of considerable size; about ten feet by three or four in a large house. At night a series of small sticks are laid transversely across it and a mat, called k'ai'ictis, which is rolled up at the rear end of the hatchway during the day, is stretched over the opening to close it. Any kind of ste'holos mat may be used for the hatch, but swamp grass is preferable to tule which becomes brittle when dry. To keep the snow from drifting into the open hatchway, a, mat may be set. up to windward on sticks driven into the dirt covering. Houses face the southeastern quarter, because the prevailing winds are from the west. The front of the house is the side having a set of steps leading up the sloping roof to the hatchway. It seems doubtful that a house on the east bank of a river, for instance, would have its entrance away from the stream, but so I wa.s told. The circumstances is that a-long the Williamson river valley, the only narrow valley where this arrangement might prove awkward, the majority of set- tlements are on the west bank. The outer steps (astono'lluks) are formed of two small tree trunks lying upright along the sloping roof. Short limbs left projecting from these hold a series of cross-sticks serving as rungs (waki's), with .dirt packed behind them to make a set of solid steps. The ladder reaching the hatchway from within is made of similar poles (ste'stkls) with the rungs tied on. There were ten rungs, a foot apart, in a large house described to me.371 This ladder is set at the southeastern end of the opening, that is the end opposite the rolled-up mat door. It is 371 This may well be the operation of the pattern number five. 1930] Spier: Klamath Ethnography 2t1 nearly vertical and protrudes considerably above the roof level.'LM-" ders are never made of a log with holes cut through it, nor were the central posts notched for this purpose.372 The rough planks used in roofing the structure are made by split- ting hollow logs with elk-horn wedges. These are driven with a hard- wood club, oak, or mountain mahogany. Every man has three or four such wedges; axes are wholly unknown and stone mauls not used. Obviously such timbers cannot compare with the broad planks of the Northwest peoples, but are rather of the crude Yurok type. Procuring the planks is quite a little expedition. They go in four to six canoes, spending a whole day splitting enough for a single house. The people of the du'kwa settlements at the mouth of Williamson river, for instance, cross Klamath lake to its timbered western shore for their supply. I I \ | 2 ~~~HATCFHWAY / \ / _ _: Fig. 19. Named parts of the house. The several parts of the house interior as set. off by the central posts are named. That portion at the front of the house directly under the outside steps is called ste'kis (M, fig. 19); the opposite end, tka'lum (N). These spaces are used for storage. The sides of the int.erior (0) are la'lEstal and the corner sections (P) are stc!oka'ctal. These, the living quarters, a-re preferable to the end spaces presumably beca.use the narrowness of the hatchway yields a greater extent of roof cover on each side, thus affording greater protection. The sleep- ing quarters of shamans are on the northern side of the lodges, hence I assume that this segment is always occupied by the head of the household. The shaman dances principally on this side of the central area. 372 Compare Barrett, 244. . I - -1 SF.? 19 , - I-IN f 202 University of California Publiodtions in Am. Arch. and Ethn. [Vol. 30 The summer house (wu'ke'ploks) differs. from the earth-lodge only in the absence of the pit and of the dirt covering on the roof. It is circular and built on the same four-post frame. The covering is of mats alone. Dirt is banked around the base. Occasionally one is erected over a. shallow pit, when it does not differ materially from an earth-lodge. These houses are smaller than the earth-lodges, ranging down to the dome-shaped houses in diameter. They have the height of the earth-lodge. No ladder is necessary with these houses, since the entrance is on the ground level. A space is left for a doorway between two of the slanting roof poles at the front of the house. A mat hung from a cross-stick halfway up these poles serves as a door. When it is not wanted, it is rolled up and tied above the opening. Sometimes a mat is hung from one of these poles inside the house as a. partition (hocto6ptls) to protect a bed from the opening. The dome-shaped lodge (sti'n'a'c) is in use the year round, stand- ing beside the earth-lodge as a. cook house. When the people are trav- eling, it may be hastily constructed of mats carried for the purpose. The more permanent structures are ten feet in diameter and eight or nine feet high-as high as one can reach to put things away overhead. Poor families intending these for winter use make them of heavy construction and dig out the interior to a depth of a foot for the warmth thus afforded. A heavy snowfall is shoveled off with canoe paddles. The fra.me of this lodge is built up of a series of a.rches (stce'cgiuc or stce'cElga), each made of two willow poles stuck into the ground at opposite points of the circle and with their upper ends lashed together overhead. The two main arches are parallel and rise to the full height of the lodge. Another is set on each side of this pair, parallel to them, but rising to a lesser height. Then the other two quadrants are filled in with poles which are bent over and lashed to these four arches. These poles are set close together to form a firm structure. There are no horizontal ribs. A space is left on the south- eastern side for the doorway. The frame of both houses is- shingled with mats; rarely these are also placed on the inside. Tule mats, or swamp-grass-which is warmer-are always placed with the stems vertiial so that the rain and snow will run off readily: Sewn mats (st!ops) are preferred to twined for this reason. The mats are tied on; two or three overlapping tiers reach to the top. They are sometimes so long that two will reach around the lodge. For winter use, two, three, or four thicknesses are Spier: Klamath Ethnography used ;373 then grass or pine boughs, covered with dirt, are heaped half- way up the side. The outer mats are held in place by tying them along the edges to poles propped against the side of the house. The rope (pekl'os) used for this is three strand, braided of grass or the outer layer of tule. In placing the mats, a space about two feet across is left at the top for a smoke hole, and another only three and a half feet high for a doorway. One has to stoop to enter. The door is a mat like that of the summer house; it is not weighted nor tied down. Drying racks are not placed inside the houses. All fish and meat are dried on racks near-by. Houses are not decorated, nor are the house posts carved or painted except by shamans. These may paint them with horizontal stripes, alternately red and white. One informant had it that men's faces, snakes, lizards, and various other animals, not spirits, are painted on the posts, but I am doubtful. Shamans also hung stuffed bird and animal skins about the interior. A wooden image and a stuffed crane skin are set up outside on the northern side of the hatchway, i.e., above the shaman's bed. Houses are not given individual names, Yurok fashion, though the localizing tendency of the Klamath might have found an outlet in this. The fireplace (sne'lOuks) in the earth-lodge is in the middle of the floor, between the four center posts and under the open hatchway. Neither a pit nor stones are used to confine the fire. Most of the time there is no fire on the hearth, only in the morning and evening; at other times the hatch is closed. A huge fire is built when it is nearly sundown, and after the mat is drawn across the opening the heat is so great that blankets may be dispensed with. This despite the severe winters; at this altitude (4000 feet) lakes and streams are frozen solid. Fires are usually made by piling short lengths of wood in log- cabin (crib-work) fashion to a height of eighteen inches. They cook when this is reduced to a bed of coals. A piece of rotten wood is ignited to carry fire from one place to another. Cooking ordinarily takes place outside the dwelling in a dome- shaped house (sti'na'c) erected near-by. An earth-lodge on lower Williamson river, for example, set some fifty yards from the stream, had its cooking lodge close to the bank. All the women of the house- hold prepare their food together, even where several families have joint tenancy. It is carried to the earth-lodge and there passed about 373 Barrett gives the successive layers as made of "a kind of reed, " triangular-stemmed tule, and circular-stemmed tule (op. cit., 244). 1930] 203 2,04 University of California Publication in Am. Arch. and Ethn. [Vol. 30 as in "one big fa.mily. " This is inevitably a matter of personal pref- erence; some dwellings have two cooking lodges attached, the families eating separately and sharing food only as friendship dictates. The mortars and grinding slabs are kept in the cooking-lodge. The intro- duction of flour has wrought a change in these ha-bits. Loaves of brea.d a-re baked in the ashes in the earth-lodge itself. Flour is always sha.red among the inmates. I was assured tha.t there was no earlier custom resembling this. But two meals a day are eaten, morning and evening. I was also told that there is no regular period for meals. Visitors are given something on their arrival, such as a little pond-lily seed, but no stress is placed on this phase of hospitality. A sort of gra.ce is sa.id before eating. The entire floor of the earth-lodge or summer mat-house is carpeted with mats except for the space within the four central posts. Thicker piles of these for beds are placed all around the walls, but the best locations are at the sides (0, fig. 19). Children might sleep in the corners. People always sleep with their feet toward the fire, no matter in wha.t direction their heads then lie.374 Beds (slai'is) are of mats of various sorts. Rich people use tule, the twined mats (slai'is) being preferred to the sewn (st!6ps). One or more of these laid on dry grass serves as a mattress, or they sleep directly on the grass. One or two more form the covering. The few who have them use grizzly bear, coyote, wolf, lynx (? slowa'), or deer pelts, or severa.l Bassariscus (? slet) skins sewn together, for mattress and blanket (sa'walc). A blanket for a little child is ma.de from the skin of a red fox, a st.retched raccoon skin, or several marten skins sewn together. Blankets (called solumtai') are also made of tule and cat-tail twined to produce a checkerboard decoration. Others are made of tules with the wing feathers of water fowl (pelicans, geese, swans, ducks, etc.) caught in one face. These are called l'a'cda- lap'o6tE, wing feathers (lac) put together. Another type of feather blanket (called k'a'lyE) is made of the dried skins of waterfowl t.wisted with a cord to form a fluffy rope. This is twined with cords inserted at intervals to form the blanket. Pillows (sE'o'lhas) are a hea.p of grass, or anything else, even wood, but the wooden headrest of the lower Klamath river tribes is unknown. Gatschet records a sack stuffed with mallard duck down for a pillow.375 374 I could find no notion that a sleeper is affected by moonlight. 375 1:144. 0Spier: Klamnath Ethnography It is customary to rise (b6tlgal) before the sun has risen, at least in winter. At that season when they sit late telling stories, they retire (k'ta'ndja) when Orion's belt has risen two-thirds of the dis- tance to the zenith (lo'lal6ks, bedtime). Old people sun themselves on the roof tops. Those who are decrepit and cannot climb into the earth-lodge are housed in mat-houses near-by. SWEAT-LODGES The earth-lodge is so warm, at least at night, that it might well have taken the place of the smaller sweat-lodges.. These however are regularly made, preferably beside a body of water into which the bathers can jump after sweating. The lodges (spo'kliks) are of two Fig. 20. Winter sweat-lodge. A portion of a summer sweat-lodge appears at the right. types, differentiated, it was suggested, for summer and winter use. There are now more of the former type in use than the latter, which suggests a case of substitution. The "summer" lodge is dome-shaped and mat-covered, hence sometimes called sti'na'c like the dome-shaped dwelling. Willows are set in the ground about a circle four to six feet in diameter, arched, and bound together at a height of three and a half feet. There are no horizontal ribs. Tule mats form the covering, with layers of green grass between to prevent the escape of the steam. When the summer lodge is kept for winter use, dirt is banked against the sides. The car- peting is of grass or leaves, green or dry according to season. The "winter" lodge (ke'lowac, a descriptive term?) is built over an elliptical pit, eighteen inches or less in depth. It is smaller than the summer lodge, and up to five feet in length. The roof over this 205 1930] 206 TUniversity of California PulbUcations in Am. Arch. and Ethn. [Vol. 30 is gable-shaped, the ridgepole being supported at each end on a pair of struts thrust obliquely into the margin of the pit. Sticks and bark resting on this form the roof; a layer of dry grass and dirt covers the whole. This stands three and a half feet above the ground. The entrance, which is quite small, is covered by a mat (fig. 20). Stones, heated outside, are rolled into a hollow in the lodge. There does not seem to be any prescribed place for this, as among other peoples. The stones are said to be at the rear in the winter lodge, and anywhere in the summer affair, but in a winter lodge that I saw, the hollow was near the doorway to the right of one entering, and at the back in a summer lodge. For the use of five or ten men, I was told, a large summer lodge has the stones at the center. Five is the Klamath pattern number, but it was said that there was no prescrip- tion of the number who may enter at one time, nor of the number of times they sweat. Water is thrown on the hot stones after they are somewhat cooled, otherwise the steam would scald the inmates. There is a certain amount of rivalry on the part of those who have the water to drive the others out by raising steam; sometimes four or five have it at one time.376 They talk to the stones when sprinkling them, asking to be cured, to live long, work well, to have luck in gambling, and to get spirits. This is the same as talking to the water when swimming on a vision quest. "The stones are 'doctor'; the water is 'doctor.' They talk to the wind, the mountains, the water. Everyone outside is quiet. Men and women go into these lodges together. Coville observes that "the branches and twigs of the 'cedar,' (Librocedrus decurrens Torr.) are often used in administering a sweat bath to a sick person. "377 The three sweat-lodges used solely by mourners closely resemble the "winter" lodge. In fact, the lodge just north of Klamath Agency does not differ in any respect. I have seen but one of the other two, that near Algoma. This is now a U-shaped wall of stones which once bore a rough gable roof covered with dirt. It is much larger than the winter lodge, enclosing an area ten by twenty feet.378 It may be sug- gested that conservatism has preserved the type for this use, while for secular uses it has been, or is being, supplanted by the dome- shaped lodge. 376 For a tale describing such rivalry, see p. 37. 377 Notes, 88. 378 For its description and use, see pp. 73 f. 0Spier: Klamath Ethnography CLOTHING AND ADORNMENT Costume is scanty for both men and women. Poorer individuals are equipped with simply a fringed skirt and nothing more beyond basketry hats. The wealthier, however, affect a dress which strongly resembles that of the Plains tribes. Such men are clad in breech- clout, leggings, moccasins, and on gala days buckskin shirts; women, in the long dress of Plains type, with short leggings and moccasins. This Plains costume seems to be of recent introduction; traditional feeling to that effect is reflected in the proscription of its use during shamanistic performances and in the Ghost dance of 1870-73. This might also be inferred from the fact that the Klamath represent its most westerly extension, or nearly so, in this quarter. Plains influ- ences have reached the Klamath through the Columbia river tribes, with whom direct contact dates only from the beginning of the last century. We may hazard the guess that this costume does not date back before 1800 among the Klamath, and is perhaps not so old. The fringed skirt is a garment which completely encircles the wearer's waist, not an apron nor two aprons. It is formed of tule, or sage-brush bark, or possibly of buckskin strips, braided in cords which hang thickly from the belt. These pendants are closely set at the front and back, and hang but-scantily over the hips. Distinctive names for the front and back parts (respectively hisci'sks, hangs between the legs, and histco'llc, hangs behind) suggest its relation to the familiar two-apron garment of California. Men's skirts do not reach the knee; women's hang below it. The man's shirt (tculi's) is of the Plains poncho type. This is a large skin with a hole in the middle, put on over the head. It reaches to the upper thighs and hangs loosely over the arms, without any attempt at sleeves, but is tied together under the arms. When only small skins are to be had, two of these are sewed together along the shoulders. Front and back are the same, the neck hole is round, and they are indifferent as to which side hangs in front.379 379 A shirt illustrated in a photograph shown me had deep fringes along its lower border and on the seams joining the sleeves to the body of the garment. The sleeves indicate that this is a modern derivative of the old garment. I was told that a fringe is sometimes placed horizontally across the chest. A man's shirt in the collections of the Field Museum of Natural History (no. 67883) has a short fringed opening down the breast from the neck hole, and a border of beads on the low collar rising around the neck. 1930] 207 2,08 University of California PubUcations in Am. Arch. and Ethn. [Vol. 30 The familiar triangular fla.ps below the neck in Plains shirts are lacking. Shamans alone paint their shirts, others who attempt it would be poisoned by them. The designs are spots, circles, and lines, not pictures. While I do not know the character of these designs, this suggests the protective shirts of the Ghost da'nce of 1890. Inas- much as we must assume that the shirt reached the Klamath from the east, and it is possible tha.t Plains influences are quite recent, it is possible that the shirt, or at least. its decoration, came only in Ghost dance days. Nor are shirts worn at all times; they are too valuable, "they might wear out"; which represents their unusualness in the view of the Klamath. Breechelouts (hacwi'skls) are made of buckskin, and nothing else. They a-re worn by women too, but not with the buckskin dress. It seems that. the flaps which hang over the belt are short, for they are said to be long only if the skin happened to be long. The belt is only a thin string, presumably a leather thong, for poor people make it of the soft bark of the tule. Men's leggings are thigh-length or calf-length (called respectively snasni'ks and mld&'s) ;380 women wear only the latter. These are of buckskin or fur, but poor people make them of tules. Barrett illus- trates one of the latter sort made in twined technique. The full-length leggings are tied to the belt at one point, the hips. In a photograph shown me, little deer dew claws were hung pendant from the front of the leggings just below the knees. Moccasins are worn at all times by the men, at home or when trav- eling, whereas women sometimes go barefoot at home in the summer. For the most part, moccasins are woven of tule or swamp grass; very few are made of skin. Barrett observes that the buckskin moccasins are for summer wear, the winter ones of tule.881 This is very probably the case, as the tule affair is intended primarily to hold a layer of dry grass next the foot. Tule moccasins (wipgd') are made in open twine. The fabric is brought together over the toe, but is quite without a heel. A cord laced through the loops along each side draws it up around the foot. They resemble those of the Paviotso.382 For traveling they make moccasins of this type of swamp grass, which is stronger than tule; those for home use are of tule and bear decorations. 380 Gatschet gives mitash, mitas, from Chinook jargon mimdash, derived from French mitasse (2:221). 381 Material Culture, 255, pl. 17, 19. 382 Lowie, Notes on Shoshonean Ethnography, 218, fig. 5b, c. Spier: Klamath Ethnography The skin moccasin (duimi'wE) differs from that of central and northwestern California, in being made of two pieces. Like the Cali- fornian affair, it is a one-piece moccasin in pattern, with a seam up the top from the toe and another up the heel, but it also has a D-shaped inset over the instep which is continued upward into a tongue. Some specimens ha.ve an additional strip around the ankle, sewed to the edge of the main piece, and which can be worn turned up or dowin. The drawstring passes around the ankle and across in front of the tongue.383 Moccasins are ma.de of deer or elk hide,384 with the hair inside, or even of beaver or badger skin. The last, is a very strong skin; D D C B a ~~~~~~~~~~~~~b Fig. 21. Patterns of skin moccasins. the moccasins will last a year or two. (Otter and raccoon skins are not used. ) Cut to pattern (fig. 21a.), they are folded lengthwise, stitched up the instep, across the toe, across the heel, and up the heel. The tongue (stai'skopk) is then sewed in over the instep. The pro- cedure in connection with sewing the heel is peculiar. The slit A-A is first made; the sides brought together so that the points-A, B, and C match; the cross seam (A-A) sewn, then the vertical one (B-C). This leaves a loop inside the moccasin at the heel. The moccasin is then turned inside out, so that the loop is on the outside. The corner of this is then cut off on the lines D-D. This forms a tab protruding at the heel. These seams are not lapped but sewn with a through-and- through stitch, hence when the moccasin is turned inside out, the seams at the heel, at least, are on the outside. Sewing is done with an awl and sinew. Another type of deer-skin moccasin (wokena') made in imitation of Warm Springs and Yakima style, has come into use since the boy- 383 Illustrated by Barrett, pl. 17. 384 The same informant stated on another occasion that elk hide is not used for footgear. 1930] 209 210 University of California Publications in Am. Arch. and Ethn. [Vol. 30 hood of my elderly informant. This is a simple one-piece affair, shown diagrammatically in figure 21b. The upper is cut to form a long tongue. It is sewn along the toe and one side, and up the heel.385 Snowshoes may be conveniently described at this place. The snow- shoe (ni') is the circular Californian form; a hoop of tough service berry wood, about twelve inches long by six wide.380 Within the hoop is a net of thongs on which the foot is placed; tie strings are drawn up about the ankle. My woman informant drew a sketch showing five parallel longitudinal thongs crossed transversely by five others, also parallel, thus forming a simple rectangular web. This may well have been merely diagrammatic, since five is the pattern number. Kroeber illustrates a more intricate pattern of webbing, a central lozenge filled by transverse lashings and connected with the frame by others.387 Barrett illustrates a snowshoe with still a third style of web. This seems to be a single longitudinal strip of skin, crossed by three others.388 Such shoes are also used when wading in the marshes. Mittens (wa-ka'pks) of coyote or other fur are worn in cold weather, possibly by men alone. They are also used for protection in war. These are sock-like, without fingers, and of arm or elbow length. A muffler (sau'kaks) is also made of skin. The woman's skin dress (gokc) is the long woman's garment of the Plains tribes, completely covering the body to the mid-calf. Unlike the corresponding man 's shirt, these are worn most of the time, especially in winter. In summer the lighter unbeaded dresses are used. This is not essentially a dance costume, although worn at the war and scalp dances. 1But these are held outdoors at night; and during the shaman's dances in the earth-lodge, the fringed skirt only is worn because of the great heat. It is possible, of course, that this is a rationalization of etiquette, for if the shaman's dance is older than the forms of war and scalp dance now in use, there may be an association of the older fringed skirt with one and the newer skin dress with the other. There can be little doubt that the fringed skirt is the antecedent form. Plains influences appear to be recent, and 383 This has been redrawn from a sketch made by my informant. On the Yakima reservation moccasins of this type have a tab at the heel, which my informant may have omitted. 386 Coville (Notes, 94) states that "the frames of snowshoes are usually made of willow wood. The mesh in old times was commonly made of the Rocky Mountain flax, Linum lewigii, less commonly of nettle fiber, Urtica breweri." 387 Handbook, fig. 68a; evidently taken from Mason, Aboriginal American Basketry, 215. 388 Material Culture, 255, pl. 17. Spier: Klamath Ethnography the Klamath use of the typical Plains woman's costume is an extreme marginal case in the west. Being of buckskin, not of elk hide,. two skins are necessary for.the dress. These are sewn together across the shoulders, leaving a short space for the insertion of the head. The neck ends of the skins are always across the shoulders, since this is the widest part of the skin. The contour of the dress is the familiar one, the natural contour of the dressed skin with the legs lopped off. Cape-like exten?ions, formed by the skin of the forelegs, hang loosely over the arms.. All the edges are slit into fringes and the front and back are tied together by these under the arms and down the sides. These are not belted in Plains style. A woman's belt made of strips of bark fiber is illustrated by Bar- rett.; others are made of human hair.389 While Barrett is probably right in stating that these are part of everyday dress, I was told that the belt (k'ai'llis) is worn at the puberty dance, when a first child is born, or when an infant dies. The dress of earlier times had no ornamentation, but it now has the yoke-shaped decoration of Plains type and rows of pendants. The yoke is not a separate piece, but the quill-work design, and in more recent days the bea.dwork, is applied directly on the dress. Formerly rows of elk teeth were sometimes used. There is no special name for the yoke; like any beadwork, this part is called iti's or ita'ntkos. The yoke of one dress seen dipped below each armpit and again markedly at the middle of breast and ba.ck. It was comprised of a series of horizontal stripes in beads of contrasting colors which paralleled this wavy lower margin of the yoke. Three horizonta.l rows of pendants crossed front and back; one following this lower margin, another just above the knees, and the third near the bottom of the dress. These pendant strings, which hang in pairs, are wound with colored por- cupine quills; thimbles and Chinese cash are sometimes threaded on them. A deep fringe (bowi's) is sometimes made to fasten along the lower margin of the yoke, at the bottom of the dress, and around the short sleeves. This is fabricated from a piece of buckskin, nine or more inches deep, slit into fine fringes so as to leave them connected by only a half-inch strip of skin. Beads may be strung on these. Similar strips of fringe are attached to a man's shirt. Hair that is cut off in m.ourning is tied in little rolls all over the dress; a man uses 389 Op. cit., 258, pl. 17. 1930] 211l 2,12 University of California Publication in Am. Arch. and Ethn. [Vol. 30 his in the same way to decorate his shirt. Sometimes bits of captured scalp are sewn in rows down each breast of a woman's dress.390 Two women's dresses of buckskin were observed in the collections of the Field Museum of Natural History. One of these (no. 61683) had a beaded yoke decoration that dipped downward sharply on the breast. This consisted of irregularly horizontal, parallel bands of small trade beads; the bands alter- nately black and white were about one inch wide. These were sewed directly to the body of the garment, there being no separate yoke-piece. The thread or sinew on which the beads were threaded passed through the buckskin only at each edge of the band, i.e., the rows of beads in each band formed taut loops an inch long. The shoulder seams were hidden by extra strips of skin. The fringes under the sleevelets and down each side of the body were twisted in pairs. Pendants on the front of the garment and on the right sleeve bore brass beads. The second dress (no. 61981) was "said to be over 100 years old." It was decorated with glass beads, thimbles, coins, and other trade articles. "The Chinese brass coins are dated in the periods of Shun-chi (1644-61), K'ang-hi (1662-1722), K' ien-lung (1736-95), and Kia-k' ing (1796-1820)." Pendants below the yoke bore green glass beads and brass thimbles; those near the bottom of the dress had strings of black glass beads with a central blue bead. The beaded yoke-decoration did not dip so markedly on the breast as that described above. The bands of beads were of the same width as those men- tioned above and sewed in the same fashion, except that a quarter-inch of bare space was left between the bands. These spaces were then painted red. A red band crossed horizontally below the yoke, again at the waist, and near the bottom. Each band of beads consisted of alternate rows of colored and white beads; in order from the top to bottom the bands bore colored beads as follows: red, green, black, (1), red, blue, blue, red, dark blue, red. The dress originally had two more bands below these but they have been replaced by rows of pendants. Excluding the second row (green) this looks like a sym- metrical color scheme; red, blue-black, red, blue, and reverse. Barrett observes that '.'the women also wore a cape or, more prop- erly speaking, a blanket, made of shredded tule or of sage-brush bark, or of a combination of the two." We may hazard the guess that the technique is open-twine like the mats.391 In the Field Museum collections are certain cloth bands (nos. 61985-86) bearing bead disks which are described as "dress ornaments" but which are probably ornaments attached to the middle of robes or blankets in Plains fashion. These are bands of red flannel, thirty inches long and about three wide. One bears four disks: one at each end, the other two equally spaced between. The second band has only two disks, both near one end but spaced like those of the first band. The beaded disks, four inches in diameter, were made up of spirally arranged rows of beads; the ground color was blue and white, with an outer row of yellow. It is noteworthy that each new colored 390 The general style of the women's dress is illustrated by a Klickitat specimen (Hodge, ed., Handbook of American Indians, 1:713). 391 Op. cit., 255. Spier: Klamath Ethnography row always began at the same radius, as in the manner of basket decoration. Pendants were attached to the periphery of each disk. These were formed of a continuous strand of black and white beads attached at intervals so as to hang in long loops. Basket hats are worn by both sexes. That of women is called k!Emui, of men, k!ane'o. Women and old men wear them at all times, younger women more than the elderly, and the younger men only when they are needed as protection against the pack strap. The small, snug-fitting hats are quite decorative with their contrasting designs, and are clearly thought so. Gala hats are studded with beads. They come halfway down the forehead, are nearly hemispherical yet flat on top. These are made in twined weave of various materials. Wom- en 's hats bear woven designs, men's are painted. Men's hats are said to have had short brims even prior to the coming of the whites. Widows must wear old hats blackened with pitch, if they wear any; otherwise they are open to ridicule. Men wear a cylindrical hat or headdress (tc'uyas) in war, ten or more inches high, and as low as the eyebrows. Made of cottonwood bark (wula'l), it is open at the top, but may be brought together in a point if made of skin; otter, coyote, beaver, skunk, or elk hide. Badger and raccoon skin is not used. These bear painted decorations: large zigzag lines around the hat; and several feathers are fastened in the top. Gatschet speaks of them as "wide brimmed. 13192 A cylindrical bark hat, said to have been worn by a chief, is in the Mield Museum of Natural History (no. 61725). This is about seven inches in diame- ter, nine high. The ring of bark is lapped and sewn with a running stitch. The decoration on the light surface is by a series of triangles along the top and bottom margins, with circles at intervals between these. The triangles have their bases on the margins; each consists of a black line within a red line; there are no base lines. The circular decorations consist of a black spot within a red ring. A pendant is fastened on each side at the top, consisting of a bunch of long hair wrapped at the proximal end with fur. Other hats are made of the scalp of deer or bear, including the ears, but not the face. Rawhide vizors, such as those used by other tribes are unknown, but Barrett mentions a similar sunshade of plain- twined tule.393 Hairdress is simple. Men wear the hair long, parted on the left side, with a braid lying in front of each shoulder or a single queue behind. The braids are wrapped with strips of otter fur. This is the style of the Plains and Plateau. Women and girls also wear their hair 392 1:92. He may have had in mind the basket hat. 393 Material Culture, 255, pl. 17. 1930] 213 2,14 University of Califonia Publications in Am. Arch. and Ethn. [Vol. 30 long and in two brEids, but parted in the middle. This again is reminiscent of the Plains. Their braids may be looped up to hang only to the shoulder; these are not fur wrapped. The hair is cut short in mourning, especially that of the women, daubed with black pitch, and generally neglected. A man who kills a rattlesnake ties the rattles to his hair; women do not do this. The hairbrush (slitci'llks) is a porcupine tail, stuffed with dry grass or sage-brush bark. A thong draws the ends together so that it dries in an arc. This thong fits over the back of the hand when the brush is in use. The open, underside of the tail is closed with a thong lacing. The ends of the quills are burned off evenly, so that only the stiffer body of the quill is left. Barrett notes that the longer quills are removed.394 My informant described the comb used by the Molala as a true double-edge comb. It is fashioned of a thin stick bent double. Little pointed sticks, as fine as our toothpicks, are inserted between the halves of this doubled stick, so that they protrude equally on each side, and are bound in place. The hair is greased; ordinarily with fish oil, but with deer tallow or that of skunk, otter, beaver, or mink, when they can get it. Frontal flattening, the only form practiced, is not universal among the Klamath, although it is general. My information differs at this point from Hrdlicka's who was told that non-deformed heads are derided as slave-like.395 I was told that it was merely "so that they would look pretty; they have nice round eyes because the forehead is flattened. "39ff A deformed head is called leple'pli nos, flattened head. The septum of the nose is pierced and the ear lobes, the latter twice or even more frequently. Both sexes insert dentaliium shells horizontally through the septum. These are placed so that the butt ends meet in the septum, the points of this curved conical shell pro- jecting on each side. If the septum hole is large, two may be placed side by side in it. Sometimes the point of one shell is thrust into the hollow butt of a second. The root of a mountain plant, called yayo'kalk, may be used in place of a dentaliuim shell. I was told that Molala women choose boldly curved dentalia, placing them so that the points 394 Op. cit., 258. 395 Hrdlicka, Head Deformation among the Klamath. 896 The informant who stated this may have rationalized her own case, an undeformed head. My other principal informants, two men, had flattened heads. Flattening is not noticeable among some of the older Indians, so that the state- ment that some escape the treatment may be true. Spier: Klamath Ethnography almost meet, forming a ring. Ear pendants are a group of four dentalia, hung in a bunch by their tips. Such an ornament. is called stosto'mnls. Neither dentalia nor any other plugs are put directly through the holes in the ear. The la.bret is unknown. Men pluck out their beards.897 Nose and ears are pierced in early childhood, even while still in the cradle. But it is usually done by the mother when the child is five or six years old. Anyone may perform the service; a man or woman usually does it for an own child. Coley's mother pierced his ears and nose when he was ten or twelve years old. He was afraid to have the septum pierced, but finally summoned courage. Piercing is done with a small sharp bone awl (so'kta); the lobe is massaged and the awl suddenly thrust through. A little charred sage-brush twig is put in the hole to keep it from closing and withdrawn when it is healed. There is no celebration of this event among the Klamath, although they are aware that this is the custom on the Columbia river. Holes are made solely for adornment; no esoteric significance attaches to the procedure. Fighting women sometimes tear out. their opponent's septa or ear holes. This is called p'eta', to tear open a hole. Necklaces are made of dentalia and other beads. Claws are not used with the exception of grizzly claws (steks) which are worn by shamans. The Molala make necklaces of yayo'kak root, mentioned above. This root has a length of eighteen* inches. It is split, lengths are tied together, and braided with sinews from the back of a deer. Three kinds of bead are in use; dentalia (t!u't!as), a small twisted shell, a half-inch long, called is'elka, and red stone beads (tcokE b'djE: tcok is the service berry) flattened like a kernel of corn. In addition there is an abundance of old style, glass trade beads on the cremation piles. Two types of dentalia may be recognized, but I am not certain; large shells called skals, and those up to three-quarters of an inch in length, t!u't!as. Dentalia are not marked with incised designs, as among the Yurok.3"8 They are, however, wound with human hair for a space at the small end, according to one informant, although another denied it. As was common on the lower Klamath river, they are st.rung end to end on a cord in lengths somewhat under a fathom: the fathom 397Gatschet, 1:90. 898 Kroeber, Handbook, 24. 1930] 215 216 University of California Publications in Am. Arch. and Ethn. [Vol.:30 length is "a little too much." The unit, according: to one informant, is the length of the extended left arm to the upper end of the humerus, or across the chest in addition to the point of the right shoulder. The other gave the unit as a doubled string from the extended left hand to the sternum; that is, a fathom. One thing is clear, that there is no great concern for dentalia as money, as among the lower Klamath tribes. The strings are not standardized either as to length or size of the individual shells. Long dentalia are more valuable than the short, but my informants never knew anyone to argue over the precise length of a shell. Both men agreed that the several kinds of beads, or at least the best, are not worked by the Klamath, but come from the south. They are either found in raided Achomawi-Atsugewi camps or are traded from these people or possibly the Shasta. Both were certain that the shells are not obtalaed from the north. This is some- thing of a puzzle, because the Columbia river people have dentalia and use them extensively in trade. Furthermore, the ultimate point of origin of the shells is the coast of Washington and British Columbia. The discrepancy in the statement is further accentuated by the Yurok statement that the shells came to them both along the coast and down the Klamath river.399 Finger rings are not made, nor bracelets, except for young chil- dren. The cast skin of a water snake is wrapped around their wrists so that they will swim well. Face painting is general, a daily usage, not confined to gala events, but more obvious at that time. The shaman's body is painted red during his winter performances. Red is the color in general use, white only at dances, and black alone in mourning. Those who go to war paint themselves red, not black, with their faces white. I have pointed out elsewhere400 that red paint is associated with warfare by the Basin tribes and their Californian and Arizonan neighbors. Here it may be added that black is used by Plains warriors. At dances white pigment, found along the creek banks, is smeared on the palms which are then pressed on the face or on a woman's smooth hair to form patterns. Red paint is difficult to get. It comes from a few localities in the mountains near Crater lake. The Northern Paiute are said to have quantities of it, however. In winter women smear their faces, hands, and arms with a mix- ture of charcoal and pitch, or deer grease, or of pitch and grease 399 Kroeber, Handbook, 22. 400 Havasupai Ethnography, 207. 1Spier: Klamath' Ethnography mixed, to prevent chapping and sunburn. Red paint mixed with this grease is similarly used. Putting on a pitch coating is called sutlu'mE. The pitch is roasted and later scraped with knives.40' Simple tattooing (sl'tcpale'as) is common to both sexes. It has no connection with puberty rites, although it is usually done in youth; nor any esoteric significance. Anyone may tattoo, even boys and girls tattoo their companions, but it is usually performed by women, who receive no payment. Charcoal is rubbed into incisions cut with fragments of stone. Women have the face tattooed, as well as breast and thighs. The face decoration is regularly a line down the center of the chin and one descending obliquely from each corner of the mouth. One Klamath woman has a line drawn horizontally from each corner to her cheeks. My female informant has a few random lines, including a poor cross, on the inside of her left forearm. A man's decoration is ordinarily confined to his arms. One man has a line running up the back of his hand and a T-figure on his wrist, each of its arms formed by several parallel lines. Another had short lines crossing the inside of his left arm just above the wrist. This is remi- niscent of the lines on a Yurok 's forearm to measure dentalia shells,402 but has not that purpose here. A few men have a line across the nose endinig in a circle beneath each eye, or the circles alone. This is the sum of face decoration; they lack the large black patches on chin and cheeks of the Shasta. Upland Takelma women, I was told, conform rather to the northwest California style. They tattoo the entire lower part of the face below a line drawn across the cheeks. TIME RECKONING, DIRECTIONS, AND NUMERATION As with most Indians whose ceremonials do not depend on the exact recording of time, the Klamath have little development of time reckoning. The result is the confusion and conflict between native authorities familiar elsewhere. Intervals of the day are of little importance. These are indicated by the position of the sun (sa"bas) for which a few names have been developed. pA'6ttk1s, early dawn; "this is the time to attack." ni'lka, sunrise. Embo'sant, early forenoon. 401 Gatschet, 1:150. He stated that pitch, "wakinsh grows on the pan-tree," by which I presume he meant the lodgepole pine. 402 Kroeber Handbook 25. 1930] 217 2,18 University of California Publications in Am. Arch. and Ethn. [Vol. 30 dino'l6pka, later forenoon. s6w6'tkE, midday. gawilo'la, mid-afternoon. di'ntkE sa"bas, sun nearly down. The movement of the stars gives time at night. For instance, I was told that in winter, when they sit late telling stories, it is bedtime (lo'laloks) when Orion's belt has risen two-thirds of the distance toward the zenith. This is about ten o'clock in mid-December. Mid- night is literally (?) pslndotklum. There is no year count, but these people have of course no difficulty in expressing spans of years. Similarly the seasons are of little sig- nificance. There are however four period names: sE'a'lum, the begin- ning of fall, lo'ldom, winter, when snow begins to fall in the moun- tains, sko'w'a, the beginning of spring, po6tda, when roots, etc., begin to ripen. It was emphasized that these names refer properly to the opening of the seasons. There is no generic word for seasons; the year is called momo'watc or e'lo'la. Gatschet observes that as an alter- native to the use of month names, they reckon time by the seasons in which natural products are harvested, as udsaksii'mi, "in the big sucker time," i-umii'mi, "in the berry season," etc.403 This seasonal count indicates that the year begins in early fall. It is marked by the end of the wild fruit season and the withering of the leaves. This corresponds to late August or early September.'04 The exact date is indefinite; it is marked by the appearance of a new moon, but bears no relation to the procession of the constellations. Ten months are named which purport to pass for lunar months. That is, the Klamath month (sa."bas, month, moon, sun) is a period indefinitely fixed in the solar calendar, marked by the waxing and waning of the moon, and omitting the dark phases. These people are fully aware tha-t ten named periods do not fit the twelve plus luna- tions of the year. No. attempt is made to make them fit, to reckon an intercalary period. The ten months are merely ten roughly marked intervals of the year-round, whose beginning, duration, and end are left indefinite as of no particular concern, but which serve the con- venient purpose of designating the several periods of the year. It is quite clear that while the Klamath are fully aware of the discrepan- 403 Gatschet, 1:76. 404 It must be remembered that winter comes very early at this altitude (4000 feet). By mid-September ice forms on the ponds during the night. Gat- schet's information confirms that of my best informants that August-September marks the opening of the year (1:74). This disposes of mid-winter and February- March as opening dates as mentioned by my other informants below. Spier: Klamath Ethnography cies of their calendar, they feel no need for correcting them. For instance, my most satisfactory informant, Nancy Phillips, named over the ten months when asked the number of lunations. When her atten- tion was called to the true number, she agreed that there were more than ten, but exactly how many she did not know, for no one keeps track of them. She thought there were more than twelve, certainly she knows that there are more than ten. The months are named for the fingers and counted in their order commencing with the right thumb, according to Nancy Phillips, and passing to the thumb on the left hand. This brings it about that the five names are repeated. 1. tkIo'po, thumb (approximately August-September and March; the ground thaws and the grass grows a little under the snow). 2. sp6'lwis, forefinger (September-October, the leaves fall; April, geese and fish return). 3. d6tklu'meni, middle finger (October-November; May-June). 4. k! 6'ptc!aalim sinW'kstls, ring finger (lit., partner of the little finger; December-January, the coldest month, the lake freezes and timber cracks with frost; June, fruits are fully ripened). 5. k!6'ptc!a, little finger (February; July-August, midsummer). The moon, which was full on September 21, 1926, was the second month according to this informant. On a previous occasion she had given the months' beginning with the left thumb, and February- March as the opening of the year. This is at least consistent with her later naming of the months. Which hand one begins with is evidently immaterial. A check on Nancy's data is given by her statement that Orion's belt appears on the horizon at dawn in the first month; this would be about August. She also stated that'the winter solstice came in the fourth month, which is correct by her reckonings. But she added that the summer solstice was in the thumb month, which would be incorrect unless she began counting on the left hand, from the little finger, which stood for February-March. Pat Kane placed the beginning of the year shortly before Christ- mas, the lunation in which the winter solstice occurs. He began the count with the little finger of the left hand and continued with the right thumb. This agrees with the procedure recorded by Gatschet. A year of twelve months was recorded by Gatschet.405 His inform- ant noted the beginning of the year about August. He named the months first on one hand, followed with the other, in each case begin- ning at the thumb, and ga-ve as eleventh and twelfth months the thumb 405 1:74-77. 1930] 219 220 University of California Pubhcationm in Am. Arch. and Ethn. [Vol. 30 and index finger of the first hand again. Naturally. Gatsehet's twelve months and my ten do not. correspond; the following tabulation. shows that there is close agreement only in the earlier. part .of the year. Finger Gatschet's description Identifia Thumb Berry gathering; dry the Late Sep fruit; fnares foal Index Inaugurate winter house by,. October feast: stack. hay, Middle Leaves fall October- Ring Snows Novembi Small Snows heavily - Novemb Thumb Lake frozen; snows DecembE Index Much rain; dances in dance DecembE house Middle Kill large suckers March-) Ring Take large suckers; dig ipos April Small Dry suckers; leave home to April-Mt dig camas and ipos says al Thumb Store camas; gather wokas Mid-Aug Index Return from.wokas harvest; gather berries . .. Septen Late.Aup tembei 2ble as tember Nancy's dating August-September September-October -November er er-December er er-January kpril .ay; Gatschet ,bout May gust t.o end of rnber gust and Sep-. r October-November December-January' February March April May-June June .. July-August The I.ndiaus quarrel a.great deal over the month reekoning. One old man, who is so deaf that he cannot. be argueq with, so I was told, held that September, 1926,. was k!6`ptc!alum sin6'kstls, ring finger month, tha.t. is, either fourth or ninth. He was counting too fast, according to Nancy. If he meant the ninth month., he was in ..sub- stantial agreement with Pat Kane in putting the beginning of the year at the end of our calendar year. In this connection we note. that women do not reckon the intervals between menstruations, nor recognize their recurrence with the luna.- tions.406 A tale, in which the number of months. is fixed by a council of animals, follows.407 The animals each tried to make so many sticks to represent the number of months they wanted. One made ten sticks for a ten-month winter. Another objected that this was much too long: "If you make a winter of ten months, people will freeze and starve to death before it ends." So he broke the ten sticks that stood for winter. If he had left them there would! have beeA twenty months in a year. The solstices a-re recognized but their exact dates not noted. They are unnamed but the apparent movement is called lo'lduim (or gapgo'- pli), going away, and skowha' (or toka'lka), returning. 406 Knotted string records, such as are used on Puget sound, are not kept. 407 Told by Nancy Phillips, who knew it only imperfectly. Of. Gatschet, 1: 105. Spier: Klamatih Ethnography Both. sun and moon are caUedT sa"bas, but they are clearly rmcog- nized as separate entities. There is a_tale accounting for the confusion. When Coyote was the sun at one time, daylight came so early that people did not get enough sleep. He shouted' at them every morning, "It has been daylight a long time." The sunshine was too bright and hot. Coyote abused the people because he was sun. Finally they deposed him, putting the moon person in his place.408 The Klamath say that during an eclipse the Grizzly Bear devours the sun or moon, hence it is called lok slo'ki, grizzly eats. They gather and shout, calling moon's wife, the Frog, to use her' power on the bear. The Frog can be seen in the moon.Yo9 The. Klamath share the habit of most other Indians in distinguish- ing only the more obvious or isolated constella.tions. So far as I identify them these are Orion's belt, Gemini, Big Dipper, and Pleia- des. -It is notable that thes'e are all winter constellations. Perhaps the Milky Way is na.med; I neglected to enquire. Polaris is known t.o be stat.ionary; its name was unknown to my informant. The evening star is called pse'geknls; the morning -star has no significance. But I judge that other stars (k'tc!ol) are not signaled for naming; Cas- siopeia, for instance, is not recognized. The constellation sliuks (Orion's belt) comprises six. stars; the three of the belt and those pendant forming the sword. The Gemini are twins (wiwi'aak), boy and girl. When they barely appear on the eastern horizon in the evening, they look over the lake and it freezes (December). Later when they stand higher, it. indicates that spring is coming. Their bow rises a considerable interval before them, a bent arc of six stars, with the string supposedly indicated. As' nearly as I can discover this group includes the western part of the constellation Gemini, the stars :, r, N, M, and another directly above them in Auriga. Rabbit (kai) is another star following these. He looks over the lake, making it freeze. Possibly this is Altair or Fomalhaut, although this does not. well fit the time of its appearance: on the east- ern horizon before dawn in dotklIA'meni (May-June), somewhat south of east. The Pleiades, called Eandando'ksni, are a group of little chil- dren who announce the dawn when they first appear on the morning horizon. Ursa Major (ko'kiuks, the divers, i.e., loons) is a group of people in a row diving into the water one after the other, gambling to see who can dive farthest. 408 Told by Nancy Phillips. 409 For the appropriate tale see p. 141. 1930] 22;1 222 University of California Publcations in Am. Arch. and Ethn. [Vol. 30 The Klamath notion is that they name their directions from the winds. This seems unusual until we realize that this is the habit of the peoples to the north, e.g., the Wishram by my own information. It is also intelligible from the role played by personified winds throughout this area, Shasta to Puget sound. Seven of the eight compass points are named, omitting northeast, from which quarter there is probably no wind. lopi'tal, east. ye'watal, southeast. mo'atal, south. 4opa'stilskni, southwest. tEka'lamdal, west. teffi'nksdal, northwest (cf. te6kan, service berry region, lying to the south- west of Crater lake). ya'm6ksda, north. This is the order in which my informant, Nancy Phillips, gave the directions. She did not offer zenith and nadir, and, on direct ques- tioning, held that these were not directions. The numerals are as follows: 1, nas. 2, lap. 3, nnda'n. 4, wuni'p. 5, tu'nip. 6, n6'tsks6pt. 7, la'pks6pt. 8, da'nks6pt. 9, n6tsRA'6ks. 10, te'6wfinlp. 11, na'slal' E. 12, la'pial'E. 13, Endan'iallE. 14, wunIi'pial'E. 15, tu'nipial E. 16, nW'tsks6ptial E. 17, la'pks6tial E. 18, da'nks6ptial E. 19, n6tski'&esiall'z. 20, la'puini te'ew'un'ip. 21, la'puini te'6wuimp pe'6nas.410 26, la'puini te'6wfinp pe'6n6tsks6'pt. 30, Endani' te'6wumnlp. 40, wuinlpni te'6wfinlp. ; 47, wumnpni te'ewuinip pe'en lapks6'pt. 50, tunmpni te'6wu'mp. 410 By analogy with twenty-six, forty-seven, etc., this looks as though the word for one should be suffixed, but the form given is consistent with one hun- dred and one. Spier: Klamath Ethnography 60, n6tsks6pt d6'nkni te'ewiunip. 70, la'pks6pt d6'nkni te'ewiunip. 80, da'nks6pt d6'nkni te'6wfinip. 90, n6tslki'fks d6'nkni te'6wfinip. 98, n6tsRA'6ks d6'nkni te'6wfinip pe'en da'nks6pt. 100, te'6wfin1pni te'6wfinlp. 101, te'6winipni te'6wiinip pe'6nas. The numbers eleven to nineteen are abbreviated forms, the full form for sixteen, for instance, is n0`tsks6ptial'FE ha te'ewuinlp ont. Even without extended etymological analysis, which I am not pre- pared to make, the basis of this numeration is transparent. One to six, nine, and ten are individual stems; seven is based on two, and eight on three. This suggests a quinary and decimal basis. From twenty up the basic count is by tens. Intermediate integers are so many tens plus the units, as in English. The exception is such an integer as twenty-one which looks like "two tens plus." Why the form for six,ty and above differs from the lower round numbers, I do not know. These numbers are used in counting all classes of objects. Gat- schet remarks that classifiers "are appended to the numeral to indi- cate the shape or exterior of the objects counted. '41. The only other number I found was a score, the ten fingers and toes together, na'sat. There are no similar collective terms for the fingers of one hand or of both. Only a circumlocution can be used, like lugawa'wals, the fingers, or petc6`mnAlugawa'wals, the toes all together (petc is foot). Ordinarily the ordinal numbers are not used in counting the fingers; one does not usually say one finger, but specifies it by name. The toes are enumerated like the fingers.412 The following list of colors was recorded: tokto'kli, red. t I o'tlalwi, green. m6tcmz'tcli, blue. spa'lptc!i, yellow. k6p' kauli' brown. b6'cb6cli, black. ba'lbali, white. ge'lopci, gray (English, gray ?), means dirty. My informant said that the word for blue was in use before the coming of the whites; she denied that there was a single word for blue-green. This implies that the familiar blue-green confusion was lacking here. I did not test any of these people for their color sense with color-wools. 411 1:523. 412 For the finger names, see the list of months above. 223 1930] 224 University of California PubU.atiozsw in --Am. Arch. and Ethn. [Vol. 30 IV. THE KLAMATH AND WE-STRN CULTURE THE PLACE OF KLAMATH CULTURE The pla.ce of the Klamath in western culture has' been obscure and has, I think, been mistakenly' appraised. In this section I 'have gone beyond the immediate task of defining it in order to build out'on its northern border the synthesis of Californian cultures furnished by Kroeber. Any a.ttempt at placing'the Klamath is rendered difficult by the pauc-ity of information for large areas in the northwestern states. There is little enough on western Washington,' but even less for the region east of the Cascade range, and practically nothing for the whole of Oregon. The southern Plateau'is unknown save for the Nez Perce; its eastern section (Kutenai to Flathead) is another complete blank. Satisfa.ctorily complete accouints now available can be num- bered on the fingers: Gunther's' Klallam, Haeberlin and Gunther's southern Puget Sound groups, Olson's manuscript notes on the Quin- ault, 'another manuscript on the Wishram by Sapir and the present writer, Teit's slight but highly suggestive notes on the Middle'.Colum- bia Salish, Spinden's Nez Perce, and the valuable collection of 'his- torical references gathered by Lewis.' The situation is better'for the Fraser river section of the Plateau, thanks to Teit's several accounts of the interior 'Salish. The stock of information on the Basin is slender and that for northern California none too voluminous. Never- theless it seems possible to place the Klamath by a cautious use of these fragmentary data. The present section will attempt to summarize the situation, with particular reference to material culture. The sections that follow analyze the shamanistic complex, social organization, and the girls' rites in greater detail. The food habits of the Klamath, and in fact their general material culture, places them squarely with the Plateau peoples. They. are as much a river people as the Thompson, Lillooet, and Shuswap, although the river in their instannce is a minor stream (the Williamson river) and substituted in part by flooded marshes. As a river people, their primary foodstuff is fish; vegetal foods, water-lily seed,.roots, and the like, form a considerable part of their provender, yet a lesser one. 10Spier: Klamath Ethnography 2 This is undoubtedly true equally -of all the tribes lying in the Columbia drainage from the Rockies to. the Cascades. It is known to .be so of tribes as far apart as the Wishram and the Lower Kutenai,"8 and of others lying between, as the Middle Columbia Salish and the Nez Perce in their original condition. The Nez-Perce case ca.lls for explanation. The present culture of the Nez Perce is Plains-like. But there can be little doubt that much, if 'not the overwhelming part. of the Plains 'traits among them, dates only from the introduction of horses into the Sna.ke-'Columbia basin', sometime between 1750 and 1800. With the horse' eame not only horse gear and all the elements of Plains nomadism, but the opportunity for buffalo hunting and trading on an extensive scale in the Plains. Many of the Snake-Columbia tribes began to make periodic visits beyond the. Rockies from this time. Teit's account, of the Middle Columbia Salish makes it quite clear that prior to the coming of the horse, typical Plat;eau culture, practically devoid of Plains elements, pre- va.iled over the entire area. Even the picture of life among the Nez Perce and. their neighbors presented to Lewis and Clark as late as 1804-6 was far more that of the Plateau at large than it is in our own day. The earlier situation is largely obscured by the fa.ct that the Yakima. and Klickitat in their eastern Washington loca-tion have many Plains-like features. Teit's investiga.tions make.it reasonably certainy however, that the situation prior to the end of the eighteenth century was quite different. At that time eastern Washington to south of the Columbia river was occupied by Salish tribes with typical Plateau culture. Subsequently the Yakima.-Klickitat moved northward, prob- ably from east central Oregon, to their historic habitat.414 It seems reasonable to assume that up to that time they, like the Nez Perce, had little in the way of Plains culture. Even today the food habits of the Sahaptins are not those of the Pla.ins. What I am suggesting here is that the balance of: vegetal products and fish, now nearly even, rather fa.vored fish in the days before hunting was so readily followed. So far as general habits of life go, the river culture of the Klamath links them a.lso with the' tribes on the lower rea.ches of the Klamath river. But their food habits do not quite coincide. Yurok, Hupa, and Shasta all ma.ke a larger use of acorns than of fish, although observers 413 Chamberlain, Kootenay, 183. 414 Teit, The Middle Columbia Salish; a map showing the early distribution is included in Haeberlin, et a., Coiled Ba.sketry in British Columbia. 1930]. 225 226 University of California PubUcatioms in Am. Arch. and Ethn. [Vol. 30 all agree that salmon form a far larger share of their dietary than in central California.415 Bulbs, roots, and seeds are eaten but seem- ingly to a less extent than among the Klamath. Deer and other large game are abundant, but curiously little used; an attitude toward this source of supply not only like that of the Klamath but exactly that of typical Northwest Coast tribes. Some of these northern tribes, in fact, will not touch venison. This non-usage is really not so strange when we recall that a plentiful supply of fish or fish and acorns renders the pursuit of deer unnecessary, and that the dense rain- forests and steep slopes of both areas make it unprofitable. To judge by the many references of Klamath informants to the habits of the Modoc, the latter must share much the same primary dependence on fish. It is just possible that roots, bulbs, and seeds play a slightly larger part in their economy, if the somewhat drier terrain they occupy has any influence. This is apparently the case so far as the Pit river tribes are concerned. Vegetal products seem to bulk larger than fish, although salmon is taken in abundance in the western sections, other fish in the east. Large game is abundant, but the Pit Rivers are not great hunters. Their general status is indicated in that, despite the availability of salmon, the western groups made much of the acorn. Southward the Maidu also place chief reliance on the acorn in the face of the considerable quantities of salmon taken and the abundance of other vegetal products and game.416 This is the general situation throughout central California, primarily a region of bulb, root, and seed gathering, of large and small game and insects, yet where acorns are the staple. It is of course obvious that the utilization of one or another food source will vary as they are available. Yet arbitrary habits and pref- erences enter. There is no reason why any foodstuff should have been completely excluded from the diet, as the Kwakiutl avoid venison, nor why a particular one should be exploited to the great exclusion of all others, as the acorn in most of California. From this point of view, the Klamath leaning toward fish as their staple is arbitrary and not dictated by geographic conditions. Their food habits are such primarily because they share the habits of the Columbia-Snake tribes as a whole. 415 Kroeber, Handbook, 84; Goddard, Life of the Hupa, 21-27; Dixon, Shasta, 423, 424. 416 Kniffen, Achomnawi Geography, 300-2; Dixon, Notes on the Achomawi, 211-13; Northern Maidu, 181 f. Spier: Klamath Ethnography Their material culture at large follows the general alignment of their basic economy with the Plateau peoples. . This can be made clear by the following t.abulation of ma.terial culture traits shared by tribes in adjacent culture areas. A few items of social culture not discussed elsewhere are also entered. Only those traits for which comparative data could be readily found are included, which makes the list random rather than truly representative. It must be understood that the reference is only to the specific form in which the Klamath hold to the trait. Thus, the compound fire-drill stick is entered, and not the simple fire drill. In fact, traits of this sort, having a distribution over the whole continent, have been excluded as adding nothing to our 1KLAMATH TRAITS SHARED BY TRIBES IN ADJACENT AREAS Klamath form Earth-lodge Mat-lodge Dome-shaped Earth-covered Dip nets Dam; weir Two-prong spear Cross-handle; curved point Slight use Extensive use Stone Shovel nose Considerable use Twine Minor use of coil Bags Woven Links with: Plateau (Chilcotin to Middle Columbia and Wishram), N.E. California (where it merges into central Californian forms). Wishram, Nez Perce, Middle Columbia? N.E. California, Basin, Plateau, Southwest rancherias, Plains, and eastward. Washington, Oregon (merges with central Californian type). California, Plateau, Northwest Coast. N. California, Northwest Coast, Plateau. California, Northwest Coast, Plateau. Northwest Coast, Plateau? Oregon, s. Plateau, Basin, Cali- fornia. California (late in N. Califor- nia?), Basin, Plateau?, South- west and southward. S. and central California (pre- historic in N. California); wood- en mortars in Plateau. Strait of Juan de Fuca and the Columbia basin to N. Cali- fornia. California, Basin, Plateau. Northwest Coast s. into N. California. Plateau, Basin, central and s. California. Made east of Cascades, s. to Klamath. Houses Sweat-lodges Fishing Digging stick Insect foods Metate Mortar Canoe Baskets 1930] 22.7 228 University of California Publicatios in Am. Arch. and Ethn. [Vol. 30 Mats Tanning Fire drill Spoon Bow Arrows Armor Shield Dress Hairdress Combs Head deformation Tattooing Klamath form Extensive use Mat needle Smoke-tan Compound Paddle Sinew-backed Flat; constricted grip Stone heads Tunic Slat or rod Round Fringed skirt Men: Plains shirt Women: Plains dress Decorated band on robe Tule slippers Three-piece moccasin Basket hats Braids for both sexes Porcupine tail Fronto-occipital Women: chin lines Links with: Oregonr?, Northwest Coast to Columbia r., Plateau. s. Northwest Coast (and north- ward?), Washington, Oregon? Modoc, Shasta, Achomawi, Wish- ram, Nez Perce; Basin, Pla- teau, Plains (and elsewhere?- probably not Northwest Coast). Puget sound, Takelma, Modoc, Washo, Lemhi, Wind R. Sho- shoni, Crow, Navaho, Hopi. California s. to N. Miwok. Arctic, Northwest Coast, Pla- teau, Washington, Oregon, Cali- fornia s. to Tehachapi, Basin, Southwest, w. Plains. N. w. California, Northwest Coast. (Not the bone heads of North- west Coast and w. Washington, where stone heads are scarce.) Northwest Coast, Plateau, Ore- gon, N. California, Basin, South- west (and southward?). Alaska to Maidu (Tlingit, Pla- teau, Oregon, Washington, N. California). Plateau, Basin, Plains (not Cali- fornia). Coast area west of Cascades- Sierras, probably Basin. Plateau, Basin, Southwest, Plains. Plateau, Basin, Plains. Plateau? (Wishram, Nez Perce), Plains. Modoc, Achomawi, Atsugewi, Paviotso. Plateau (moccasins are also N. and central California, Basin, Southwest, Plains, and east- ward). N. California (=Plateau type?; other types, Northwest Coast, Basin, s. California). Plateau, Plains. Achomawi, Northern Maidu. s. Northwest Coast to Alsea, Siuslaw, and Klamath. w. Arizona, California, Takelma (Oregon in general?). (Not Pla- teau? and Northwest Coast.) 0Spier: Klamathl. EtMnography Dentalium shells Cradles Games Pipes Musical. instruments Numerals Ritual number Calendar Scalping Head trophy Scalp dance War dance Disposal of dead Klamath form Ornament Sitting Board Beaver teeth dice Four stick Hand game Modified hoop-pole Ring-and-pin; rush ball Spherical; discoidal Angular Hand drum Deerhoof rattle in girls' rites Notched rasp Clap stick Open flute Finger names Five Numerical names Whole scalp Unusual With notched rattle Cremation Link-s -with: Ultimate source, coasts of Wash- ington and British Columbia. N.w. and N. central California. N. Basin, Plateau. Oregon,, Washington, British Columbia. N.E., central, and s. California (and elsewhere?). Some central Californians, Ba- sin, s. Northwest Coast, Pla- teau, Plains. Minor game of coast from Alaska to Oregon (not N.W. California). Modoc?, Wishram. s. Plateau (Nez Perc6 to Okan- agon). Wasco, Wishram, Middle Colum- bia, Nez Perc6 (links with Plains calumet?). Recent in Plateau; Basin, Plains, (Northwest Coast?). N. California. Plateau, Basin, Southwest, Plains. N. and central California. California, Southwest; Takelma, Nez Perc6 (and probably gen- eral in the west). Takelma. s. Northwest Coast, Washing- ton, Oregon, California s. to Sinkyone and Northern Maidu, Basin. N.W. California, Fraser r., N. Northwest Coast. California, w. Southwest, Wish- ram (no scalping in N.W. Cali- fornia E. to Shasta, and North- west Coast). California, Northwest Coast. General (except N. w. Califor- nia) (over head trophies on Northwest Coast). Upland Takelma?, Wasco, Wish- ram, Nez Perc6; without rattle, Chinook, N.W. California. Modoc; some cremation in N.E. California; central California; formerly in E. Washington; N. Plateau. 1930] 229- 230 University of California PubliUations in Am. Arch. and Ethn. [Vol. 30 understanding of the place of the Klamath in western culture. Specific tribal reference is made in only a few instances; principally where it is necessary to define the distribution in the neighborhood of the Klamath somewhat more closely than the names of culture areas allow. It does not seem desirable to cite specific references beyond this, since my notes are far from exhaustive. By far the majority of linkages are with the Plateau, particularly with its southern section. Northeastern and northwestern California are close seconds, but the general impression is that when details of specific form are emphasized, the relationship with the mountain tribes of northeastern California is much closer. The southern Northwest Coast, that is, largely western Washington, is a bad third, and, as is readily intelligible, central California and the Basin are culturally even more remote. Southern California and the Pla.ins have only a third as many elements in common with the Klamath as have the Plateau peoples, and the Southwest very few. The total impression might be considerably modified if we knew more of Oregon cultures, but I do not believe this would change the verdict of basic resem- blance to the Plateau. House forms are Plateau, both earth-lodge and mat-lodge. Mat- covered lodges are widely used, but the specific type here, resembling the earth-lodge structure, is a localized type of the southern Plateau. Plains dress and hairdress have come in through the same region; the older fringed skirt is typical of the whole area west of the Rockies or a.t least west of the Cascades and Sierras. Of personal adornment, the basket hat and tattooing are Californian, frontal flattening is from the Northwest Coast by way of the Columbia. Fishing habits are those of the larger northwest: northern California, the Northwest Coast, and the Plateau as a whole. The shovel-nose canoe is char- acteristic of the southern half of the same area. The preponderant use of basketry and the almost. complete absence of wooden utensils affiliate with California., the Basin, and Plateau, rather than with Northwest Coast. However the prevailing technique, twining, is that of the northern coast. The manifold use of mats again seems char- acteristic of the north. Cradles show a nice combination of the basketry sitt.ing type of northwesterly California and the wooden fiat forms of the Plateau. Arms a.nd armor are northern: the round shield is a Plains feature, probably transferred across the Plateau. Things less grossly material, such as games, affiliate in all direc- tions. The Californian four-stick game is balanced by the east.ern hand Spier: Kilamath Ethnography game; bea.ver-teeth dice are northwestern. Rattles are California'n, the hand drum a.nd notched rasp, eastern. The ritual number five is known in a swathe of territory from Vancouver island to the Basin. The war or incitement dance links the Columbia river with north- western California. More important is the practice of cremation which is partly followed in northeastern California and was, according to archaeological evidence, the earlier practice of the southern Plateau. This links the Plateau, northeastern and central California in one vast area which formerly at least was given to cremation. Specifically local elements beyond those. already noted, that is, those of northeastern California and the southern Plateau, are pipes of several forms, tule slippers and three-piece moccasins, perhaps the bundle form of ring-and-pin game, and the porcupine-tail comb; none of them more than superficial details. There are a number of features characteristic of nea.r-by culture areas, which in their distributions are quite within the range of influences impinging on the Klamath, and yet in which the Klamath do not share. Some of these are enumerated below. It must be understood tha.t no attempt has been made to list a.ll the items char- acteristic of each of the adjacent areas. The list includes only those features whose known distribution is close to the Klamath and which might have been expected among them. It is rather surprising to find that the several methods of deer hunting, the seed beater, rabbit-skin blankets, and tubular pipes are not present among the Klamath. All would fit well into Klamath cul- ture and their habitat imposes no difficulties. On the other hand, fish- ing stages are not needed on their relatively quiet streams choked -with rocks and with low banks. Pulverizing of salmon may be absent because comparatively few salmon are taken, and possibly other dried fish do not crumble like the flakes of dried salmon. Cocoons from which rattles can be made probably do not occur in Klamath territory. Why the Klamath do not beat seeds directly from the bushes is not clear; they ma.ke considerable use of them but prefer to gather the plants whole and shake out the seeds. Similarly the absence of tubular pipes and the tanning frame is inexplicable. Parfleches are not manu- factured but are traded from the Columbia. This explains the absence of their manufacture, for the tribes with whom they come in contact do not. make them either. This is certain for the Wishram and undoubtedly so for Wasco; the Klickitat and Yakima make them but seldom ;417 all trading from the Nez Perce and their neighbors. 417 Haeberlin et al., Coiled Basketry, 357, 360. 1930] 231 232 University of California PubUication in Am. Arch. and Ethn. [Vol. 30 The fashioning of utensils from wood may well have been inhibited by the development of basketry substitutes and the recency of direct conta.ct with a woodworking culture on the Columbia. Other features are perhaps too distant to suggest their imitation: the plank house and blanket weaving of the Northwest Coast, the tipi, FEATURES OF ADJACENT AREAS NoT SHARED BY THE KLAMATH Houses, plank tipi Dance-sweat-house Fishing stages Fashioned salmon club Pulverized salmon Hunting, impounding deer snares pits rabbit nets Seed beater Canoe sails Rabbit-skin blankets Blanket weaving Parfleche Wood carving Ladles and spoons Tanning frame Toothed comb Trough cradle Cocoon rattle Tubular pipe Tobacco cultivated Names, inherited and received at puberty Coastal from N. Alaska to N.W. California (e.g., Shasta, Takelma, and Wishram). s. (and N?) Plateau, N. and E. Basin, Plains, Macken- zie, N. Eastern Woodlands. Central California. Northwest Coait to N.W. California, Plateau s. to Columbia river. Northwest Coast to Columbia river. Plateau s. to Wishram (not Nez Perc6) (links with Plains pemmican and Californian pounded-meat?). Nez Perc6, Plateau?, Basin, N. Plains, Mackenzie? N.W. California, Takelma (Northwest Coast for mountain goat). N.E. California (and elsewhere?). Nez Perc6, N.E. and s. California, Basin, w. Southwest. California, Arizona, Basin, occasional in Plateau. Northwest Coast (aboriginal?). California, Southwest, Basin, Oregon (Modoc, Wish- ram: not Klamath and Nez Perc6), w. Plateau, Mackenzie. Northwest Coast to Columbia river, w. Plateau. s. Plateau, N. and E. Basin, E. Southwest, Plains. Northwest Coast to Columbia river. Northwest Coast to Columbia river. Wishram, Plateau, Plains. Quinault, Twana, Nez Perc6, Molala, Northern Maidu. Northwest Coast s. to Tillamook (links with N.W. California sitting type?). N.E. and central California. Takelma, Wasco, Sahaptins at Celilo, Nez Perc6; California, Southwest. Yurok, Hupa, Wintun, Maidu, Takelma, Wishram?; Northwest Coast, s. California, Southwest and southward, Plains and eastward. Northwest Coast, Plateau s. to Columbia river. even in its mat-covered form on the Pla.teau, the dance-sweat-lodge of central California. Yet even here some influence of these may be seen among the Klamath. Riven planks for earth-lodge timbers were derived, from Shastan and lower Klamath river forms of the plank house. The central Californian dance lodge did indeed reach the Klamath as part of the Ghost dance of 1870, but too late to influence their culture before its disintegration. Spier: Klama&th Ethrnography Taken as a whole Klamath material culture is of Plateau type with strong leanings toward northeastern California. But their social and religious culture has a different aspect. While basically of the extremely simple Basin-Plateau type, social habits reflect the North- west Coast, religious ones rather those of northern California and western Oregon. The basis of Klamath religious experience is that of the northern interior, the Basin, and the Northwest Coast, in distinction to Cali- fornia a.nd the Southwest. The experience is deliberately sought and the power obtained is of a generalized, unspecialized na.ture. In this they resemble as well the generality of North Americans. Even their spirits are of the sort universal east of the Pacific coast. Specific Northwest Coast influence appears in the circumscription of general spirit repossession to the midwinter month. On the other hand, the shaman novice's ina.ugural dance, which gives at least as much color to Klamath shamanism, is characteristic of northern California and western Oregon. The Klamath form of it is, rather, Oregonian. Details of shamanism, on the other hand, link more firmly with northern, especially northeastern, California. Social organization is extremely simple, and it is quite patent that as late as the end of the eighteenth century, it resembled that of the Basin and Plateau in every particular. Even now social stratification is only weakly developed. Chiefs were only recently set off from the mass, their functions vague and their influence slight. At the very moment tha.t Klamath society began t.o break down with the coming of the whites, they first imported from the Columbia river judicial func- tions for the chiefs. Chiefs are the wealthiest individuals, but. they are not chiefs because they are wealthy. Their selection is still based on principles of personal ability. The basis of this differentiation, wealth, came from the same direction, but the accompanying principle of the Northwest Coast, caste stratification, never reached here; nor did the chief's adjunct, a speaker or advocate, although this func- tionary is known in the Plateau and Basin as well as in the Northwest. At the other end of the scale, slavery is apparently an old institution, but was of little importance until wealth began to figure among them. The Northwest Coast concept of status based on wealth appears aga.in in marriage, here regularly by purchase. But social estimation did not rest on the amount paid, which was in fa-ct very little. With so slight an emphasis on rank and wealth, those northwestern corollaries, local exogamy to mainta.in the one and the family compact following 1930] 233 234 University of Californi Publication,s in, Am. Arch. and Ethn. [Vol. 30 on the other, were entirely absent among the Klamath. Nevertheless the mechanism of the Northwest Coast for the exchange of presents, the potlatch, does a.ppear here in the form of prescribed reciprocal visits between the families of bride and groom. It has not been clearly recognized that these features of Northwest Coast social life have' spread even farther from that focus into the northern Plains and to north central California. Of the slender stock of Klamath ceremonials, the scalp dance is widely known, the war or incitement dance is commoni to northwestern California and perhaps all of Oregon. Kroeber has pointed out that so far as California is concerned scalp and war dances are mutually exclusive in their distribut.ions.418 But the Klamath, like all others of eastern Oregon, carry out both dances. The girls' adolescence rite of the Klamath is one with those of western Oregon and northern California. This is a loca.l phrasing of similar rites known from Alaska to the Southwest, but lacking east of the Sierra-Cascades. Specific details of the Klama.th affair are closer to northern Cali- fornian forms than to those of Oregon. The Ghost dance of 1870 was evanescent among them. Its first form was that characteristic of' the original dance of Nevada; la-ter it was overlaid by features char- a.cteristic of central California.41' Some historical relief emerges in this survey. It is definitely traceable in some details and can be established inferentially for more basic elements. The most recent stratum of adopted traits is post- reserva.tion, i.e., later than 1865. This includes two types of basketry derived from the Northern Paiute settled on the reservation, a sifting basket and a coiled water bucket, and the Ghost dance. The earlier form of the Ghost dance was instituted by the same group of Paiute in 1870 or '71. This was the familiar circle of dancers, whose leaders drea.med new regulations; its ideology the return of the dead. In 1873 a Pit River introduced central Californian elementg; a dance lodge, probably the clap-stick rattle, new songs, etc. To the same stratum belongs the imitation of the Warm Springs-Yakima one-piece moccasin. Other elements are recognized as pertaining to pre-reservation days, but probably not earlier than the middle of the last century: the hand drum derived from the north, and perhaps the social dances in which it is used; the hand ga.me brought by a, Spokane ('?) man. 418 Kroeber, Handbook, 144, 844. 419 Spier, Ghost Dance. Spier: Klamath Ethnography The judicial function of chiefs, dispensing fines and punishments, is known to have been derived from the Columbia river tribes at this time. Some trade articles not manufactured by the Klamath also per- tain to this era: parfleches, skin bags, and buffalo hides from Warm Springs and the Dalles, flat basketry bowls from Pit river through the Modoc, and beads, even dentalia, from the same southern source or from the Shasta. The second quarter of the last century saw the coming of the horse. Ogden remarked on only one among them in 1826, but they began to be obtained in numbers, probably from Warm Springs, about 1840. There can be little question that with the horse ca-me elements of Pla.ins culture from the half-Plains tribes of the Columbia-Sna.ke, the adoption of predatory habits, and the great stimulus to slave- raiding and to sla.ve-trading with the people at the Dalles on the Columbia. Both tradition and inference assign Plains traits to this period. The relative newness of Plains style of dress is attested by the incom- pleteness with which it was adopted. It was never the universal garb, but confined to the wealthiest, and, most significantly, proscribed at the performances of sha.mans and the Ghost dance. Further, men did not wear Plains costume to war and set out on their raids afoot, even after the acquisition of horses, which suggests that the new accouterment and habits had not taken firm root. The circular shield, which also was never in general use, may well have come in at this time. The earlier types of Plains garments here were only simply decorated or not at all. It is known that the elaborate beading of the yoke and the use of scalps, etc., for appendages on women's dress, and the pa.inted designs on men's shirts, are much more recent than the first a.doption of these garments. These local changes probably reflect a similar transformation in decoration in the southern Plateau. Possibly the dome-shaped swea.t-lodge was introduced at this early period. It seems to have been displacing the small earth-covered affair for some time. But the latter is also generically the eastern type, not the Californian, and is without doubt older than this incur- sion of Pla.ins elements in the early nineteenth century. Finally, I cannot avoid the feeling that the sense of tribal solidarity, in which, even though but slightly developed, the Klamath differ from their Californian neighbors, dates from this contact with Plains-like peoples. 1930] 235 236 University of California Publications in Am. Arch. and Ethn. [Vol. 30 At least one minor element, assigned by tradition to this period, was derived from another direction; the notched rattle used in war and scalp dances in imitation of the Upland Takelma. Direct conta.ct was opened with the Upper Chinook at the Dalles only after 1800. By this time Snake raids had dispossessed the Sahaptins of central eastern Oregon, driven them north of the Columbia, and cleared the south bank of that stream by scattering Cayuse-Molala east and west. Direct contact quite across Oregon brought the Klamath in touch with the great trading center at the Dalles and revealed to them the superior a.nd stimula.ting cultures of the Ch.inookan tribes and the semi-Plains groups from above on the Columbia. The whole rise of the chiefs, the growth of social stratification based on wealth, the concern with slave-holding, all seem to date from this time. Wealth began to be considerable: its form was new: horses and slaves. Chiefs grew in wealth and prominence over shamans. The basis of their selection changed in some degree from war to wealth. Slave-raiding to the south increased in frequency: the booty was carried to Warm Springs and the Dalles. It is doubtful if the material differences in wealth and slave- holding existed prior to 1800. Shamans were dominant over chiefs and were but slightly more wealthy than the majority. The historic tribal subdivisions were in existence, the village groups and their dis- tribution established, and the basic social relations, as reflected in kin- ship terminology for example, much as we know them. Marriage may well have been by purchase, but possibly without the pseudo-potlatch custom of reciprocal visiting. Material culture was prevailingly southern Plateau in aspect, and religious practices, while more north Californian, had already some impress of the Northwest Coast. Is there a, Lutuamian culture province? I am doubtful. Kroeber has somewhat hesita.tingly posited one, set off by a certain distinctive- ness from that phrasing of Northw,est Coast culture found localized on the lower Klamath river, and favoring rather the Basin and Plateau, to which the Klamath country lies open. "The Lutuamian culture or subculture, as represented by the Klamath and Modoc, corresponds well with this setting. It reveals some specializations, such as its wokas and tule industries, that are obviously founded on peculiar environment. There are some north- western influences, but rather vague ones. The basis of the culture is perhaps central Californian, with certainly some Great Basin or 1930] Spier: Klamath Ethnography Platea.u admixture. Since the introduction of the horse, the Lutua.mi mode of life has evidently been modified analogously to that of the Plateau peoples of the Columbia., although less profoundly; and with the horse came a number of cultural elements from the Plateau, or even from the Plains; of which some went on to the Shasta and Achomawi. This recent modification appears to have given Lutuami culture a more un-Californian aspect than it originally possessed. Neither the Kalapuya nor the Klamath-Modoc were a numerous enough people to have possessed a truly distinctive civilization. The Kalapuya are gone, but nearly a thousand Lutuami remain, and as soon as their society and religion are seriously inquired into, their precise cultural affiliations will no doubt become clear."420 The fidelity of this characteriz.ation to the situation as it is now known is considerable, but the scantiness of information has permitted the supposition that the local culture is sufficiently distinctive to be given the dignity of a separate subprovince. The impression is that Kroeber has been led astray by the paucity and false balance of the materials on which he had to rely. Barrett's Material Culture of the Klamath does indeed present a picture of a culture in which the use of tule is accentuated. The predominant role given to wokas, the water-lily seed, is a readily intelligible infer- ence from Coville's detailed description of the specialized technique of its preparation. The brief and disjointed texts recorded by Gatschet allow only a puzzling insight into Klamath religious life. At the same time the true character of the southern Plateau, or at least the picture of several generations ago, is obscured by the half- Plains-like culture established for the Nez Perce by Spinden. Klamath technology is emphatically not one based on tule, with the limitations or suggestions of form it imposes. Nothing is clearer than that tule was in almost every case a substitute for more desirable material. Basketry techniques and forms are one with those of the Basin and northern California, and, we might hazard, the Plateau, despite the frequent use of this substitute material. Barrett's paper, which allows the inference of a developed tule industry basic in their material culture, is strictly speakiing not an exposition of their mate- rial culture at all, but a description of a collection of objects, which in his day had already degenerated to the use of convenient substitute material. What further inference may rest on the Klamath occupa- tion of a tule marsh is false. Klamath culture is not a marsh culture, but one of a river people. The norm of Klamath material culture is 420 Kroeber, California Culture Provinces, 163; Handbook, 334-35. 1930] 237 238 University of California Putblioation8 in Am. Arch. and Ethn. [Vol. 30 seen in their river home, not in the marsh which occupies a distinctly secondary place in their life. Correspondingly, while the place of wokas and the intricacies of its preparation are impressive, it must be recognized that their staple food is fish. There can be no doubt of this. They are far more a fish- using people than those on the lower Klama.th. And in this they link without question with the Plateau people. The use of wokas does indeed distinguish them from others of the area, but we recognize that this is no more than the utilization of a local food source readily avail- able and abundant, quite in harmony with the tendency in the whole area to emphasize one or another seed supply as the environment dictates. It must now be apparent that Gatschet's textual material presents the peculiarities of Kla.math phrasing of things religious without adequate explanation of their oneness with the Plateau and north- eastern California. There is little that he gives besides the songs, but their abundance and their nature can be duplicated among the Wish- rani and, I believe, were they known, a.mong the Nez Perce and Achomawi. I am not in a position to say how far their mythology is peculiar. Teit's paper on the Middle Columbia Salish makes it clear that Plains elements in this area are not older than the end of the eighteenth century. Prior to that southern Plateau characteristics prevailed with fair uniformity among the Upper Chinook of the Dalles, the Interior Salish of eastern Washington, and, we infer, the Waiilatpuan and Sahaptins of eastern Oregon and central Idaho. Such Plains cha.racteristics as the Klamath display are demonstrably late and they are themselves aware that much of Northwest Coast practice came to them only a.fter the opening of direct contact with the Columbia tribes barely a hundred years ago. The fina.l placing of Klamath culture is, perhaps, with the Modoc, as the southernmost outpost of Plateau life. Kroeber's supposition of a distinctive subprovince is excusable in the light of the misleading character of the available descriptions and the absence of Teit's epoch- making contribution. We do not yet know much of the southern Plateau but it does not seem likely. that this perspective will be seriously disturbed.42' 421 Now that a Lutuamian culture province fades away, we may hope to see an end to that barbarism, the word Lutuami, itself. No Kiamath or Modoc would recognize it, and should it prove that Klamath-Modoe, Molala-Oayuse, and Sahaptin form a single linguistic group, as has been suggested, we may hope to hear the last of it. 1]Spier: Klamath Ethnography THE SHAMANISTIC COMPLEX However simple and consistent a picture the religious practices of the Klamath may appear, there can be no doubt that they represent an amalgam of historically diverse elements. It is not necessary to indi- cate the exact sequence in which they were acereted in order to prove this. We can show that these elements have different distributions among neighboring tribes and that they enter into combinations differ- ent from that of the Klamath. The Klamath complex is best inter- preted as a function of their geographic position, without ignoring the fact that the peculiar form in which the Klamath ha.ve these ideas is duplicated nowhere else. The characteristics of Klamath religious practices and beliefs may be briefly recapitulated.. There a-re a host of weakly defined but sharply localized spirits from whom power is sought. These are chiefly birds and animals, some natural phenomena, and a few anthro- pomorphic supernaturals. The powers they bestow give success in all things difficult. Power is deliberately sought, it. rarely comes involun- tarily, and always according to a rigid formula. It is sought by almost every Klamath, perhaps by all, and the majority seem success- ful in their quest. Shamans are then set off from the laity only in the possession of a grea.ter number of powers. It is better to make the distinction by their acts: those who cure or practice weather control or clairvoyance, no matter how few their powers or the nature of their familiar spirits, are shamans. The quest is undertaken at any time of life, although usually at puberty. The most propitious times in later life are at periods of stress, as in distress or on the death of an imme- diate relative. The formula prescribes fasting for a number of nights on the mountains, continually running about and piling up rocks, combined with diving into lonely pools, or following the latter plan alone by diving beneath whirlpools in the river or lakes which spirits are known to haunt. The suppliant loses consciousness and wakes to find himself on the shore bleeding profusely from the mouth and nose. He does not ordinarily have a vision, but success is manifested in a song heard in a dream on a night soon following. The dream experience is essential. So closely is the spirit identified with the song, that but a single word indicates both. The song is the embodiment of the spirit, which hardly exists apart from it. The spirit is not the suppliant 's guardian and later quests are not necessarily for renewed 1930] 239 240 University of California Publications in Am. Arch. and Ethn. [Vol. 30 experiences with it. Power ma.y also come involuntarily, when a dreamer hears a spirit song. There is no obligation to a.ccept, but resistance brings illness. The midwinter month (December-January) has a curious place in the Kla.math calendar. While a shaman must of necessity evoke his spirit. to cure or foretell at any time of the year, this is the period when every shaman feels called on to exhibit his powers. The novice shaman also waits for this time to give a public test of his possession, and only after this does he attempt to cure. These seances are solely for displaying power, manifested by the songs and a series of feats. During these performances, those among the audience who have spirits also sing their own songs if the spirit impels. In other words, the midwinter month is the period par excellence when spirits regularly resume contact with their communicants. The essential of the winter seance is a dance by the shaman as close a.s possible to a. huge fire in order to call his spirits to assist him in the feats that follow. These are stereotyped tricks, of which two, swallowing fire and arrowheads, are the test of the novice. Other tricks are of the type of making things appear and disa.ppear. A few other details should be mentioned here. There is an assistant at this performance, a layman, who repeats the shaman's words. Food brought to the seance is thought first. to be eaten by a frog spirit, and until she ha.s eaten no one may touch it. The performance takes place in the shaman's own dwelling. Weather control, like the display of powers at the midwinter seances, offers an opportunity for shamanistic rivalry. Clairvoyance appears in connection with the war dance preparatory to an expedi- tion, and, curiously, the shaman accompanies the party to continue his clairvoyant function, and, what is still more unique, to serve as surgeon. Illness is the result. of the intrusion of something foreign, usually a spirit and commonly the spirit of a malignant shaman. A cure con- sists in extracting a tiny object which is either the disease itself or embodies it. The curing procedure is highly formalized: spirit after spirit is invoked to search the patient, beginning always with the Old Frog Woman, until the culpable spirit is found. The shaman 's familiars are thought t.o stand beside him ready to assist. When he succeeds in sucking out the disease object he swallows it, while blood streams from his mouth. Very little of this is unique with the Klamath, as we may see by comparison with the sha.manistic complex of neighboring peoples. ]Spier: Klamath Ethnography What is unique, however, is the peculiar conjunction of the elements affected by them. Even the closely related Modoc differ, to judge by what little we know of them. By way of orientation it is desirable to survey the situation north and south.422 Among the Kwakiutl the guardian spirit is inextricably linked with the hereditary caste system. Rights to a particular guardian are fixed by inheritance, so that the spirits function essentially as badges of clan rank and secret society affiliation. Hence they are few in number. Apart from this, they have definite mythological character, unlike the bird and animal spirits and the few anthropomorphic supernaturals of the Kla.mat.h. Acquisition of a guardian is not the relatively random quest. of the Klamath, but is definitely fixed with relation to the midwinter ceremonial. 'the novice, a lad, is sent at this period under the tutelage of an old woman in quest of the heredi- tary guardian. He returns to the village so strongly imbued with power a.s to be of danger to himself and others, hence the essential of the ensuing performance is to reduce his ecstasy to a normal state. A potlatch is offered on his behalf. Such individuals, men or women, may become sha.mans in this manner; only the dances they perform on their return apparently differ from the ancestral dance forms. In a Kwakiutl tale, a man wandering in the woods is initiated by the animals. On his return his overwhelming power is found dangerous and is dulled by contamination with menstruants.423 During the mid- winter sacred period, spirit repossession is general and at the cere- monies all a-re impelled to display their hereditary spirit songs and dances. The grouping at this time of those having the same spirits is of a piece with the grouping by clans during the remainder of the year. Both groupings rest essentially on names acquired in the one instance from spirits, in the other by clan inheritance, names through- out giving rank and privilege.424 The Kwakiutl form of quest, its relation to the midwinter cere- monies, and its social significance are peculiarities which do not pre- vail south of Vancouver island. Various elements of this complex, however, are found in the area to the south, such as the secret societies as far as the Quina.ult on the Washington coast, and the winter dance 422 See also the summaries of Benedict, The Concept of the Guardian Spirit, 10 f. 423 Boas, Ethnology of the Kwakiutl, 733-749. 424 Boas, Social Organization and Secret Societies, 393 f., 418 f., 501. 1930] 24.1 242 University of California Publications in Am. Arch. and Ethn. [Vol. 30 period even to the Klamath. But this interrelation of spirit quest and secret society ritual, with the social grouping that rests on it, is lack- ing. The Kwakiutl must be looked on as a highly specialized form of the spirit quest-shamanistic complex, of which the more normal forms, in the sense that they are less differentiated and widely dis- tributed, are found in western Washington and the interior. As an example of the more common form, I will present that of the Thompson. The Thompson complex425 is more nearly that of the Klamath but is even simpler. The vision quest, which takes place at puberty, is essentially the prerogative of boys, although girls may obtain some slight powers. Hence shamans are men and but few women. The vision is sought in the mountains where the boy remains praying, dancing, running about, and fasting. His training does not end with the vision, but continues after power has been acquired, the whole being directed toward magical training for adult life. In fact the powers bestowed by the spirits are specifically for particular pro- fessions and arts. The spirits are legion: natural phenomena, animals, birds, and a random collection of objects. The seeker sometimes receives power from the same spirit as his father and shamans some- times directly inherit their powers. It does not appear that the shaman novice establishes himself by a performance of fixed kind, nor do the shamanistic seances, though they include feats like those of the Klamath, occur only in midwinter. Shamanism in the area to the east of the Thompson must be of this type, and in the Basin southeast it is clearly so. South of the Klamath the shamanistic complex takes quite another turn. In California power is not ordinarily sought, hence the blessing descends on the few, the shamans, and for the mass there is nothing. This is an about-face from the situation in Oregon and northward, or in the Plains. In other words, power is a thing that comes involun- tarily. Further, the form of the complex has in the northern half of California two quite distinct phrasings. In northwestern California the person destined to be a shaman is visited in a dream by a spirit or ancestor who inserts the power in his body. This power or "pain" is animate, has personality, and is conceptualized as a small spindle-shaped object. Such "pains" are disease objects, yet their possession by shamans gives the power to remove other malignant "pains" from a patient. Accordingly the shaman-to-be is made ill by the "pain" he has received, until he is 425 Teit, The Thompson Indians, 317 f. Spier: Klamath Ethnography restored by an established shaman and learns to control the gift. The dance in which this takes place is, as Kroeber remarks, essentially a professional initiation.426 For example, the Yurok novice dreams of a dead person putting a " pain" in her body. (In northwestern California, shamans are most frequently women.) A dance is made for her, when under the direc- tion of established shamans, she withdraws the "pain" and having thus shown control of it, swallows it again. After a brief dance in the mountains, this rite is repea.ted and followed by one around a hot fire to further gain control of her "pains. 2427 The northwest California complex has its ramifications as far north as the Wishram on the Columbia and southeastward to the Southern Maidu. The novice dance is found over this whole area, although in the north being rather a demonstration of power than an occasion for acquiring control of it. The concept of the "pains' as embodiments of power or agents of disease extends eastward across the state. Yet the general aspects of shamanism in northeastern Cali- fornia are appreciably like those of the Klamath and Thompson. The second form of shamanistic complex is central Californian. Here individual shamanism is overshadowed by group activities of the Kuksu cult ceremonies which are shamanistic at bottom. Some dis- tinction is ma.de between the sucking shaman, who cures in the com- mon manner, and the outfit shaman, whose ability is, rather, clair- voyant. This distinction is also made to the northwest but is by no means impressive. The outfit shaman is usually the' heritor of the necessary paraphernalia from a parent or brother who instructs him in their use. His power clearly lies in the objects themselves. On the other hand, the sucking shaman obtains his by a visita.tion in a dream. Not only are central Californian shamans overshadowed by Kuksu impersonators, to whom has accrued all elaborate display, but the shamanistic complex proper is strongly infiltrated from the northwest. There is in California at large some further' specialization among shamans beyond those who cure. Weather-control shamans seem most clearly differentiated in south-central California, according to Kroe- ber. The Klamath fall in with the majority of peoples beyond this focus in looking on weather control as only a somewhat specialized function of any shama.n. Similarly with rattlesnake shamans, who are general in California. save in the northwest. Bear shamans, found in all but peripheral California, are not known to the Klamath. 426 Handbook, 852. 427 Op. cit., 63. 1930] 243 244 Uaniversity of California Publications in Am. Arch. and Ethn. [Vol. 30 From this perspective the Klamath complex takes its place with the relatively simple shamanism of the northwestern interior and the Basin, overlaid with elements shared by tribes of the northwestern coast and those of northern, or better, northwestern California. Shamanism bulks large in the social scheme of the Klamath 'as it does in the int.erior Plateau and the Basin, for the reason that com- peting rituals are lacking. Beyond the girls' rite, the war and scalp dances, and for a brief time the Ghost dance, the Klamath have noth- ing but shamans' performances. The winter ritual of the Northwest Coast extends only to central coastal Washington, the potlatch only to the Columbia river. The wea.lth display dances of northwestern California do not penetrate even to the Shasta; those of the central Californian Kuksu cult not beyond the head of the Sacramento. Ele- ments of these ritual systems do indeed appear among the Klamath, but the dances as such a-re absent. The Klamath hold the acquisition of supernatural power within the reach of anyone. In this they accord with the generality of North Americans and differ emphatically from the Ca.lifornians for whom power descends only on the few. The Klamath form of quest fol- lowed by a dream in which power comes links the common quest of America with the dream experience of California. The Northwest Coast, its interior, and the Basin are a unit with the Klamath in that the'quest is for power of a general sort, not for visions yielding spe- cific sanctions as in the Plains. The spirits whose benevolence is sought are birds and animals as in the interior, the Plains and eastward, not the mystic supernaturals of the Northwest Coast. The source of power and disease among the northwest Californians, the "pain" objects, do not a.ppear among them in any form. The Klamath con- cept of disease is the very general North American one' of intrusion; the soul-loss concept of the Northwest Coast and tribes to the Colum- bia is unknown here. The Klamath novitiate dance for shamans is shared by groups from the Sacramento to the Columbia. It is the northern demonstration of ability rather than the initiation and mastery of power of the northern Californians. The repeated dance of the Yurok prevails through Shasta and Modoc,'but the single Klamath dance is of the interior type. The Northwest Coast winter ritual ma.kes its appearance here only in its general aspects. Klamath shamanistic seances are held only in midwinter and this is also the time for repossession by spirits among 10pier: Klamath Ethnography the people at large as well as the shamans. Inasmuch as both these features are northwestern and non-Californian, they must be looked on as influences from the north. The following sections attempt to substantiate these generalizations. Spirits Those spirits of which the Klamath seek power are conceived as animals, birds, reptiles, fish, sometimes as natural phenomena, and in a few instances as supernaturals, anthropomorphic and akin to the characters of their mythology. Their number is indefinitely large. Each may be said to constitute a genus of spirits; e.g., "the Coyote spirit" is properly a group of coyote spirits. Members of the class are not singled out for individualization. The spirits of other tribes of the northwestern sector are of the same type. All the interior tribes on into California and all others of Oregon for whom we have information hold to this conception: Thompson, Lillooet, Wishram, Nez Perce, Takelma, Klamath, Modoc, Achomawi, Atsugewi, Maidu, and Wintun. There are undoubtedly differences among them in the type of spirit most frequently sought; thus, natural phenomena bulk larger in the Thompson mind. Further, objects, some as trivial as the nipple of a gun, figure as spirits among the Thompson, Lillooet, Snohomish, Nisqually, Nez Perce, Achoma.wi, and Atsugewi.42 A concept of a different sort holds sway on the coast from Van- couver island to the Columbia river. There the prineipal spirits are supernaturals of mythological character. These are necessarily restricted in numbers since they are highly individualized, and cor- relatively the powers they bestow are particularized. Where informa- tion is available, these crea.tures are credited, as a.mong the Klallam, with a social existence of human pattern: they live in houses, are provided with servants, and travel in canoes. This is the predominat- ing spirit concept among the Kwakiutl, Klallam, Snohomish, Nis- qually, and Quinault. The slight information on Chinook spirits sug- gests supernaturals of the same sort. This is not to deny that animals and birds figure among the Klallam and Puget Sound Salish, and that among the Quinault there a-re a few bird spirits that reta.in their 428 Teit, The Thompson Indians, 354; The Lillooet, 283; Spier and Sapir, Wishram Ethnography; Spinden, Nez Perce, 247 f.; Sapir, Religious Ideas of the Takelma, 35; Meacham, Wi-ne-ma, 130 (skins and feathers of spirits); Dixon, Notes on the Achomawi and Atsugewi, 218; Angulo, Religious Feeling, 356. 19303 245 246 University of California -Publications in Am. Arch. and Ethn. [Vol. 30 avian characteristics. The really unique group may -be the Kwakiutl, for there supernatura.ls alone appear.*2- A second specialized concept is found in northern California among Shasta, Yuki, and Maidu, at least. Shasta guaardians are described as human in form,.but "rather shorter than the ordinary st.ature." The Yuki "personal spirits are small, like boys, with gray hair." One of the Klamath guardians, which every shaman appar- ently has, is probably related to these. It is Tcakia'k, "little boy," which frequents the water, is no bigger than a baby, and has hair falling to its waist. A Maidu may have various animal spirits, but the truly important ones are in a.ppearance like people. To the Pomo there appears "a being, human and speaking one's own language but a stranger." Similarly some among Achomawi and Atsugewi spirits are "disease persons" or spirits in human form.430 These con- cepts fall in with the animal and bird spirits. For although these supernaturals are human, they are not particularized like those of the Northwest Coast and are not restricted in numbers: on the con- trary, they constitute an apparently large, undifferentiated class. The feeling cannot be avoided that these manikin spirits are related in some way to the beings that appear to, others of northern California. The Yurok dreams of a dead person, usually, if not always, a. shaman; the Sinkyone sometimes of a dead relative. Yokuts spirits have quite another phrasing, but we should not overlook that among them are dead relatives.43' These cases are rather obviously related to the inheritance of power by a visitation from a dead relative, but there seems also a continuum from the Shasta. manikin to the Sinkyone ghost. Kroeber has pointed out that there is a Yurok-Hupa substitute for the familiar spirit in the so-called "pains." This will be dis- cussed below. A note may be appended on certain Klamath spirits. Mention was made above of Tcakia'k, a water dwarf. Dwarfs figure commonly enough among the mythological concepts of western Indians, but this one is specific: a dwarf living in or about the water. Gifford has 429 Boas, Social Organization and Secret Societies of the Kwakiutl, 394 f.; Gunther, Kilallam Ethnography, 291 f.; Haeberlin-Gunther, Ethnographische Noti- zen, 58 f.; Olson, Quinault field notes; Boas, Chinook Texts, 220, 230. 430 Dixon, Shasta, 470, Northern Maidu, 265, Notes on the Achomawi and Atsugewi, 218; Kroeber, Handbook, 198; Freeland, Pomo Doctors and Poisoners, 63, 58. The spirits that help Quileute shamans restore souls are usually dwarfs (Frachtenberg, Eschatology, 335-7). 431 Kroeber, Handbook, 63, 149, 514. Spier: Klamath Ethnography remarked their presence in the mythology of various Shoshonean groups (Western Mono, Northern Paiute of central Oregon, Northern Shoshoni, and Serrano), pointing out their absence in typical central Californian groups.432. But they are not confined to Shoshoneans: the Yurok believe in dwarf-like spirits that. haunt. overgrown spots in the creeks, the Hupa speak of little beings resembling men in the water in remote places, and the foothill group of Northern Ma.idu of one- legged dwarfs found in or about water.433 So far as these data go, the concept is largely north Californian and Basin in distribution. Klamath Dwarf-Old-Woman, who figures in one form as a rock on Mount Pitt, may be identifiable with Takelma Rock Old Woman, associa.ted with a rock north of Rogue river.434 Inheritance of Spirit Power On the Northwest Coast proper there is a, tendency to fixed inheri- tance of experiences with ancestral guardians. Among the Tlingit, e.g., there is some tendency for men to be blessed by spirits according to their clan affilia.tions; the spirits descending from uncle to nephew. Among the Kwakiutl, inheritance is rigidly fixed, for a youth expects to find only those spirits that gave powers to his clan.433 Something of this prevails southward but in dilute form. A Quinault seeks the spirit his ancestor acquired, and we are told tha.t the guardians of the Chinook are hereditary, like t.he Kwakiutl. A Nootka may also inherit a spirit with whom an ancestor had an expe- rience.436 The southern Puget Sound Salish hold tha.t human social organization is reflected among the spirits; that a powerful spirit would appear only to a man of high rank. While this does not define the system as one of strict inheritance, it must approximate this because both the number of persons of high rank and of powerful spirits is small. Inheritance of a guardian and of the particular guardian thus prevails generally north of the Columbia river. Elsewhere inheritance is present but in less fixed form. There is for the Thompson seeker 432 Gifford, Western Mono Myths, 304; Marsden, 183; Lowie, Northern Sho- shone, 234; Notes on Shoshonean Ethnography, 297. 433 Kroeber, Handbook, 67; Goddard, Life of the Elupa, 64; Dixon, Northern Maidu, 265. 434 Sapir, Religious Ideas of the Takelma, 45. 435 Swanton, 465; Boas, Secret Societies and Social Organization of the Kwakiutl, 393, 418. 436 Boas, op. cit., 393; Sapir, Vancouver Island Indians, 593. 1930] 247 248 University of California Publications in Am. Arch. and Ethn. [Vol. 30 the possibility of being visited by the spirit his father had met. Inheri- tance is apparently common among the Nez Perce; at least shamans inherited their status and one could hardly become a shaman save through inheritance. As the number of spirits that give shamanistic powers is small, it seems probable that the particular spirit is fre- quently inherited, although it is not so stated. Comparable to this, anyone among the southern Puget Sound Salish might involuntarily receive a blessing from a spirit acquired earlier by an ancestor. For instance, if one's forebear had been a great shaman there was a pre- disposition to acquire a shamanistic spirit. Rather different from these cases is the case of the Klallam shaman who, dying, may will some of his spirits to a younger man, yet it reflects the tendency toward transmission in this area.437 Inheritance occurs again in northern California, where children of shamans are ra.ther automatically visited by their fathers' spirits. (It must be borne in mind that in this region acquisition of powers sets off shamans from the laity, who do not ordinarily acquire them.) It is intelligible why inheritance does not appear in northwestern California, where a shaman is made, not by the visitation of spirits, but by the insertion of "pains" by established members of the craft. The distribution of inheritance is in the interior: Klamath, Shasta, Chimariko, Pomo, Southwestern Wintun, Maidu, and Paviotso. Inher- itance does not appear again, so far as present data show, in all of central a;nd southern California until the Akwa'ala438 are reached. Inheritance of position is by no means the only way one becomes a shaman even among these north Californian peoples, and the visita- tion by spirits of one's father is quite limited among them. In other words, there is here little of the rigidity of inheritance known in the northwestern sector. Shasta and Northeastern Maidu show the closest approximation to the northwest; if the parent is a shaman, the child becomes one, and the particular spirits are heriditary guardians of the family for generations. Among Northeastern Maidu the succession is inevitable, among Shasta there is some latitude. The Klamath situation resem- bles the Shasta: a shaman's child is the involuntary recipient of his father's spirits, which he may or may not accept. There is nothing to show that the same spirits continue with successive generations.439 437 Haeberlin-Gunther, Ethnographische Notizen, 56; Teit, The Thompson Indians, 320; Gunther, Kilallam Ethnography, 297; Spinden, Nez Perce, 247, 256. 438 Gifford-Lowie, Notes on the Akwa'ala, 345. 439 Dixon, Northern Maidu, 274, 275; Shasta, 471, 477. Spier: Klamath Ethnography The Sacramento Valley Maidu hold that inheritance is not neces- sary, while the foothill group of the same people have a grea.ter ten- dency to inheritance. Among the Chimariko, shamans similarly may or may not inherit their positions. The Paviotso shaman's powers are sometimes inherited from father to son. In these cases heredity is simply one way of achieving shaman's sta.tus.440 In this second group of cases it seems that susceptibility to experi- ence with spirits is heritable, if not the inheritance of powers from specific spirits, but in a third group the inheritance is that of position and paraphernalia alone. That is, the individual Pomo, or at least the Northern and Eastern Pomo, and the Southeastern Wintun sha. man instructs his son, brother, or some other relative and delivers to him amulets and other objects necessary to him as successor. Among the Pomo this is the manner in which "outfit doctors," not "dreaming or sucking doctors, " are made.44' This affiliates with the general habit of instruction of shamans and secret society members in this quarter, where in general the hereditary element is absent. It seems quite uncertain whether the northern Californian cases as a whole are linked historically with the northwest. Possibly they are, since the hereditary element is lacking in more southerly Califor- nia. But as the data stand there is a hiatus of distribution in Oregon. On the other hand, there are few Oregonian dat.a save for the non- hereditary Takelma. It seems significant, however, that the Wishram did not mention inheritance in this connection to either Dr. Sapir or myself. The Power Quest In a discussion of the vision in the Plains, Benedict points out that there is there a distinction between visions (quests in my termi- nology) from which sanctions are received, such as the success of a projected war party, and those by which a guardian is acquired. The former are of more frequent occurrence in the life of each individ- ual.442 Among the Klamath every quest that is successful brings power that abides, not a sanction for a particular project. Every quest corresponds to the specific guardian quest of the Plains, though a guardian in the sense of a familiar is not acquired. So far as 440 Dixon, Northern Maidu, 267, 269; Chimariko, 303; Lowie, Notes on Sho- shonean Ethnography, 294. 441 Freeland, Pomo Doctors, 59; Loeb, Pomo Folkways, 320; Kroeber, Hand- book, 361. 442 The Vision in Plains Culture, 12 f. 1930] 249 250 University of California Publications in Am. Arch. and Ethn. [Vol. 30 guardians or familiars can be spoken of, they pertain to shamans alone, since they alone conjure up their familiars to their aid. But theoretically, every Klamath who has had an experience could call on his spirit for aid. The Klamath concept apparently holds for all tribes of the Pacific coast. There does not appear anywhere the Plains type of quest for particular sanctiops. The spirits of the experiences are not guardians, they a-re not invoked for assistance (sa.ve by shamans), and they give continuing powers to perform the arts and professions or for success and wealth. Accordingly, shamans are differentiated as a class more by their acts and the posesssion of powers to perform the shamanistic arts, than by the unique possession of helpers. This statement does not apply, however, to California, for there this is precisely the difference between shamans and laity. The Klamath deliberately seek power by self-isolation and morti- fying a.cts. Their attitude is that it is to be had by a voluntary quest, save in those rare instances which may be described as inherit-ance and in occasional involuntary dreams. The terms quest, vision, and drea.m have commonly been used as though they were fully interchangeable. It is preferable to distin- guish the voluntary quest and the involuntary acquisition of power; to confine the term vision to sensory experiences including seeing and hearing a spirit, and the dream to its everyday significance. These are, to be sure, by no means mutually exclusive. The quest and the involuntary dream as sources of power have different distributions in western America. The quest is general over eastern North America and in the Plains. But as Kroeber has pointed out for California and Benedict for the Southwest the deliberate quest is absent and its place is usually taken, so far as individual acquisi- tion of power is concerned, by the dream.443 I will attempt to define their respective areas of distribution somewhat more precisely. From this point of view a distinction must be made between the eastern and western sections of the Southwestern area. Personal powers figure only in a minor way among the Pueblos, where priestly craft functions in place of shamanism. Yet at Laguna and Cochiti, at least, there is some occasional questing for powers in isolated spots. 443 Kroeber, Religion of Indians of California, 327 f.; Benedict, The Concept of the Guardian Spirit, 36 f. Fig. 22. Distribution of the quest and dream as sources of spirit power in western North America. Squares indicate the quest; circles, the dream. 2.52 University of California Publications in Am. Arch. and Ethn. [Vol. 30 At the same time, among the'western Apa.che, who have shamanism full-fledged, those who do not become shamans by the instruction of .established practitioners, seek power in the same way.44' West of the Apache the dream is the source of power: this is known at least for Havasupai and Pima. Havasupai shamans alone have familiars, which they acquire involuntarily in dreams, and even when they inherit the power it seems necessary that they dream. Pima shamans become such generally by heredity, or by being bitten by a rattlesnake, but they may'also have persistent dreams in which they acquire power. Russell makes it clear that this involves no deliberate quest.'" Dreaming appears with the same function among the Shivwits, hence may have been general among other Southern Paiute. Prior to the introduction of the Sun Dance among the Southern Ute about 1900, shamans got their powers in dreams without fasting; since then they deliberately seek power in the Sun Dance."6 Dreams are the source of supernatural sanctions on the lower Colorado and in southern California alike, despite the difference of setting in the two areas. Mohave, Yuma, Akwa'ala, and Desert Ca.huilla shamans derive their powers by dreaming of the great mythological characters, sometimes under the stimulation of jimson- weed, but not-as part of adolescence rites. The Cahuilla, indeed, hold the boys' rite but it is reasonably clear that jimsonweed drinking at this, or at other times, does not furnish the setting for acquisition of power. On the other hand the boys' rites provide the'proper setting for a majority of southern Californian peoples: the narcotic of jim- son weed causing sensory disturbances which are interpreted as dreams of spirit beings, prevailingly animal in character. This is known for both Southern and Northern Dieguefio, Luisenio, Juanenio, Kawaiisu, and may be inferred for others. The Ka.waiisu, unlike their neighbors, administer jimsonweed regularly to girls as well as boys and with the same end in view. Inasmuch as the candidates for these rites do not present themselves but are selected, we have here a tom- mon basis of involuntary dreaming for the region as a whole. This is not to deny that the phrasing of the experience itself is markedly 444 Benedict, op. cit., 39; Goldfrank, Organization of Cochiti, 58; Bourke, Medicine Men, 453. 445Spier, Havasupai Ethnography, 276 f.; Russell, 257. 446 Lowie, Notes on Shoshonean Ethnography, 291. Spier: Klamath Ethnography different among Mohave-Yuma, Akwa'ala-Cahuilla, and the westerly southern Californians.447 The quest on the other hand is characteristic of the northwest. It is of course voluntary and the looked-for response probably always takes the form of a vision. The Klamath are a partial exception, which I will discuss later. . The quest has been noted among all tribes for whom there is information: KIamath, T'akelma., Alsea, Tillamook, Chinook, Wishram, Quinault, Twana, Nisqually, Snohomish, and Klallam. It is at the bottom of the initiation practices of the Kwakiutl and Bella Coola. In the interior, it is characteristic of Thompson, probably Lillooet, Nez Perce, Chilcotin, and Tahltan.448 As these data stand the areas of distribution are quite distinct. The voluntary quest is characteristic of the northwest and the Plateau, as well as the Plains; the dream, involunt.ary in its occurrence, of the western section of the Southwest, southern California, and the south- ern Basin. On the border common to the two areas they occur con- current.ly, na.mely in northern California, southern Oregon, and the northern Basin. In northern California the dream basis is clear for the Pomo and the Northern Maidu of the foothills, apart from Pomo shamans who are such by initiation alone.449 There is a suggestion that the Wailaki are one with them.450 All other groups have some kind of combination of dream and quest. Northeastern Maidu and Yokuts hold dream and quest equivalents; the Yokuts stressing the quest in contradistinction to the Maidgu. Sinkyone, Yuki, and the Sacramento valley group of Northern Maidu differ from them only in that their alternative to the deliberate quest is an involuntary waking experience.45' Of the Paviotso, Lowie remarks: "A shaman gets his song a.nd regalia through a vision in a dream. Annie says in the Walker River District there was a mountain where people went in quest of a vision. 447 Kroeber, Handbook, 604, 640, 775 f., 783, 793; Preliminary Sketch of the Mohave, 279; Gifford-Lowie, Akwa'ala, 344-5, 347, 348; Hiooper, Cahuilla, 334-5, 340; Spier, Southern Dieguefio, 312 f.; Waterman, Religious Practices, 296; Du Bois, Religion of Luisefio, 80. 448 Farrand, Alsea, 245; Boas, Tillamook, 5, Third Report, 10, Chinook Texts, 214; Eells, Indians of Washington Territory, 674; Farrand, Chilcotin, 19; Emmons, Tahltan, 112. 449 Freeland, Pomo Doctors, 58, 63, 64; Dixon, Northern Maidu, 269. 450 "The Yuki state that the Wailaki shamans often dreamed of a spirit coyote" (Kroeber, Handbook, 154). 451 Dixon, Northern Maidu, 267, 275, 278; Kroeber, Handbook, 149, 196-198, 514. 1930] 253 254 University of California Publications in Am. Arch. and Ethn. [Vol. 30 This may mean only that the quest is undertaken and that the vision appears in a dream at that time, but I interpret it as meaning that both dreaming without a quest and the deliberate quest occurred among them. His description of the Wind River Shoshoni ideas, is, I think, more convincing as to the simultaneous existence of dream- ing and the quest.452 Shaman candidates among Yurok, Hupa, and Shasta experience in dreams, but for the laity there are opportunities to acquire power of the quest type. "There are, however, [among the Shasta] one or two very simple ceremonials, whose purpose is merely to acquire luck. For example, on very dark rainy or windy nights, men go out alone, and naked, and walk all night, praying for luck, singing, seeing strange things pass them, and having strange voices calling. Or on very cold nights in the middle of winter, a man will go out just before dawn, and after praying for luck, will plunge into the river, and swim." "The [Hupa] men used to swim a certain course in the icy water [of Trinity river] and then lie on the frozen shore as long as possible that they might win great good luck." "A Yurok man who wishes to be brave and fierce . . . . goes at night to a lonely mountain pond, swims, and is then swallowed or taken below the sur- face by a monster. "453 The combination occurring among Klamath, Modoc, Achomawi- Atsugewi, Northeastern Maidu, and Chimariko is most interesting. The regular Klamath formula is that the seeker may have his quest rewarded by a vision, but the power is not obtained at this time. It comes some nights later, when in a dream he hears the spirit's song. I am impressed by the fact that the quest experience and the dream are separated. Indeed the seeker may not have a vision at all. Further, dreams of the same sort, conveying power, come to one unsought. The Klamath notions seem to me most intelligible when considered as the amalgamation of the two kinds of experience. Those of the North- eastern. Maidu are explicitly the same: a man on his lonely quest experiences the presence of spirits and on his return they appear to him in dreams. The Achomawi-Atsugewi case seems to be the same: "While he is bathing during the night, he may find something in the water, or may dream of it later when he takes a little sleep." Again the Chimariko seeker is visited in a series of dreams, which were "often the result of solitary visits to remote mountain lakes." Modoc 452 Lowie, Notes on Shoshonean Ethnography, 294-296. 4553Dixon, Shasta, 489; Goddard, Life of the Hupa, 88; Kroeber, Handbook, 40, 68. Spier: Klamath Ethnography ideas are appreciably close to those of the Shasta in that the novice has an involuntary seizure, apparently while at home, but following his inaugural dance, he goes to the mountains in quest of power.454 Corresponding to the absence of sharp distinctions between sha- man and layman in the northwest, the quest is open to everyone throughout that area. Yet there are some differences with respect to sex. This is measurable in part by the relative number of men and women shamans. Men are more frequently successful in acquiring power than women and are accordingly more often shamans among the Thompson, Klallam, Quinault, Snohomish, Nisqually, Wishram, and Klamath. The evidence is quite definite on this point. Among the Tillamook and Nez Perce both sexes obtained powers: it is conceivable that men predominated. The statement that Takelma shamans are equally men and women is perhaps significant in view of the situation to the south. Shasta, Yurok, and Wiyot shamans are chiefly women, with but few men. Goddard merely mentions both men and women shamans among the Hupa, and Dixon among the Chimariko, but Kroeber implies that they were normally women.455 Elsewhere in California it seems that men predominated as sha- mans; specifically Achomawi, Atsugewi, Northeastern Maidu and those of the foothills region, Yuki, and Salinan. In southern Cali- fornia, where the equivalent of the shaman is one who has had visions induced by jimsonweed, this situation holds even more stringently because the narcotic is rarely administered to girls. Yet in north- eastern California the northwestern emphasis on feminine shamanism is reflected. Dixon observes of the Northeastern Maidu: " There were women shamans as well as men; the former being, however, more numerous in Big Meadows, where the Maidu come nearest in contact with the Achomawi, among whom women shamans were very numer- ous. 456 Angulo457 states that their number was about equal among the Achomawi, which is not a contradiction of Dixon. This implies that the Achomawi-Atsugewi and Big Meadows Maidu were both influenced by Shasta custom. Elsewhere here (Yuki and Foothill Maidu) women were rarely shamans. 454 Dixon, Northern Maidu, 279; Notes on Achomawi and Atsugewi, 216; Chimariko, 303; Meacham, Wi-ne-ma, 129-136. 455 Goddard, Life and Culture of the Hupa, 66; Dixon, Chimariko, 303; Kroeber, Handbook, 111, 137. 456 Dixon, Northern Maidu, 274. 457 Angulo, Psychologie religieuse, 564. 255 1930] 256 University of California Pu-blication0s in Am. Arch. and Ethn. [Vol. 30 The Period of the Quest The Klamath deliberately seek power both in adolescence and in later life. In the discussion of the Plains vision referred to .above, Benedict points out that it is there an affair of maturity and not of adolescence.458 Everywhere in this area, on the other hand, the quest may ta.ke place in adolescence, at a time sometimes varying widely from actual puberty, and in the majority of tribes at any time in later life as well.. This is consonant with the difference in the result accruing from the experience. This is as much characteristic of Klallam, Sno- homish, Nisqually, and Modoc, for example, as it is of the Klamath. The absence of suitable specific data makes it difficult to judge how far this is true of others. At least among the Wishram and Achomawi-Atsugewi there is a difference; power is sought in youth alone and should the seeker fail then, he has no opportunity in later life. Descriptively the Kwakiutl align with the Wishram, for the novices who acquire spirits at the winter ceremonial of the secret societies are adolescents. Powers are also acquired later but the few references indicate that this is not common. The situation is different, however, for the Kwakiutl does not freely seek the spirit, but automatically acquires it at a time fixed by the round of a predetermined social system. There is a suggestion that the Thompson, Lillooet.. (who are described as identical with the former), and Nez Perce have their experiences only at adolescence. The Tilla.mook evidence is equivocal: from the very brief account we learn only that a boy of twenty is sent out. The situation is different in northern California because of the dream source of power in place of the quest. Nevertheless, this might have been restricted to adolescence, but that seems not to be the case. On the whole, data fail for the region, but our accounts read as though the shaman novices were always adult. To take the Shasta as an instance: it seems unlikely that the novices' experiences would occur to adolescent girls. Such at least is the impression, but we can- not be sure. The quest, open to the laity among Shasta, Yurok, and Hupa, is not restricted to any period in life. In southern California, where a shaman 's career is initiated in the jimsonweed rites, this is an affair of adolescence. 458 The Vision in Plains Culture, 3. Spier: Klamath Ethhography The Conduct of the Quest Basically the manner of seeking power is the same throughout this area. This is necessarily so because it means simply fasting and praying in lonely spots in the mountains, beside pools,. etc. There are more specific details having definite distributions. IFor example, the Klamath youth isolated on his quest runs con- stantly a-bout piling up rocks: so does the Modoc, Achomawi, Atsugewi, Wishram, and Nez Perce. This is a wholly arbitrary and pointless procedure, unless it be thought that any arduous task will heighten the seeker's immolation and thus hasten the benefaction. The Thompson have rationalized this type of activity (it does not include piling rocks specifically) as training for later life. The same point of view appears more widely in the comparable puberty rites for girls. Pos- sibly related to these rock piles is the seat or little monument to which the Yurok retiresAto acquire luck in gambling or power to bewitch.459 More significant than this, the Klamath dives into the water, where he is drawn below by a spirit. He loses consciousness to wake on the shore with blood streaming from nose and mouth. Diving for a spirit is a practice widespread north and south. The Wishram sometimes dives to reach certain under-water caves where spirits dwell. A Kwakiutl shaman's song tells of being carried under the sea where he received power. In several Chinook tales the hero swims, is carried under by a supernatural to its house, and wakes on the shore. Precisely the same formula is recorded in an experience of a Klallam; here too, blood flowed from the man 's mouth. The Snoho- mish and Nisqually youth regularly dives into whirlpools, carrying in his hands a rock on which he spat, to weigh himself down, and wakes on the shore. I recorded exactly this curious procedure in a tale of the Kalispel. Boas obt.ained a legend of the mainland Cowichan which suggests the same form: an orphan went swimming and diving every day in order to get strong. "One day he made a fire near a lake and accidentally spat. into the water." He lost consciousness, visited the supernaturals below, and obtained power.460 While I think 459 Dixon, Some Shamans, 24; Sapir, Wishram Texts, 187; Spinden, Nez Perce, 247; Kroeber, Handbook, 63, 64. 460 Boas, Ethnology of the Kwakiutl, 1294, Chinook Texts, 220, 230; Gunther, Klallam Ethnography, 294, 302; Haeberlin-Gunther, Ethnographische Notizen, 57-8; Boas, Lower Fraser River, 2. 1930] 257 258 University of California Publicationm in Am. Arch. and Ethn. [Vol. 30 that it is hazardous to use incidents incorporated in tales as evidence of customs, it is at least suggestive in the absence of other information. Southward of the Klamath the same practice of diving in moun- tain lakes, finding "something" in the water, and recovering on the shore is recorded for Achomawi, Atsugewi, and Northeastern Maidu. The case of the Northern Maidu of the Sacramento valley differs only slightly in phrasing: the seeker is rescued by his friends. Again, the Yurok swims and is taken below by a monster.46' In view of this distribution it seems to me significant that the Shasta and Hupa seeker is described as swimming for luck; the Hupa man, at least, then lying as long as possible on the frozen shore. The Chimariko is briefly described as bathing at dusk in remote mountain lakes, and of the Wiyot we learn only that some people were pitied by powerful lalke spirits. The significance of all these cases lies in the recognition of a class of under-water spirits from whom power is to be had, and further that it is attained by diving to reach them. If we assume that South- ern Maidu and Miwok shared the belief in under-water spirits (and our scant data do not deny it), the Yokuts case becomes intelligible as a southern outpost of the belief. These people sometimes derive their powers from animals or monsters inhabiting the water, and Kroeber, in fact, recorded the experience of one Yokuts who dreamed he was taken below the water by spirits.462 Not the least curious part of some of these experiences is the notion that the effect is so great as to cause hemorrhages at mouth and nose. This was referred to above among Klallam and Klamath as a result of diving. It is also the regular effect of other types of experience among the Klamath. The identical notion has been noted for a variety of other tribes of this area: Quinault, Modoc, Shasta, Sacramento Valley Maidu, Sinkyone, and Yuki.463 The Klamath are not at all reticent among themselves as to the nature of the experiences they have. Thus, one who has had a suc- cessful quest at once may sing what he hears in his dreams. On the other hand, a Wishram carefully guards his experience, betraying it only by songs at the spirit dances and by his personal taboos and 46lDixon, Notes on the Achomawi, 216, Northern Maidu, 267, 278; Angulo, Psychologie religieuse, 157-158; Kroeber, Handbook, 40, 68. 462 Dixon, Shasta, 489, Chimariko, 303; Goddard, Life and Culture of Hupa, 88; Kroeber, Handbook, 117, 514. 463 Olson, Quinault field notes; Meacham, Wi-ne-ma, 129;- Dixon, Shasta, 473, Northern Maidu, 267-268; Kroeber, Handbook, 149, 197. Spier: Klamath Ethnography behavior, and finally revealing it only at the point of death. These contrasting attitudes have different distributions. A number of Wash- ington tribes align with the Wishram: the Puget Sound Salish does not reveal his experience until late in life, the Klalla.m and Nez Perce waits until the spirit. dances are held, and there is a suggestion that the Chinook also maintains silence. Similar reticence crops up again in northeastern California: the Achomawi-Atsugewi never tells, the Sacramento Valley MIaidu only as an old man.464 Considering their geographic situation we might have expected similar reticence of the Klamath, but on the contrary they share the freedom of speech of the generality of Californians. The Novice's Performance It is obvious that there must always be an initial performance by a shaman to demonstrate his ability, such as a first cure. It does not follow, however, that this need be a performance of a special type, as it is among the Klamath. The Klamath novice waits for winter to prove his power and only after this does he attempt to cure. The dance he then performs is much like those he. will use in subsequent demonstrations of his power, but includes the demonstration of two feats; eating fire, and swal- lowing and disgorging arrowheads. He sings his spirit songs as he dances close to a huge fire, while an assistant calls on his spirits t.o help him perform the feats. It will be observed that this has nothing to do with curing. An initiatory performance is by no means confined t.o the Klamath. Kroeber has previously pointed out that this is characteristic of north- ern California, centering in the northwest.465 It has also a northward extension beyond the border of that state through western Oregon. Records are available for Wishram, Tillamook, Klamath, Modoc, Shasta, Tolowa, Yurok, Hupa, Chimariko, Sinkyone, Yuki, Northern Pomo, the northerly Wintun, the Sacramento valley branch of the Northern Maidu, the Northeastern Maidu, and the Southern Maidu. There is no reason for supposing that it was wholly lacking among other western Oregon and northwest Californian groups, for most of whom there is meager information. It is apparently absent among 464 Haeberlin-Gunther, Ethnographische Notizen, 66; Gunther, Klallam Eth- nography, 290; Spinden, Nez Perc6, 247 f.; Boas, Chinook Texts, 215; Dixoin, Some Shamans, 24, Northern Maidu, 267. 465 Handbook, 852. 1930] 259 260 University of Califormia Publications in Am. Arch. and Ethn. [Vol. 30 Achomawi, Atsugewi, more southerly Pomo, Miwok, and Yokuts, and in the north among Quinault, Klallam, and the Puget Sound Salish, which sets limits on the south and north. It appears again, however, among the Cahuilla. The dance has by no means everywhere the same function or form. For the Klamath and Shasta it is merely a demonstration of power; the same is implied for the Modoc and in our very brief note on the Tillamook.4Y6 The Wishram case belongs to this category in part: the novice demonstrates his power to cure, but there is this difference, that he must first be restored from his seizure. That is, from the Shasta northward the inaugural dance is primarily a demonstration. The inaugural dance has quite another purpose in northwestern California: it is to enable the novice to acquire control over his powers, in this case the rather material "pains," which correspond to the spirit powers of other tribes. The dance is, as Kroeber remarks, essentially a professional initiation. The novice is aided by qualified shamans to retain in his body and control the "pains," which up to this point make him ill as they do any layman. Prior to the dance, he receives instruction from older shamans. This type of performance is credited to Yurok, Tolowa, presumably Hupa (and Karok?), Chi- mariko, Sinkyone, Yuki, Northern Pomo, northerly Wintun, Sacra- mento Valley and Northeastern Maidu, and the Southern Maidu. There are then two types of inaugural dance, so far as function is concerned: the demonstration type of Oregon and the control type of northern California. Kroeber's designation of the dance as north- west Californian is not quite adequate in view of the Oregon data. From a wider point of view, the control dance of northern Cali- fornia and the Wishram case link with the southern part of the North- west Coast proper. A feature of repossession during the winter sacred period of the northern tribes is the overpowering strength of the spirit which makes its possessor "ill"Y until he is relieved by dancing. This is also described among a number of them as characteristic of the condition of the youth who has newly acquired a spirit. It is the essential point in the novice 's part in the winter ritual of the Kwakiutl and Quileute: he must be relieved of his superabundant power. Simi- larly, among the Klallam a boy who has received a strong power must remain away from home for some months until he has learned to con- trol or dull it.467 I do not know how widespread in western Washing- 466 Meacham, Wi-ne-ma, 136; Boas, Notes on the Tillamook, 10. 467 Gunther, Klallam Ethnography, 290. Spier: Klamath Ethnography2 ton this idea is, but it crops up again among the Wishram. A lad who acquires a spirit is "sick"; a shaman is employed to relieve and restore him. This is one of the essentials of a Wishram novice sha- man's inaugural dance. It must be noted that in all these instances the condition of the possessed is described in one way: he is "sick." This is precisely what is at the. bottom of the north Californian con- trol dances, although the sickness is due to an agent of another sort. It is altogether regrettable that there is no information on the coastal tribes of Washington and Oregon, since the similarity of the ideas suggests that there may have been a continuous distribution along the. coast. At least in the interior, the "sickness" of the repossessed does not hold beyond the Wishram. Klamath and Shasta have the form of the recovery ceremony without the substance. To turn to some comparisons of detail, the dance made by the Shasta immediately on the spirit seizure is not duplicated among the Klamath. When the Shasta novice revives, she dances and may be carried or swung over the fire. This is repeated on two nights follow- ing. At the last dance the spirit shoots a "pain" into her, which she masters, withdraws, and reinserts in her body. The later inaugural dance is, however, much like that of the Klamath. The novice waits until winter to demonstrate. This dance takes place in a dwelling just as among the Klamath. She dances and sings while an assistant calls on her spirits to help her.468 What is lacking to complete the analogy are the tricks of legerdemain, but these among the Klamath are the stock in trade of all shamanistic performances and appear to be unknown to Shasta practice. There is just a possibility that the Klamath arrow-swallowing trick is related to the Shasta novice 's demonstration of her ability to extract and replace the "pains," described as small clear objects pointed at each end. Another difference is that her dance is under the aegis of qualified shamans; a feature linking the Shasta with the typical north Californian performance. The Modoc novice dance which Meacham described469 is externally like that of the Shasta yet smacks of Klamath ideology. The spirit seizure takes place apparently at home. When this is manifested by hemorrhages, the novice is made to perform ten dances while an 468 Dixon, Shasta, 475. 469 Wi-ne-ma, 129-136. As the original is difficult of access, it is reprinted below in an appendix. Despite Meacham 's naivete and his obviously faulty knowledge of the Modoc tongue, as evidenced elsewhere in Wi-ne-ma, all his descriptions of Modoc life ring true. 1930] 261 262 University of California Publications in Am. Arch. and Ethn. [Vol. 30 established shaman calls various spirits until the tutelary is discovered. This is repeated for five nights. The novice then goes to the mountains to seek further power. On his return he dances during a month, aided by others, to prepare for his profession. This terminates in another ceremony under the shaman's direction. In the opposite direction the Wishram also have an inaugural dance for a shaman novice which, unlike these southern examples, occurs at any time he receives power. As in the Klamath affair, he performs in the characteristic manner of a shaman of his tribe, even demonstrating a cure. When he dances to call his spirits, others per- form to help him acquire strength. There is a difference, however; the lad is overcome by his power so that a shaman must dance to relieve him and help him regain control. While this is foreign to Klamath-Shasta thought, it is characteristically Wishram. There everyone on returning from an experience is overpowered by it and must be restored by a shaman. As Kroeber justifiably points out, the unique characteristics of the Californian control dances center among the Yurok and Hupa, and gradually diminish in number as far as the Maidu.470 It may be profitable to reexamine the data. There are two distinctive features of these complexes: the "pains" by which the novice is possessed and the dance for control. It is obvious that there is no inherent connection between this particular type of possession and such a dance. Among the Yurok, Hupa, and Shasta "pains" take the place of spirit familiars among the Klamath, for example. Among the tribes to the east and south (Achomawi, Atsugewi, the several Maidu groups, Wintun and Yuki) the shamans operate with spirits, but 'pains" appear as causes of disease. All these have the dance, which is far wider in its distribution than "pains" as sources of shamanistic power. The concept of the "pain" has interesting variations among the several tribes. It is everywhere animate, sometimes with personality, and while ordinarily immaterial, has sometimes a material form. When its material characteristics are described, it is commonly a clear double-pointed object, like an ob1sidian arrowpoint. Thus the Shasta "pain" is a small, clear double-pointed object; they are the same as spirits, yet each spirit possesses one. The Chimariko disease object is small, double-pointed, and animate; the Achomawi like a little 470 Kroeber, Handbook, 852. Spier: Klamath Ethnography needle of glass. The Wintun "pain" is a spirit missile, a flint or obsidian arrowpoint, and like its Yuki analogue, is dispatched to cause illness. The Yuki "pain" is an invisible obsidian point. The object which the Sacramento Valley Maidu shamans insert into the novice to see if he can get it out, is neither a disease object. nor the embodiment of power, yet is obviously related to these; an invisible double-pointed object as long as the finger. The disease objects of the Nort-hea-stern Maidu are not described in this fashion, but they have the animate, personal character of the "pains" elsewhere. Southern Maidu initiates are "killed" with a "piece of medicine" a half inch long. No description is available from Yurok and Hupa.47' It is unfort.unate that the term "pains" has crept into the litera- ture to describe things which, it seems to me, are not relat.ed to these concepts. Thus, the "pains" of the Takelma are neither substitutes for spirits nor have they personality. Of the Sinkyone, it is recorded: "Ordinary disease was cured by sucking out of the body the sinsing or material 'pains. ' " In neither case do these differ one whit from the disease objects of peoples outside this area. The case is somewhat different for Achomawi-Atsugewi: the "pain" is described as "a small bla.ck thing, like a bit of horse-hair" which a shaman sucks out in a cure, but "the 'pains' grow in size and strength by killing people. If the shaman does not catch the 'pain' when it returns to him after killing the person it has been shot into, he loses all control over it, and it goes about killing people of its own initiative. This is the cause of epidemics." This closely resembles the Northeastern Maidu concept.472 The inaugural dance of California is of considerable proportions only among Yurok, Hupa, Shasta, and Northeastern Maidu, and as such is comparable to the Klamath affair. On the other hand those of the more southerly peoples seem relatively unpretentious: Sinkyone, Yuki, Northern Pomo, northerly Wintun, Maidu of the Sacramento valley, and Southern Maidu. The Shasta performance has been described above. The Yurok novice dreams of a dead shaman who puts a "pain" in her body. A dance is made in the sweat-house for ten days during which she dances under the direction of older shamans. The "pain" is induced to leave her body and is re-swallowed, thus demonstrating control. She 471 Kroeber, Handbook, 111, 197-199 302, 361; Angulo, Psychologie religieuse, 562; Dixon, Northern Maidu, 268, 280; Gifford, Southern Maidu, 244. 472 Sapir, Religious Ideas of the Takelma, 42; Kroeber, Handbook, 150; Dixon, Some Shamans, 25, Notes on the Achomawi and Atsugewi, 218. 1930] 263 2.64 University of California Publcations in Am. Arch. and Ethn. [Vol. 30 t.hen goes to a seat or little monument on the mountain for reasons obscure. On her return the dance is repeated and followed by one about a large fire to "cook the pains." The Tolowa dance is like that of the Yurok. The Hiupa dance is but barely described. "At night a dance is held in the sweat-house. The candidate dances around the fire as many times as possible. This is continued until the prospective doctor gets the desired clairvoyance, or the sucking doctor gets the 'pain' in his mouth. During the nightly dance the candidate makes the motion of catching the pain and putting it into his mouth. Some- times the desired power is long in coming. The candidate goes in that case to a high mountain and dances in the rarer air. " When dreams come to the Northeastern Maidu novice, or he has had an answer to his quest in the mountains, he becomes ill. A performance is then held at which older shamans attempt to appease his spirits. He then sings his spirit songs. This may continue through the winter, when, after his ear lobes are pierced, he sets off to the mountains to acquire more spirits.473 The southern performances are simpler. The Chimariko dance takes place out of doors, where an established shaman put. a "pain" into the novice's mouth. Of the northerly Wintun, Kroeber writes: "Shamans are 'finished' in a dance held in the sweat-house at, night. Older doctors suck the novices' bodies clean; then call the yapaitu or spirits, who enter the neophytes and render them temporarily uncon- scious or maniac"; of the Sinkyone: "A new shaman fasted and danced at nights [sic] for a period of some duration in a conical house erected or reserved for the occasion, while more experienced colleagues accompanied him, interpreting his symptoms and advising him in the gradual acquisition of mastery over the difficulties of his involuntary art." The Yuki dance takes place in summer; older sha- mans dance with the novices for five days. The novices' spirits come to the older shamans and instruct them how they can be made man- ageable. The spirits shoot the novice with spirit obsidians to make his heart light and clean. Among the Northern Pomo a shaman is called in to diagnose the novice who has had an experience. They dance together and the shaman sucks out a feather as.a sign that -the novice will become a shaman. The Maidu novice of the Sacra- *mento valley division falls sick after his experience. Shamans sing over him and give instruct-ions. They insert long spirit objects in his 473 Kroeber, Handbook, 63, 127; Goddard, Life and Culture of the Hupa, 65; Dixon, Northern Maidu, 274-279. Spier: Klamath Ethnography nose, over which he must demonstrate control by withdrawing them. Southern Maidu are initiated by established shamans. "They are placed in pits with 'medicine' and 'killed' with a piece of 'poison medicine' about a half inch long."474 It does not seem probable that the Cahuilla novice dance is related to these northern forms, since it is not only different but is not known in the intervening area. The Cahuilla youth gives a feast and then dances for three nights to gain ability to sing his songs.475 There are some rather interesting linkages in these data. Klamath and Shasta have the dance in winter. Possibly the Northeastern Maidu novice's dance through the winter is related but that is doubt- ful. Presumably summer is the dance time for others, as specifically the Yuki. Its duration is five days among Klamath; Shasta, three for each performance; Modoc, five in at least the first performance; Yurok, ten; Yuki, five. Klamath and Shasta hold it in a dwelling, others in the sweat-house (Yurok, Hupa, northerly Wintun) or special dance house (Sinkyone). The Klamath novice dances as close to a huge fire as he dares; so the Shasta in the earlier dance; the Yurok dances around a. hot fire to " cook the pains, " the Ilupa around a fire. Shasta, Yurok, and Yuki believe the aspirant is shot with "pains" by his spirits; the Yurok draws them out. and swallows them to show his con- trol. The Hupa novice duplicates this by seizing and swallowing the " pains. " The older shamans' part in the Yurok, Sinkyone, Yuki, and Northeastern Maidu dances is to show the novice how to gain mastery of his "pains" or spirits. Among the northerly Wintun, Northern Pomo, and Sacramento Valley Maidu they treat him for his disorder. They first suck his body clean; the Pomo suck out a feather. The Valley Maidu combine this with an opportunity for the novice to demonstrate his control by drawing out "pain"-like objects they put in him. The Yurok, Hupa, and Northeastern Maidu novice seeks further experiences in the mountains; the last after his ears are pierced. Ear piercing is the preliminary to a power quest among the neighboring Achomawi and Atsugewi. * There can be little doubt that the north Californian dance, which is essentia.lly one of instruction and control, is related to the tutelage exercised over neophytes in the Secret Society-Kuksu complex cen- tering farther south. 474 Kroeber, Handbook, 149, 197, 199, 361; Loeb, Pomo Polkways, 326; Dixon, Northern Maidu, 267, Chimariko, 303; Gifford, Southern Maidu, 244. 475 Hooper, Cahuilla, 334. 1930] 265 266 University of California Publications in Am. Arch. and Ethn. [Vol. 30 The Winter Performance Shamanistic performances among the Klamath could take place only in midwinter (December-January). Every shaman made a point of exhibiting his power at. this time, several of them sometimes joining for a performance. Midwinter was the most propitious moment to recall the spirits. In fact, it was at this time that laymen of the audience, who had some spirit power, would feel impelled by their own spirits to sing their songs. This is the essential element of the Kwakiutl secret society per- formances in midwinter. The society members are those who are repossessed by their spirits and the novices those who are newly pos- sessed at this time. This is probably general among all adjacent groups having secret societies. Boas remarks that "among all the northwest coast tribes these ritualistic performances are confined to the winter months, and the season is set off from the rest of the year as the sacred season.476 Southward of here there are secret societies but they are not spe- cifically connected with spirit repossession; such as those of the Qui- leute, whose performances take place in the winter months. The possession is general among all individuals. Again, among the Klal- lam the guardian spirits come to their owners once a year, generally in December or January. The person becomes "ill," calls on his friends to help him sing, and "when the guardian spirit is satisfied, it leaves the owner, who then becomes well again. " Identically the same is recorded for the southern Puget Sound groups (Snohomish, Sno- qualmi, Nisqually, etc.) as occurring from the middle of November to the beginning of January.477 Both among the Klallam and on the Sound these dances take place at. the winter potlatch, but they also occur at other potlatches through the year, merely as dance perform- ances and without repossession. This renders Olson's observations on the Quinault intelligible: their potlatches are usually given about Christmas or New Year, and at these people sing their spirit songs. This suggests that here t.oo repossession ta.kes place at this t.ime. On the other hand, his notes state specifically that spirit power is not stronger at this time. The spirit dances of the Wishram are very 476 Tribes of the North Pacific Coast, 244. 477 Frachtenberg, Ceremonial Societies of Quileute, 322; Gunther, Klallam Ethnography, 290, 306; Haeberlin-Gunther, Ethnographische Notizen, 56, 59, 73. Spier: Klamath Ethnography similar to the Klallam-Sound Salish affairs. These likewise are held only or primarily in midwinter, that is, from December. to March... Potlatching on this occasion is represented by some trifling gifts made by the singers. It is not clear that the man who rises to sing his spirit song at this time is repossessed, but it is suggested by a statement I recorded that when the auditors heard their songs sung, they "became like fire inside, wild, " offering bits of flesh to be cut off. The guardian spirit dance of the Nez Perce closely parallels this performance. This appears again among the Lillooet, and it may be that Teit's reference to the "manitou dance" of the middle Columbia Salish refers to a similar affair. (These people formerly occupied the territory between the Wishram and Nez Perce.)478 A Wasco analogue may be reflected in a tale describing just such a singing and dancing festival among the animals; each of the five nights they dance closer to the flames.479 This is similar to the Klamath practice of dancing as close as possible to the fire, which I know is also the custom among Puget Sound groups, hence may have been universal in these performances. There is a suggestive connection with these northern ideas in the Alsea custom of telling traditions in one month alone, January, and never a gain during the year.480 It is significant that the novice among Klamath and Shasta waits until winter for his inaugural dance. Among the latter this is the period in which the candidate is most likely to be possessed: "doctors generally begin their dreams and dancing early in the winter, as it is then that the Axeki [spirits] are always about the camp. X'481 If then the Klamath winter performance is construed as related to northern rites, by reason of its time, and the repossession of both shaman and layman, it is unique in one respect. I have been unable to find that elsewhere there are such regularly recurring opportunities for shamanistic display. To be sure, these occur elsewhere and quite widely, but not as regimented performances fixed as to time. The Shcaman's Assistant The Klamath shaman regularly employs an assistant who serves as interpreter of his rushing mumbled speech. Similarly, at both winter performances and while curing, a second individual is selected to call the spirits and prompt them to act. 478 Spinden, Nez Perce, 262-264; Teit, Lillooet, 286, The Middle Columbia Salish, 98, 128. 479 Sapir, Wishram Texts, 311-312, cf. 95-99, 129-131. 480 Farrand, Alsea, 245. 481 Dixon, Some Shamans, 24. 1930] 267 2,68 University of California Pubticationm in Am. Arch. and Ethn. [Vol. 30 Elsewhere in the area from the Columbia river into northern Cali- fornia the shaman is provided with an assistant of this type. He participa.tes, however, only in curing ceremonies. At the southern fringe of the area his functions fuse with those of the clown attendant in secret society performances. On the other hand, in western Wash- ington he is not an interpreter or repeater of speeches, but serves purely as a lay assistant. The Modoc assistant comes nearest to the Klamath in type. Meacham refers to him as an orator who repeats loudly the shaman's speeches and who supports the shaman novice while dancing. Since the specific reference to him is in connection with the novice's dance, we may presume that he regularly a.ccompanies the shaman in all his operations.482 Takelma, Wishram, and Achomawi assistants function during a cure like those of the Kla.math, repeating the shaman's incoherent speech. A shamanistic dance, purely for display, occurs among the Wishram, but I did not learn that an assistant of the Klamath type had any part in it. This appears, however, among the foothill group of Northern Maidu, where, while a shaman shows his ability to com- mune with the spirits, a layman "apes the shaman in everything, repeats after him everything he says, and in every way tries to make the spectators laugh. It is considered a compliment to laugh, and a sign of appreciation." This "clown" has also full rein at the cere- monial for the dedication of a new dance house. Powers reports for the Southern Maidu a boy who repeats every utterance of a shaman during his performance. This may refer to a clown like that of the Northern Maidu or to an assistant who helps at a cure.483 There is no trace of the clown in the dances of the Northeastern Maidu, but he appears again in the performances of the Sacramento valley branch. He is speaker for the dance leader, mimics the words and actions of the dancers, and attempts to make the spectators laugh by his antics. Such clowns a.ppear regularly in Kuksu cult dances a.mong other Sacramento valley peoples and are known as far south.as the central and southern Yokuts.484 In the region west and north of the Wishram, shamans have assist- ants of a sort but they are not speakers. For example, the Chinook 482 Meacham, Wi-ne-ma, 129-136. 483 Powers, Tribes of California, 334. 484 Sapir, Religious Ideas of Takelma, 43; Angulo, Religious Feeling, 356, Psychologie religleuse, 568-569, 574-577; Dixon, Northern Maidu, 271, 286, 295, 311-318, 322; Kroeber, Handbook, 383. Spier: Klamath Ethnography shaman who seeks to recover the soul of a sick man uses "a manikin made of wood and cedar bark. When a conjurer wants to make use of this manikin he gives it to a person who has no guardian spirit, who shakes it for him, and they two go to the country of the ghosts." Boas states that this type of cure was derived by the Chinook from the Chehalis. Such sticks, carved and painted with reference to their guardians, and here described as manikins, are common property to those with spirit helpers from Oregon northwa.rd. When a Quinault shaman goes to cure, he is regularly accompanied by an aid who holds and rattles his carved stick. Here again illness is due to soul loss but there is no journey to the land of the dead.485 In quite another direction the Pomo appear with an assistant for the curing shaman. He shakes the shaman's rattle and sings his songs. As Kroeber remarks, the appearance of an assistant in this quarter is unusual, and in the present state of our knowledge not con- nected by distribution with the instances cited above.486 One cannot escape the impression that this provision of a speaker for the shaman is linked with t.he chief's spokesman of many tribes of the west. The Northwest Coast type of spokesman is known as far south as the Wishram on the Columbia and perhaps in coastal Oregon. In the Basin, from eastern Washington to northern Arizona, there is word-for-word repetition of formal speeches. There is no chief 's spea.ker in the intervening area from Wishram to Maidu, but there is for the shaman. When we realize that in this region, chiefs are socially eclipsed by shamans, we may well argue that a substitution has taken place. Objection may be made, however, that this is equally true in the Basin, where, so far as we know, shamans do not have speakers. Accessory Elements The distribution of analogues of the Klamath shamans' feats is quite irregular, to judge by the few notes I have collected, but this is undoubtedly due to incompleteness of information. Merely to put these on record, I present them here. The Klamath shaman swallows fire; so does the Lillooet and Thompson. Live coals are eaten by Quinault secret society performers, 485 Boas, Doctrine of Souls, 40, Chinook Texts, 207; Olson, Quinault field notes, 1925. 486 Handbook, 258. 1930] 269 270 University of California Pubicatiois in Am. Arch. and Ethn. [Vol. 30 Pomo ghost dancers, Cahuilla and Pima shamans. Perhaps the Nis- qually trick of eating cold ashes is connected with these.487 Treading on fire or hot coals is no trick of the Klamath, but occurs I believe among the Modoc. It is also known from the Quinault, the Puget Sound Salish, Northeastern Maidu, and in the boys' rites of southern California. The Pomo can plunge his hands in fire; the Yurok shaman handles hot stones.488 The Klama.th shaman can fill or empty at will a basket with live fish, pond-lily seed, or blood. The Thompson is credited with causing water to appear and disappear, the southern Yokuts with making fish appear in water, and the Pima with transforming water to wheat.489 The Klamath hold that food brought to the shamanistic perform- ances is first eaten by the frog spirit. Until this spirit has partaken, the food may not be touched for fear of sickness. Similarly in a Shasta ceremonial in which a shaman frees a village of lurking disease, his guardian spirit is thought to eat food which has been brought for that purpose. If this is not. forthcoming, the spirit may devour some- one instead. The Takelma analogue takes a different form: a shaman must not ea.t before dancing, for, since the dance is reckoned food for his guardian spirit, he, the shaman, would be guilty of the impropriety of eating first.490 It may be that this is only a special construction put on the Klamath custom of throwing morsels of food to the spirits before one eats. Certainly the Kwakiutl who puts bits of food in the fire before the feasters eat is presenting them to the spirits of the fire.49' The Klamath, and probably Modoc, shaman has set up on the out- side of his dwelling a carved image of a boy or dwarf spirit. To find a carved image here is distinctly unusual in a culture in which wood- working is minimal. This seems to me to be another evidence of influence from the woodcarving area to the north. While the analogy may not be proper, it may be pointed out that to the northwa.rd a shaman, in fact everyone with a guardian spirit, carves or repre- sents that. spirit on a stick which he carries during his performances. Tillamook, Chinook, and Quinault shamans have such carved sticks, 487 Teit, Lillooet, 289; Thompson Indiana, 362; Olson, Quinault field notes; Kroeber, Handbook, 264; Hooper, Cahuilla, 331, 335; Russell, Pima, 259; Hae- berlin-Gunther, Ethnographische Notizen, 68. - 488 Raeberlin-Gunther, op. oit., 61; Dixon, Northern Maidu, 279; Kroeber, Handbook, 68. 489 Kroeber, Handbook, 507; Russell, Pima, 258. 490 Dixon, Shasta, 488; Sapir, Religious Ideas of the Takelma, 42. 491 Boas, Ethnology of the Kwakiutl, 1331. Spier: Klamath Ethnography which their assistants hold or with which they dance. In fact, the use of special dance staffs is general among all Quinault who have spirits, as it is among the southern Puget Sound Salish. It is strange that neither Dr. Sapir nor I learned of either type of carving among the Wishram, but our information is not full.492 A tall painted pole is erected in front of the Klamath shaman's lodge from which dangles a cord bearing at intervals bunches of feathers, a piece of wood, and a miniature gambling tray, all these representative of spirits. There are analogues of this southward to the Maidu. Such a pole is also erected outside a Modoc shaman's dwell- ing; it is painted and bears the skins and feathers of various spirit animals. It is used earlier at the novice's dance. This is described as some three or four meters high and decorated with paint and a few feathers. Such a pole is set up when a Shasta shaman rids a locality of disease: it has baskets hung from it. Again when an Achomawi- Atsugewi shaman cures, a pole is sometiines set up near the house, to which is tied a bunch of feathers (qaqu). The qaqu has a significant function; it is obtained with difficulty on his spirit quest and has the power to tell him where the disease is located. The poles used for a Northeastern Maidu shaman novice are somewhat different. They are set up on the roof of the house in which the inaugural dance takes place, and inside above the spot where he stays; others are carried to the mountains where novices go on their quests. These are short wands, a meter to a meter and a half long, variously painted, with pendent strings of feathers and acorn shells, or bunches of feathers. These have some obscure but intimate relation to his spirits. Similar wands are set up at the dedication of a new dance house by the Northern Maidu of the Sacramento valley.498 Among the items of a Klamath shaman 's costume are a forehead band or neckband of woodpecker scalps mounted on a buckskin founda- tion and a bunch of yellowhammer feathers worn on hat or neck. The feathers of both these birds have value and ceremonial usage in California but not to the north. Woodpecker scalps attain the status of money among the Yurok and perhaps their neighbors. Not only have they value in themselves, but they are worked into ceremonial headdresses over most of northern California. The Yurok and Hupa use forehead headbands like those 492 Boas, Notes on Tillamook, 10, Doctrine of Souls, 40; Olson, Quinault field notes, 1925. 493 Meacham, Wi-ne-ma, 130, 136; Dixon, Shasta, 475-476, 487, Notes on Acho- mawi, 218, Some Shamans, 24, Northern Maidu, 275-278, 312. 2,71 1930] 272 University of CaUfornia Publications in Am. Arch. and Ethn. [Vol. 30 of the Klamath in their Jumping dances, and the Hupa " roll- woodpecker hats" in their Brush Dance. Dixon figures what is prob- ably a Shasta shaman 's headband with a large woodpecker scalp affixed. Northern Maidu make use of a belt or bandolier of wood- pecker scalps now mounted on buckskin, but formerly on a netted foundation. A headdress like a helmet, covered with these, is worn by the highest members of their secret society. A spirit impersonator of the Sacramento Valley Maidu wears a forehead band of these scalps, tapering at its enids. Pomo women wear "a woven band of wood- pecker feathers" (scalps?) presumably in dances: this suggests a net foundation. In all the other cases cited above the foundation of the band is a buckskin strip. Kroeber has pointed out that the Yurok headband (and I would add here all northern analogues) have skin mount.ing, while those used in the Kuksu dances of the lower Sacra.- mento valley are made on a close network. The Northern Maidu belt then represents an intelligible transformation of technique. Klamath and Shasta shamans' headbands, and the Takelma. shaman's mere use of the feathers in his hair, are attenuated usages.494 Klamath uise of yellowhammer feathers is only casual and would have no significance were these not regularly part. of ceremonial regalia to the south. There the form is commonly a forehead band. Their use is primarily central and northern Californian and they are not found in most of southern California.495 A search of the litera- ture shows the use of such bands by shamans only among Achomawi, Atsugewi, and Northeastern Maidu, and the feathers alone by Takelma shamans.496 Elsewhere they are non-shamanistic dance regalia: Chi- lula, Sinkyone, Coast Yuki, Yuki, Pomo, Patwin, Northern and South- ern Maidu, Yokuts, Koso, and Luisefio.497 It will be observed that this distribution avoids the Klamath river tribes. Circumspection of action in the presence of a shaman is held necessary in a number of groups from Thompson to Maidu. His imme- diate presence is fraught with power, so that making contact with him by his shadow or one's own, startling him, or making a misstep in passing, are all potentially dangerous. A shaman of the Northern 494 Kroeber, Handbook, 26, 56, 433-434; Goddard, Life of the Hupa, 19, 69, 86; Dixon, Shasta, 481; Northern Maidu, 149, 284; Loeb, Pomo Folkways, 157; Sapir, Notes on the Takelma, 264. 495 Kroeber, Handbook, 140, 149, 508, 665. 496Dixon, Notes on the Achomawi, 219, Northern Maidu, 283; Kroeber, op. Oit., 316; Sapir, op. oit., 264. 497 Kroeber, op. cit., 140, 149, 173, 216, .261, pls. 59 and 77, 508, 591, 665; Dixon, Northern Maidu, 152, 283; Faye, Southern Maidu, 48, 50. Spier: Klamath Ethnography Maidu of the Sacramento valley may kill merely by letting his shadow fall on his victim. A Klamath cannot pass in front of a shaman nor drop something in front of him, else the shaman suffers a hemorrhage. Similarly a Kwakiutl apparently does not pass behind an eating shaman, else he too has a hemorrhage. The Wishram shaman, angry with one who endangers him by passing behind or startling him, may bewitch in revenge. A Cowichan of the mainland group will not let his shadow fall on a sick shaman, for the latter may seize the chance to replace his own lost soul. Similarly the Lillooet avoid letting their shadows fall on a shaman or standing in his, for fear of consequences. The Thompson, too, will not let his shadow fall on a shaman (the reverse is not harmful), and dares not startle him nor eat meat with- out inviting him to share it.498 Clairvoyance Prediction is a widespread, if not universal, ability of shamans. A somewhat unique form among the Klamath is that at the dance pre- ceding the departure of a war party. The shaman dances in front of the warriors in order to discover those who will be wounded: he sees them bleeding. He also accompanies the party to watch over them and cure wounds. Something of the same sort is noted for the Modoe. This also seems to take place among the Chinook: "before the people go to war they sing. If one of them sees blood, he will be killed in battle. " Perhaps this is general in the northwest; at any rate Goddard observes of the Northwest Coast at large: "Every war party includes a shaman who by his supernatural knowledge warns of danger and points out favorable times and places for attacks. The shaman, by catching or destroying the souls of the enemy, makes the killing of their bodies an easy matter."499 As I have pointed out it is unusual that the Klamath shaman treats wounds and fractures by matter-of-fact methods. But this has also been noted of the Achomawi. Angulo thinks this true only since the advent of the whites, but it is surely an old practice of the Klamath.500 The practice of freeing a localit.y or village of disease, dispatched there by a malevolent shaman, is probably more common than I have 498 Dixon, Northern Maidu, 269; Boas, Ethnology of the Kwakiutl, 730; Lower Fraser River, 9; Teit, Lillooet, 288; Thompson Indians, 361. 499 Gatschet, 1:21; Boas, Chinook Texts, 270; Goddard, Indians of the North- west Coast, 119. 500 Angulo, Psychologie religieuse, 571. 1930] 2.73 274 University of California Publications in Am. Arch. and Ethn. [Vol. 30 been able to discover. At any rate, I note it for Achomawi-Atsugewi and Shasta, as well as Klamath. Following the usual shamanistic seance the Klamath shaman searches the vicinity for the malevolent spirit. He rushes about until he falls, blood streaming from his mouth. Several men must support him by the arms and restrain him until he can destroy it in water. The Achomawi-Atsugewi shaman forces the disease out of the ground by his power and then seizes it. He staggers about in his effort to hold it until he can plunge it into water. He then swallows or burns it, or dispatches it to kill its sender. The Shasta erect a shaman 's pole at the affected spot. The shaman dances, calling on her spirit to dislodge the disease "pains" from the ground. "If the shaman's guardian is successful, he succeeds in pushing the pains into the river, and drowns them. " Certain features of a Chinook cure are like these. When their shaman has taken a disease from a sick man, he must be held by several men to overcome the strength of the disease spirit, until he can plunge it into water.50' The destruc- tion of an extracted disease by placing it in water is a common feature among Washington tribes. I have seen it, for example, as part of a cure by members of the pseudo-Christian sect of Shakers among the Skokomish and learned that it occurred among the Okanagon. Various Beliefs I have a somewhat random series of analogues of Klamath beliefs. The Klamath hold that a storm rages when a shaman dies, when he is cremated, and again when his lodge is burned. The Northern Maidu hold that it thunders at once when a great man dies, and inasmuch as the most important individuals of the Maidu are shamans, it is clear that this is a comparable belief. The Thompson have it that the death or burial of a person causes an immediate change in the weather; that of a shaman, a sudden alnd extreme change. The death of a Wind River Shoshoni causes a whirlwind. These beliefs are apparently a single expression of a more general belief that both births and deaths are signaled by a change in the weather. For example, the Klamath also hold that it always blows at a woman's first parturition. Analogous to this the Wishram say that extreme weather marks a birth. Further, a rainbow points to the spot where the birth takes place. The latter belief at least is 501 Dixon, Notes on the Achomawi, 2:19, Shasta, 487; Boas, Chinook Texts, 208. Spier: Ktamatth Ethnography shared by other Upper Chinook (Wasco, Cascades, and Clackamas) and the Eastern Pomo. Similar to their belief in respect to death, the Northern Maidu say it thunders when a woman has a miscarriage, and the people of Cochiti say the same of a Navaho woman when the wind blows. Again, a sudden change of the weather accompanies a Thompson birth.502 An association of the moon and frog is common to! several tribes of this quarter. The Klamath say that the Moon is saved by his wife, Frog, during an eclipse, from being eaten by a grizzly. The frog can be seen in the moon. The Quinault hold an identical belief. An eclipse is caused by the "upper fisher" (a spirit animal) biting the moon. " There is a lady frog up there who helps the most and really saves the moon." A tale describes this frog as married to the moon. According to the Northern Maidu, the spots in the moon are thought to be a frog. A Northern Paiute tale describes Moon as swallowing Frog.503 One of the few acts of outright magic among the Klamath relates to causing the wind to blow. This is accomplished by poking in the hollows of a certain rock, striking another, or poking in a particular eddy. The Wishram also maintain that stirring about in a hole in the rocks near their village will raise a wind; the Quinault that thrusting a stick into a spot where gas bubbles rise in the water will bring a storm; and the Twana (or Klallam?) that shaking a certain rock will cause rain.504 Comparable to the Klamath belief that the "cat" Bassariscus kills deer by wrapping his long tail around their necks, the Northern Maidu state this of the mountain lion.505 Sulmmary The general basis of Klamath shamanism is like that of other interior Plateau and Basin peoples. Social importance attaches to shamans above even that accorded chiefs. Rituals are meager and throughout are crisis cerenlonials. As with the generality of North American tribes, acquisition of power is free for all, shamans stand- 502Dixon, Northern Maidu, 266-7; Teit, The Thompson Indians, 3714; Spier and Sapir, Wishram Ethnography; Lowie, Notes on Shoshonean Ethnography, 297; Goldfrank, Cochiti, 83; Loeb, Pomo Folkways, 316. 503 Olson, Quinault field notes, 1926; Dixon, Northern Maidu, 266 (the tale relative to this is not available to me); Natches, Northern Paiute Verbs, 257. 504 Spier and Sapir, Wishram Ethnography; Eells, Indians of Washington Territory, 674. 505 Dixon, Northern Maidu, 266. 1930 ] 275 276 University of California Publication in Am. Arch. and Ethn. [Vol. 30 ing out only by their greater capacities. The voluntary quest in an isolated place and the formulation of spirits as birds, animals, and other things in nature are equally general North American. There is, however, a local phrasing of shamanism in the interior, centering in southeastern Oregon and northeastern California, that is, from Klamath to Maidu. Inl part this is due to the presence of ele- ments shared equally with the whole of northern California; in part to specific influences from two directions, northwestern California and the Northwest Coast. These have been caught up into complexes characteristic of this territory. Certain features are, further, peculiar to this locality alone. California and the western segment of the Southwest, are char- acterized by the few individuals on whom supernatural grace descends and the dream as the means of its reception. Elsewhere east and north the voluntary quest is open to everyone and the vision its cul- mination. Northern California combines these variously; the Klamath by making the quest, permitting anyone to do so, and holding the quest a prelude to the dream. Spirits are the source of power in northeastern California, as they are northward and eastward. The curious northwest Californian con- cept of "pains," alike sources of power and pathogenic in their action, infiltrates the beliefs of the northeastern quarter of the state but does not penetrate to the Klamath. Nor do the mythic super- naturals of the northwest coastal strip known as far south as the Columbia river make their appearance here. But the conceptualiza- tion of the shaman 's spirits as dwarfs, localized in northern California, does appear a.mong the Klamath. There is also in the interior some inheritance of a predilection to acquire spirits: in this the Klamath share. But there is nothing of that rigid inheritance of particular spirits which focuses on the Northwest Coast, nor the inheritance of paraphernalia known to central California. The manner of conducting the quest has two local peculiarities. The curious practice of piling rocks in the mount-ains is known north and south of the Klamath from Nez Perce to Atsugewi. Diving for a spirit, swooning, and awakening with a hemorrhage has a, far more widely extended range. In one form or another, it is found from the Kwakiutl through northern California. The emphasis on it links the Klamath definitely with the northern members of this group. 1930]Spier: Klamath Ethnography * The time of life at which supernatural power may be gained is not limited among most western tribes. Those of eastern Washington and the interior of British Columbia, however, tend to restrict the quest to adolescence, and with them seem to link the secret society initiations of the Northwest Coast. One gains the impression that in northwestern California, where the acquisition is involuntary, it descends only on adults. The Klamath, like others of northeastern California, follow the more widespread habit of seeking power at any time. Feminine shamanism is characteristic of northwestern California. While this appears in the relatively large number of women practi- tioners as far east as Shasta and Achomawi, the Klamath are unaffected by this influence. Specialization among shamans is not marked over most of western America. Such distinctions as a-re made between those who cure and mere clairvoyants, or the really powerful shaman and one with limited abilities, are very va.riable and never particularly impressive. Such distinctions do not appear among the Klamath at all. Neither do the common Californian specialists, the rattlesnake, bear, and weather shamans. Weather control is exercised by any Klamath shaman who has the requisite spirits. Northern California and Oregon share an initial dance for the novice: a performance of initiation and mastery in the south, a demon- stration of ability in Oregon. The Klamath affiliate with the latter. Its local form on the Klamath river below the Klamath (Yurok, Shasta, and Modoc), a dance made imimediately on the acquisition of power with a second somewhat later, has no counterpart.among the Klamath. The general form of the Klamath novice dance is identical with all their shamanistic seances. In one particular they affiliat.e with the down-river tribes, namely in da'ncing over a hot fire, but even this is known on the southern fringe of the Northwest Coast proper, if not elsewhere. Northwest Coast influences appear at. two points: the time of shamanistic seances and the woodcarvings representing spirits, Kla- math seances are restricted to midwinter, the novice dance among them. This is regularly the period at which all who have spirits are repossessed and impelled to dance, sha.mans among tihem. The regi- mentat.ion of the repossessed in. the midwinter. sacred period bf the Northwest. Coast is lacking, but a diminution of t.he idea and its forms is clearly traceable from that focus to Shasta and Klamath. Wooden 1930] 2.77 2.78 University of California Publications inn Am. Arch. and Ethn. [Vol. 30 images are unique in the non-woodworking culture of Klamath and Modoc, yet this too is traceable southward from the same center. It is far less certain that clairvoyance in connection with a war party is derived from that source, yet the trait is shared by the northern coastal people. Some elements are local in the Oregon-north California interior. Such is the shaman's assistant, his mouthpiece and interlocutor. He is present at cures from Wishram to Achomawi; present at seances from Klamath to Southern Maidu. For some of the Maidu he is a clown, which links him with the clown among the Kuksu impersonators of central California. There is reason to think that this functionary is historically connected with the chief's speaker known from the Northwest Coast through the interior to the Southwest. Other local elements are the pole having spirit, reference (Klamath to Northern Maidu), the notion that food is first eaten by the spirits (Takelma., Shasta, Klamath), the ridding of a locality of disease (Klama.th, Shasta, Achomawi-Atsugewi). Other elements, while local, are somewhat more widely distributed. There is the circumspection with respect to the shaman 's shadow and to startling him.; the notion tha.t weather changes at birth, and death: all these are common from southern British Columbia to the Maidu, largely through the interior. The Klama.th shamans' dress bears some imprint of northern and central Californian influence in the use of woodpecker scalp head- bands and of yellowhammer feathers. Both are characteristic dance regalia, in the south; on the Oregon border they have passed over to the shamans. In broad perspective Kla.math shamanism is of generalized, undifferentiated North American type, sharing some features that are local, and but moderately influenced from the Northwest Coast and the central and northwestern foci of California. Its general nature is recognizable in the quest. and the undifferentiated power it yields; the most marked local elements are the predisposition to inherit power, the novice dance, and the shaman's assistant. California at large appears in the dream as a vehicle of acquiring power; northern Cali- fornia in the dwarfs that figure among Klamath spirits; central Cali- fornia. in the shaman's headdress. Northwest Coast influences are no stronger; repossession in midwinter and the use of carvings color Klamath shamanism only mildly. Spier: Klamath Ethnography If we look for elements peculiarly Klama.th, they are difficult to find. At best there is only the regularity with which seances occur at midwinter, and even here we do not know with certainty that this does not occur elsewhere. Further, if we feel that this regularity is due to recurrent. possession, this merges with general Northwest Coast practice, and its uniqueness disappears. The peculiarity of the Kla- math complex, then, rests only in the specific combination of widely distributed elements that they have effected, a combination known nowhere else in exactly this form. THE MARRIAGE COMPLEX Social relations among the Klamath, even more than the shaman- istic practices, are clearly explicable as a function of their geographic position. They are most clearly crystallized in the matters of chief- tainship, sla.veholding, concepts of wealth, and the relative social standing of shaman and chief, but especially in their marital rela- tions. Household and village groupings, puberty rites and analogous customs, have much less definite or peculiar form among them. For this reason, the discussion which follows will be confined l.argely to the earlier group of topics. As the most complex situation is that surrounding marriage, I will analyze this first. While marriage among the Klamath is contracted by purchase, it is not a negotiated transaction. Payment is socially obligatory. It is a seal of respectability and in a rough way the amount changing hands is a measure of the social pretensions of the families involved. Hence clandestine matches are severely frowned on. No obligation may be assumed for which payment. is not forthcoming. Yet such payment does not fix the status nor the value of an individuial with any pre- cision. For example, the purchase price of a murdered man's mother or that of his bride never enters consciously in assessing his blood price. Failure to pay for a bride is a blot not easily erased. It must be remembered that. the reciprocal gifts from the bride's family frequently outweigh the purchase price. Further, the bride is in no sense a chattel, but free to leave her husband at will, he having but slight recourse too recover his gifts. Nor is a wife obligated to bear children to her husband, nor does their birth cancel the marriage debt. In adultery it is clear that a man has no eye to the monetary value of his wife's affections. He kills the lover; to take payment is unthinkable. It. does not appear that a marriage brings about a con- 279 1930] 280 University of California Pub7ication.s in Am. Arch. and Ethn. [Vol. 30 tractual union between the families. The levirate is customary but not obligatory, and the sororate only occasional. It is true that the hus- band's brothers may take exception to her marriage with another, but this may mean only that they look upon it as an affront to their brother's memory. Marriage is denied only between blood relatives. There is no village exogamy nor trace of restrictions or preference of any other order. At best the evidence suggests some tendency to marry within the tribal subdivision, but this may mean only that contacts are less frequent outside of this group. The formalities of marriage prescribe a series of visits with an exchange of gifts. The prospective groom sends a male emissary with a bundle of sticks representing his offering. The bride's family pro- ceed to the groom's house where they exchange gifts with the family of the groom. In kind and in quantity these are modest indeed when measured by Northwest Coast standards. A month later the couple and the groom's family return the visit, when again gifts are exchanged. Beyond these two, further formal visits are made only by the social pretentious. These formalities may be frequently initi- ated while the couple are still pre-adolescent, the bride being carried to her husband's home after her puberty dance. The general nature of the Klamath marriage relation smacks strongly of the Northwest Coast. This is apparent in the premium attached to purchase, to a lesser extent in the weighting accorded the amount of the purchase price, in the reciprocal nature of the gifts, and the series of formal visits before the affair is considered ter- minated. What is absent here is the notion that rank and privileges of a tangible nature are acquired by a marriage purchase. This can best be brought out by instituting a series of comparisons with other tribes. For this purpose I shall select only a limited num- ber of cases. For one thing data from Oregon tribes are almost wholly lacking, and those for California and Washington are both scattered and scanty. Throughout California, save on the eastern border of the state, there is an appearance of purchase. But in reality it is only a pre- sentation of gifts of no great amount as an earnest of the son-in-law's good will, a matter of etiquette rather than a stipulated price. The formality of giving varies as much as its conventional expression. With some, bride service is in order; with others, the parents-in-law must reciprocate the gifts. It is everywhere little more than an Spier: Klamath Ethnography observance, far removed from the set contractual relations established by a Northwest Coast marriage. It is nowhere an occasion for the display of wealth, and social prestige is not fixed by it.50s Among the Basin tribes there is not even this semblance of pur- chase. What these people desire in a son-in-law is a good provider, who normally takes up residence with his wife's people. His incidental services can hardly be construed as payment in any sense. The groom slips into their life as unobtrusively as possible. Ceremonia.l is almost wholly lacking; at most. a feast in celebration.507 The situat.ion in the Plains is not far distant, resembling rather that in California. In general the approach to the northwest.ern con- cept consists only in an exchange of gifts. The groom's family bring gifts to the home of the bride, receiving an equivalent, sometimes to a greater amount. Married life begins at. once with the bride's family to whom the son-in-law renders services. But it is doubtful that either the gifts or the services are looked on as specific payment. Rather they serve to esta.blish friendly relations and no more. Large amounts of property might be given for the honor of the act, but it was equally common for an old man to marry his daughter to a suitable man to obtain a provider for the family, no compensation being thought necessary. While some ostentation is inevitable in a Plains ceremonial act, there is here no lavish display of wealth. Equally foreign to their thought is the idea of social status fixed by a transfer of property.508 If we t.urn from these t.o the Northwest Coast tribes, a very differ- ent picture is presented. We need go no farther nort.h than the Kwakiutl for a typical expression of attitudes. The rank of a fa.mily, the social status of a child, are clearly defined in a regimented series depending largely on primogeniture coupled with wealth. Marria.ge is a purchase conducted on the same plan as the purchase of all cere- monial symbols of exceptional value. Its mechanism is the potlatch, the ceremonious bestowal of wealth on a social rival to enhance one's own status, with the certainty of its return several fold if the rival is to maintain his rank undiminished. A man buys in this fashion the rank and social privileges of his bride's family for the children of this marriage or heirs or relatives whoever they be. When the bridal price is ultimately returned two or three-fold, the contractual relation is at an end. The groom has the obvious advantage in this transfer of 506 See for example, Dixon, Northern Maidu, 239; Kroeber, Handbook, 492. 507 Lowie, Notes on Shoshonean Ethnography, 275. 508 Kroeber, Ethnology of the Gros Ventre, 12; McCOlintock, Old North Trail, 184; Grinnell, The Cheyenne, 137. 1930] 281 2.82 University of California Publications in Am. Arch. and Ethn. [Vol. 30 wealth in addition to the privileges that accrue by his investment. So little is the bride concerned and so great the emphasis on the acquisi- tion of new prerogatives for his family, that a man ma.y carry out the fiction of a marriage where there is no bride. In the desire to keep the privileges within the family, a man of high degree may even espouse his younger brother's daughter. Normally in return for the huge purchase price, the wife is given him as a first instalment. The heraldic crests of her family, its social privileges, and a considerable amount of property are given later to members of the husband's family. This repayment annuls the mar- riage; the wife is now free to return to her people, but should she choose to stay it, is customary to make a new payment to the father- in-law. The chief who desires marriage for himself or his son invit.es the aid of his townsmen in a matter of mutual interest; " our bride " is to be acquired, "all the tribes [lineages] have married my daughter" is her father's phrase. Four men are dispatched to the bride's town to offer the huge bride price, four hundred blankets and more. When the groom's party arrive, their spokesmen with stereotyped speech measure out the blankets, which are carried into the father-in-law's house. Like most Kwakiutl ceremonies, the marriage ritual is interwoven with myth; at this point the son-in-law and his party are put to a test which has mythical background. They must rush through torches into the house, or they a-re tried by a great fire and shown the sea monster whose mask spews out th.e bones of unsuccessful suitors. The father- in-law now ma.kes his first repayment in self-depreciatory manner, two hundred blankets, a gesture derogatory of his rival's lavish giving. These the groom promises his followers in the name of his future heir. Soon after, when the father-in-law is ready to make full repayment, he visits the other with his followers. He makes a sym- bolical gift indicating the wealth he returns, the copper plates, the house and carvings the son-in-law demands, his own title acquired in marriage, and his ceremonial office and dance in the secret societies, which are to go to his grandson. The bride is brought out to have her valuable finery placed on the shoulders of the groom. Next day the young man distributes the wealth among his tribesmen in recom- pense for their aid at the marriage, and to start the wealth on an interest-bearing round once more. In doing so he assumes the proud titles acquired from his wife's father and bestows them too on others. Spier: Klamath Ethnography The marriage of those of lower rank is doubtless analogous but on a smaller scale.509 I shall trace the elements of this marriage complex from this area southward, in order to make clear their relation to the customs of the Klamath. At the outset I will deal with the interrela.ted attitudes toward wealth and rank, marriage as an expression of it, and local exogamy dictated by it. Later I will return to a comparison of the marriage ceremony itself, to the levirate and sororate as expressions of family contracts, and to related phenomena such as reactions to adultery. Purchase, Rank, and Exogamy The marital customs of other Vancouver island tribes duplicate those of the Kwakiutl in all essentials. Their divergence is in a less sharp definition of rank and prerogative and diminished elaborateness of marriage ceremonialism. This diminution can be traced by suc- cessive steps to the Klamath and beyond. Not only are the attitudes of the island tribes respecting rank and wealth analogous to the Kwakiutl, but mnarriage is the same kind of mech.anism for their transfer. We may also assume here the kind of local exogamy implied for the Kwakiutl. As marriages with relatives are normally pro- hibited, and mates of suitable rank and wealth are not to be found in the same tribe or village, a practice of local exogamy follows. While there is no prohibition against marriage within the community, the vast majority of alliances must be outside, since pretension to rank sinks deep in the several strata of society. There is at least some studied exogamy on the part of those of highest. rank for whom foreign alliances mean increased prestige. As expressed by Sapir "among the more southern tribes inter- marriage is prohibited only between such as are demonstrably related by blood, even if fairly remotely so. Owing to the structure of the village community, this would in many cases mean that there are few persous in a village that one is legally entitled to marry; but it is import-ant to note that the village community as such need not be exogamous, that is, does not specifically prohibit intermarriage among its members. '1510 509 Boas, Social Organization and Secret Societies, 358, 421, Ethnology of the Kwakiutl, 776, 1344, Social Organization of the Kwakiutl, 116. 510 Sapir, Social Organization of the West Coast Tribes, 367. 1930] 283 284 University of California Publicatiois in Am. Arch. and Ethn. [Vol. 30 The alternative to this exogamic practice was marriage with blood kindred. While it was generally frowned on, an exception was per- mitted if no other mate was available. This is an attitude different from that of the Kwakiut.l, who, desiring to keep privileges within the family, married his niece or cousin. When there is no available mate outside the blood group, a Klallam may marry his first cousin. Eells implies that a Twana (and Klallam?) always sought a wife first "within a certain circle of his relatives."51' There is no adequate information for western Washington tribes save from the Klallam and more moderately from the Chinook. Inasmuch as their attitudes and practices coincide in large degree, it is safe to assume that they are representative of the intervening territory. Both Klallam and Chinook view wealth as a criterion of rank and its lavish display in the wedding ceremonial as an index of social pret.ension. Marriage with a person of lower rank is unthinkable. It is apparent that most Klallam brides are paid for, yet elopements occur and slave alliances are formed. Once such marriages are con- tracted the parents make no effort to separate the couple, but a girl who marries in this way is not highly respected. Whether payments are made following elopements to legitimatize them, we are not told. Upper-class Klallam. have a definite preference for extra-tribal mar- riages, especially with the people o-f Vancouver island. Local exogamy is even more general, holding for all ranks, for in a record of fifty- seven marriages only five were with co-villagers.512 Purchase, or its simulation, is so obviously the normal Chinook procedure also that even a successful elopement will be nullified unless a small sum is paid. And these st.icklers for form go through the orthodox ritual on this occa.sion as though the marriage was not yet consummated. For a man too poor to pay, there is no recourse but in matrilocal residence with bride service.515 There are but scattered observations on the tribes of the interven- ing territory but they suggest conformable pracatices. Since we learn that common people among the Makah paid a small amount for a bride without any ceremony, purchase must have been the regular form for all classes of society. Yet as elopements were frequent and are said always to have ended amicably, this suggests an approxima- 511 Gunther, Klallam Ethnography, 241; Eells, Indians of Washington Terri- tory, 655. 512 Gunther, Klallam Ethnography, 241 f. 513 Boas, Chinook Texts, 251; Franch6re, Narrative, 254. Spier: Ktamath Ethnography tion to the Klallam situation and perhaps the payments made in such cases by Chinook. Marriage occurred in or out, of the tribe, but with no near relative, which includes fourth cousins.514 Caste stratification was as strong among the Quinault as among their northern neighbors, and marriages had to be contracted with persons of equivalent status. This brought it about that tribal or at least village exogamy was insisted on for chiefs. The sentiment was nearly as strong for com- moners as well, although in their case emphasis was on the fact that all co-villagers, or ne.arly all, were related; marriage of relatives of even remote degree being forbidden. Not onily might a man, normally not marry in his own village, but for the same reason he could not ordinarily marry in that from which his mother came. The upper class always purchased wives from other noble families; to marry below one's rank was disgraceful in the highest degree. If a common man wanted to marry a chief 's daughter, the chief would pay someone to kill him. In contrast to the Klallam desire in extra-tribal mar- riages, the Quinault had no preference for northern tribes. All mar- riages, of whatever rank, were made by purchase, that is by the exchange of gifts of high value. Elopements were common, yet this was theft and therefore dangerous, for the girl's father might kill the man, whose relatives could not retaliate. The bride who eloped must be paid for, in fact the price then demanded was inclined to be exorbitant. Legitimatizing of marriage was Quinault theory at least., yet illegitimate children were many and not looked on with contempt.515 Purchase of the same order appears among the Puget Sound tribes, at least for the wealthy, that is the upper classes. The interchange of gifts was not compulsory, so that poor people married on agree- ment alone or deferred payment until they were able. The general impression obtained from the account of these people (Nisqually, Snohomish, Snuqualmi, Skokomish, and Suquamish) is that class dis- tinctions did not cut so deep as on Vancouver island and coastal Washington. Differences of wealth were not so great, hence, com- paratively, wealth and its function in marriage were not emphasized to the same degree. The exogamic habit is present however but apparently confined rather to the upper class. Even fourth and fifth cousins could not marry.516 The Nisqually maintain that tribal exogamy held for all; one who broke the rule was not only disgraced 514 Swan, Indians of Cape Flattery, 11 ff. 515 Olson, Quinault field notes, 1925, 192.6. 516 Haeberlin-Gunther, Ethnographische Notizen, 34-38; Spier, Nisqually field notes, 1922. 1930|] 285 2,86 University of California Publications in Am. Arch. and Ethn. [Vol. 30 but heavily fined. " Generally speaking, these Indians seek their wives from among other tribes than their own .... . It seems to be a matter of pride, in fact, to unite the blood of several different ones in their own persons ... . With the chiefs, this is almost always the case. "517 Marriage by purchase was the usual form among the Nez Perce, but it is said that a marriage by agreement because of the groom's capabilities was a more desirable and honored form. This is a new note, but it was characteristic of Plains tribes to the east. The pur- chase was. effected with little ceremony. Bride and groom were not ordinarily members of the same village.518 The exogamic habit here had not the same basis as among the tribes previously discussed. It seems due solely to the Nez Perce habit of forbidding marriages to relatives of even remote degree, which necessarily resulted in village exogamy. I am unable to find evidence of exogamic practice farther to the east, but will return later to analogous forms of wife purchase. To return to the Columbia river: the Wishram had as strong a sen- timent for caste and wealth as their down-river relatives, the Chinook proper. Rank and wealth were here not quite synonymous, for chiefs were not always the wealthiest individuals. Marriage was always with one of the same rank, so that the children would have as high standing as their parents. Payment was insisted on in every case; even in the case of marriage with a deceased brother's wife the procedure was gone through, else the children were as "nothing," i.e., not legitimate. The feeling was that the marriage must be regular- ized by payment, else the contractants were the subject of odium, and furthermore the marriage would not be permanent. So strong was the feeling, that a lower-class man guilty of adultery with a chief's daughter or other close relation would be killed; not for his crime as such, but because he was too poor to make proper recompense in marriage. Even constructive adultery, such as accidentally touching another man's wife, was heavily fined. The insistence seems to have been on the regularity of the form, for it was freely acknowledged that the gifts made on either side were about equal. The marriage of very young children was arranged and carried out in full form. Chiefs and wealthy men marry women of other tribes, but marriage within the village was prohibited to no one save a blood relative.519 Much the same situation is implied for other Upper Chinook. 517 Gibbs, Tribes of Western Washington, 197. 518 Spinden, Nez Perc6, 250; Farrand, Notes on the Nez Perce Indians, 246. 519 Spier and Sapir, Wishram Ethnography. Spier: Klamath Ethnography From the Columbia river southward the exogamic practice extended among the coastal tribes as far as the Hupa and some form of the purchase feature to the Pomo. The caste sentiment of the northern tribes existed on the Oregon coast, with its concomitant local exogamy to maintain caste in marriage. This is implied for the Alsea, where there was "a decided preference for marriage with women from another tribe. This is explicitly stated of the nobility. At the same time there is evidence that the men did not care to go too far afield for their wives, for in such specific cases as could be cited, the favored tribes were the Yaquina and Siuslaw, whose languages are almost identical with Alsea and who regarded each other as closely akin. It is more likely that the exogamous tendency was local and extended to villages rather than to tribes. XX620 In general culture the Coos resembled the Alsea, hence it may be presumed that the same exog- amic sentiments prevailed. This is the more certain since the tribes to the south, the northwestern Californians, were also locally exoga- mous by habit. Coos marriages were apparently all made by pur- chase, often being those of children. The amount transferred was pitifully small in comparison with bride prices north and south. How far there was proprietary interest in the purchase, may be judged by the fact that the husband would demand the return of the bride price from the parents of an adulterous wife.521 While caste sentiment was not so definitely crystallized among the Takelma, distinctions were drawn between rich and poor, freeman and slave. "There is no well-defined rule against marriage within the village, but as it must very often have happened that practically all the residents of a village were related [marriage of relatives being forbidden], it was customary to look beyond the village for a mate, and in many cases even to marry into some neighboring tribe of alien speech, like the Shasta or the Galice Creek Athabascans." But it was as true here as to the north that such extra-local marriages were dictated by the necessity of finding a wealthy mate, for "the social status of the children depended very largely, of course, on the price paid for the mother, so that poor people's children were looked down upon as not much better than dogs .... . The indebtedness of the husband to his father-in-law did not entirely cease with the initial purchase of the wife. " He was obliged to visit him later with further gifts and to pay him again on the birth of the first child. " This pay- 520 Farrand, Notes on the Alsea, 242. 521 St. Clair and Frachtenberg, Traditions of the Coos, 26. 1930] 2 87 288 University of California Publicatioms in Am. Arch. and Ethn. [Vol. 30 ment was considered as equivalent to the buying of the child." Here too childhood marriages often occurred.522 East of the Cascades among the Klamath, social stratification was far more fluid. There were rich and poor but the caste feeling was only slightly developed. Indeed the sentiment may not be more than a century old, for the impression was derived that differences of wealth were slight and rank absent prior to the opening of direct contact with the Columbia river tribes. This does not date back of the opening of t.he nineteenth century. Marriage here was indifferently within or without the local community, but inasmuch as there was no senti- ment dictating marriage with one of equivalent wealth and rank, such exogamy as does occur is random and without purpose. The only prescription is that some payment must be made. The amount was relatively small and no great weight attached to its size as indicative of rank, or pretension to rank. The Northern Maidu have no general rule of marriage within or without the village. Of the Sacramento valley division however it is observed that "there would appear to have been a slightly greater pra.ctice of local exogamy.". The reason for this is not clear unless it be that co-villagers were commonly related, for these people have no feeling for distinctions of rank based on wealth.523 Whether the Southern Maidu had a similar exogamic tendency does not appear. There is however some simulation of purchase, since the girl's father received more than the boy's in the exchange of gifts.524 Local exogamy of a type affiliated with northwestern forms was not found in this direction beyond the Maidu. Exogamy did occur among the south-central Californians, but it was the exogamy of totemic moieties, which was local exogamy only to the extent that the totemic groups were localized. Thus it happened that the majority of Yokuts marriages were locally exogamic. The most northerly of such tribes were the Miwok.525 Since the habit of local exogamy was so slight.ly developed among them, it is doubtful that the Maidu case should be interpreted as an imitation of their practice. I should rather hold that it is to be talken as a result of the prohibition of mar- riage with relatives, hence not a reflection of either northern or southern influences. 522 Sapir, Notes on the Takelma, 268, 274. 523 Dixon, Northern Maidu, 239, 240. 524 Faye, Notes on the Southern Maidu, 36. 525 Gifford, Miwok Moieties, 141; Kroeber, Handbook, 457, 493. 190pier: Ktamath Ethnography To return to the coastal groups of northern California, the Yurok and their neighbors had attitudes toward social standing, wealth, and marriage appreciably like those of the Northwest Coast tribes. For the Yurok every social relation was nicely valued and the basis of them all was the marriage settlement. Social status was propor- tionate to a man 's wealth and the amount he was willing to pay for the mother of his children. His own life was valued at. the amount paid for his mother. Partial payment. or a small sum put him at the bott.om of the social scale. The price itself was definitely fixed. Marriage was outside the kinship, group and commonly outside the village, but not necessarily so. Kroeber represents the smaller villages as practicing exogamy by reason of kinship of their inhabitants, but it is at least as likely, in view of the exaggerated importance of wealth, that a sufficiently large bride price could be obtained in many instances only from men of other towns.526 An analogous situation may be inferred for the Tolowa, whose little known culture closely resembled that of the Yurok. On one fundamental point we have the verdict of the Yurok, who "regard the Tolowa as rich, a distinction they accord to few others of the people known to them." Inference also places the Ka.rok in the same category, for in all other known aspects of culture they were as one with Yurok and Hupa.527 The Hupa attitudes in the matters we are concerned with here were so nearly identical with the Yurok that to cite them would be but to repeat the Yurok case. At one point there is some difference, yet one only of degree. Our authority states briefly that women went regularly to their husbands' villages on marriage, implying that this was a uni- form pra.ctice, or nearly SO.528 Yet in actual practice it may mean no more than the Yurok habit, local exogamy based on kindred exclusion and the desire for a wealthy alliance. (If the marriages were as regu- larly exogamous as we are told, then I can see no rea-son why we should not describe this as a system of unnamed patrilineal local sibs.) The lower Klamath attitude is also that of the Shasta. It is essen- tial that a wife be paid for in order to fix the value of the children. Even elopements are condoned. if the bride price is paid later. Blood money for dea.th or injury is equal to the sum paid for a man's mother. The price varied widely but tends to be fixed at a value somewhat less than that of the Yurok. The wealthier families, who arrange infant betrothals, ordinarily choose the girl from another village. 526 Kroeber, Handbook, 28, 42. 527 Kroeber, op. cit., 98, 108, 126. 528 Goddard, Life of the Hupa, 54, 58. 1930] 289 290 University of California Publications in Am. Arch. and Ethn. [Vol. 30 Poorer men offer the bride price directly, or may serve their parents- in-law in lieu of payment.529 Purchase appears again among the Sinkyone, and perhaps pre- vailed in the intervening region. By no means so wealthy as their northern neighbors, values were correspondingly less. Nevertheless we observe something of the Yurok attitude in a definite value a.ttached to a man's life. A wife was seemingly paid for, alt.hough Kroeber suggests that this may have been in the nature of a gift to be recipro- cated rather than a formal purchase price. If this is so, it suggests only the practice of the generality of more southerly Californians. But inasmuch as illegitimate children were paid for, the suggestion is rather of an approximat.ion to the northwestern norm. Something of the same regularity of purchase, with reciprocal giving, is implied by Yuki statements.530 The Huchnom and Pomo attitude toward marriage did not differ appreciably from more southerly Californians. The exchange of mar- riage gifts can hardly be described as true purchase, but appears rather "an expression of good will and dignified but affectionate etiquette." Loeb observes, however, that the man's family gave more than it received in return. It would therefore not be possible to see in it certain affiliation with northwestern customs, were it not that both people prescribed a series of gift-bearing visits of the north- western type to consummate the marriage. I shall subsequently trace such formal visiting from the Kwakiutl southward to this point. As to exogamy, there was none; marriages were contracted indifferently within the community, outside it, or with foreigners. In neither group were there serious rankings based on wealth which might pre- scribe a studied attitude toward maintaining it in marriage.53' The typical central Californians (Wintun, Miwok, Yokuts, and Cost.anoan) register negative to the characteristics of northwest mar- riage. Social stratification of any range and rigidity was absent, wealth unimportant, mnarriage gifts were presents of friendship how- ever much formality demanded their bestowal, formal visiting seem- ingly unknown, and exogamy casual, certainly not for the express purpose of maintaining social position. In a word marriage does not function to uphold social stratification nor do the northwestern rites appear.532 529 Dixon, Shasta, 261 f. 530 Kroeber, op. oit., 146, 179. 531 Kroeber, op. cit., 210, 254; Gifford, Clear Lake Pomo Society, 310; Loeb, Pomo Folkways, 277. 532 Compare Kroeber, op. cit., 357, 456, 469, 493. Spier: Klamath Ethnography Marriage practices of the Northwest Coast type can be traced into the northern Plains. At least this appeals to me as the most rational explanation for the similarities which Crow, Hidatsa, and Assiniboin show and for their divergence from the general Plains custom. It will be well to define the latter again. Such typical tribes as Arapaho, Cheyenne, and Blackfoot viewed with favor marriages in which the son-in-law was chosen for his capacity as a man of honor, of war-like prowess, and a good provider. The father-in-law's desire was com- monly to secure his services. Gifts were exchanged, but rather for the honor of the act than with thought of definite purchase and the return of an investment. The anomaly of Crow attitudes is most intelligible, in view of their geographic position in the northwestern Plains, as ultimately derived from the Northwest Coast. The most honorable form of marriage for the Crow was by purchase, which gavre to the woman a position of respect beyond that of any of her unpurchased sisters whose matches were in contrast mere elopements. The young man presented horses to her male relatives and meat to her mother. There was no band exogamy here, although marriages may have occurred outside the group almost as frequently as among the Nez Perce.533 It will be observed that the Crow attitude is appreciably like that of the Nez Perce. That the Nez Perce placed a higher value on a mar- riage contracted because of a groom's capabilities, the Grow on pur- chase, is no argument against historic connection between the two. Both tribes recognize marriages of both forms; wherein they both differ from the majority of Plains peoples, is that they have marriage by purchase, and even stress it. As a matter of fact, the differenee may be really one of formulation on the part of the ethnographers. Lowie's account of the Crow is so circumstantially detailed as to leave no doubt of their attitudes, but Spinden's description of Nez Perce marriage is of the briefest. If he was impressed by the Plains- like character of the culture of this Plateau tribe, his formulation may well have stressed this non-northwestern marriage form. It may also be that the history of marriage in the two tribes has taken opposite directions. Possibly the Crow, recently acquiring the purchase habit, stressed the novel form as giving distinction to marriage; the Nez Perce, possibly with purchase as the original form, took over Plains marriage regulations with a deal of other culture and stressed what was the novelty from their viewpoint. 533 Lowie, Social Life of the Crow, 220, Notes on the Mandan, Hidatsa, and Crow, 74. 1930] 291 292 University of California Publications in Am. Arch. and Ethn. [Vol. 30 It is also conceivable that purchase a.mong the Crow is not related to the Northwest Coast forms, but that it is a derivative of their matriliny. That is, the desire to become affiliated with a particular matrilinear lineage may have been met by insistence on purchase, the groundwork for this being present in the common Plains custom of exchanging gifts. But there is nothing otherwise in Crow life to suggest that lineage as such was important. Among other Plains tribes possible analogies to northwestern prac- tice are found only among Hidatsa and Assiniboin. "The Hidatsa and Crow agree in considering marriage by purchase as the most honorable form from the woman 's point. of view; but among the Hidatsa I got. t.he impression, which I did not receive among the Crow, that it was somewhat discreditable for a man to have to buy a wife," for it implies a lack of personal qualifications. This attitude disposes of the suggestion made in the last paragraph, for the Hidatsa are also matrilinear. This is a duplication of the Nez Perce case; in contrast to the Crow, it may be said that here the novelty, purchase, if indeed it is a novelty, has not acquired the prest.ige found among the Crow. There is nothing in Lowie 's brief account to suggest formality in pur- chase, and in fact such procedure would appear quite at odds with their attitude. Matthews' statement confirms the absence of formal purchase. Nor is the Assiniboin case one of outright purchase. There are however features which suggest the Northwest Coast. As with other Plains tribes the marriage is accompanied by an exchange of gifts. Here the gifts of the bride's fa.mily were distributed among the groom's rela.tives, who return next day "returning to the donor some gifts of superior value to those received," these then being ca.r- ried to the girl's group. This is faintly reminiscent of the potlatch. Further, the first gift-bearing party was that of the bride, whereas the Plains habit at. large is for the young man's party first to make con- t.ribution. The impression gleaned from the account of the Assiniboin is a certain formality in the terms of settlement and in the marriage processions somewhat at variance with the generality of Plains mar- riage habits.5" Typifying these, Beckwith's brief statement concern- ing the Dakot.a implies only the exchange of gifts, although in one form of marriage the groom offers horses until the bride's father is satisfied with the amount.535 534 Lowie, Notes on the Mandan, Hidatsa, and Crow, 46, The Assiniboine, 40; Matthews, Ethnology of the Hidatsa, 52. 535 Beckwith, Customs of the Dakotahs, 255. Spier: Klamath Ethnography The Amount of the Bride Price Analogous comparisons may be made of the quantity and value of the bride price. This is somewhat difficult for lack of precise data, but such as exist make it clear that the largest amount was that of the Kwakiutl, with diminishing values toward the peripheries of the area. A further difficulty arises from the fact, that from the Columbia river northward we know only the amounts paid by members of the chiefly class. We are concerned here not alone with the gifts made by the groom, but with those given in return by the bride's relatives. It is not easy to give an adequate notion of the huge quantity of wealth which the Kwakiutl gave in the course of their marriage exchange. A few ra.ndom observations make clear that it was a vast amount; four hundred blankets and other valuables were presented on the groom's behalf, and in return the bride's father made gifts of larger value, coppers valued at thousands of blankets, his house and carvings, "a large stock of food, blankets, boxes, dishes, spoons, kett.les, bracelets, coppers, and the ceremonial box lids." In addition there were the intangibles, privileges, titles, and crests. Such were the valuables transferred on the marriage of the chiefly class, which rep- resent of course an. upper limit. The price to one of lesser rank was inevitably much smaller, but in any case a very real aggregation of wealth.530 Cont.rast with this the meager amount exchanged by the Coos, where the usual price was ten fathoms of beads, a couple of blankets, an otter hide or a canoe; and the Crow, horses to the bride's male relatives, meat t.o her mother. It may be well to brief all relevant data. The Klallam gave blankets, shell money, and several canoe-loads of food; on Puget sound, gift.s of like value, including a fine canoe in one instance, in another forty blankets, two slaves, and a canoe, and in a third, five horses and $50 in cash. The Twana gave money, blankets, guns, horses, etc., to a total value of $100 to $400. The sa.me lavishness, although moderate in comparison with Kwakiutl, is implied for the Chinook. The Wishram gave horses, packs of blankets, etc., and kept the dishes of the marriage feast. In return they received baskets, beads, slaves, and a good horse. A trunk full of clothing and other finery was exchanged. 536 Professor Boas tells me that he was never able to get an account of a commoner 's marriage. 1930] 293 294 University of California Publications in Am. Arch. and Ethn. [Vol. 30 South of the Columbia river the quantity was quite moderate, even where its value in native eyes was high. For the Tillamook, the price was generally one adult and one child slave. The Coos value has been cited above. A chief's daughter came higher, being usually bought with woodpecker scalps, valuable for their use in dance caps. Such scalps formed valuables from this point to northern California. A Takelma paid dentalia and the like in moderation; the return was baskets, women's ha.ts, camas, dried salmon, and other such household articles. The impression I derived among the Klamath was that the quantity was somewhat greater, approaching somewhat Wishram prac- tice. A chief ga.ve a slave, horses, blankets, beads, food, elk and other hides, but not nearly so much as my Wishram informant implied. One poorer man gave five fathoms of beads for his bride. Values drop appreciably among the Maidu; the northern groups made a gift of beads, deer, or gave nothing; the southerner gave shell money or other valuables such as blankets, and in return obtained baskets and other property. On the northern coast of California values rise high again. The Yurok price was standardized. For a wealthy girl, ten strings of beads, fifty woodpecker scalps, an obsidian blade, a boat, etc., up to $300 in value, were given, with a return of ten baskets of dentalia, otter skins, a canoe or two, several deerskin blankets, and other valu- ables. A poor girl was valued at eight strings of beads and a boat. Hupa values may have been the same or somewhat less; Karok cer- tainly were, five to ten strings. The Hupa groom gave woodpecker scalps, several strings of shell money, to a value of $30 to $100. Pre- sumably he also gave articles of the kind returned to him, dresses, weapons, baskets, and food. The average Shasta price is set at one or two deerskins, fifteen or twenty long dentalia, ten or fifteen strings of disk beads, and twenty or t.hirty woodpecker scalps. Values dwindled again markedly south of here; the Sinkyone price was less than that of Yurok or Hupa, shell beads alone are mentioned for the Huchnom, and presumably neither group gave more, or much more, than the Pomo. The Pomo gave beads and deerskin and in return received baskets, or, according to another account, gave food, beads, deerskin or rabbit-skin blankets, receiving in return baskets, belts, and food. Eastward of the Columbia river va.lues were again low, even if measured in terms of the lesser quantity of valuables possessed by Spier: Klam.ath Ethnography these people. The Nez Perce gave a horse, elk teeth, and similar articles; the return gift comprised bags, food, and the like. Hidatsa and Assiniboin prices were similar to those of the Crow noted above. A Hidatsa man gave horses to the girl's parents, who reciprocated with as many or sometimes more; an Assiniboin gave one or more horses, received pemmican and twelve pairs of moccasins, and then gave somewhat more than he had received. The impression derived from these data, despite their paucity, is that the bridal payments were made on a lavish scale as far southward and eastward as the Columbia river peoples. Beyond this they were meager in comparison. The region north of the Columbia was one in which material possessions bulked large. Wealth was a matter not alone of relative values but of quantity. To be sure the material wealth of the Washington tribes was by no means so great as that of the Kwakiutl. I get the impression that there the bride price was high, relatively as well as quantitatively, in comparison with the southerners. A second focal area of relatively high values appears again in northwestern California, as Kroeber pointed out, rising above those of coastal Oregon and the region to the south. It is not easy to define relative value, but it does appear when we realize that the material possessions of the Pomo and the Crow cannot have been far different from those of the Yurok. The higher values, both quantita- tively and relatively, may be viewed as a corollary of a social struc- ture dependent for its standards on wealth. We note incidentally the articles that constituted valuables in bride purchase. Woven blankets (in the historic period, trade blankets) figure as far south as the Oregon coast, dentalium shells beyond to the south as far as Sinkyone and perhaps Huchnom, and southeastward to Klamath and Shasta. Woodpecker scalps are part of the marriage price in California, extending northward to Coos on the coast, to Shasta in the interior. The significance of this restricted distribution, for example, is that the Klamath made use of woodpecker scalps for headbands in California fashion but never mentioned them as having money value. Horses are of course associated with the Plains forms of purchase, but are mentioned as far northwestward as Klamath, Nez Perce, Wishram, and southern Puget sound. To be sure the horse was not carried beyond Puget sound, or there were few until after 1850. 1930] 295 296 University of California Publications in Am. Arch. and Ethn. [Vol. 30 Marriage Visits One of the most characteristic features of marriage in the north- west was a, series of reciprocal visits. These visits were more or less formally prescribed as to number and were always the occasion for making gifts in fulfillment of the contract obligations. For the more. northerly groups, they were avowedly potlatches. The peripheral groups, who did not participate in the potlatch system, had at least some approximation in the marriage ceremony. To be sure, visit.ing back and forth carrying gifts after marriage was the habit of the Californians at large, a.s well as others. The peculiarity of this series is, however, that the visits were more or less formally prescribed and st,ereotyped as to time and conduct. Their number among the Kwakiutl is not described; it may not have been fixed. But it is clear that the father-in-law visited the groom's people on several occasions to return an equivalent, of their payments and to deliver the privileges, honors, etc., due them. This was by their own account potlatching; indeed the emphasis is so much on t,his a,spect that a fiction might be substituted for an actual mar- riage. The return of the property was several-fold; the groom dis- tributed the articles among his followers who had previously lent him the amount needed to make the purchase, and so the property was once more started on an interest-bearing round. Klalla,m marriage was just as clearly a pot.lat.ch. "After six months or a year the bride's family visits the groom 's relatives, bring- ing food and gifts. They are expected to bring at least as much as the groom's people brought to the wedding. Generally they try to bring a, little more. 2 537 We may presume the distribution of these in regular potlatch fashion. As potlatching is also the custom of other western Washington tribes, we may assume that it also appears in their marriage settlements. It is t,rue that we are told that those of Puget sound were not pot.la,tches, but they do not differ one whit. from the Klallam case.538 Further, after the exchange of gifts the parents of the couple loaned each other property, expecting to get it back some time. The reciprocating visit of the bride's parents took place-a few weeks or months after the marriage. 537 Gunther, Klallam Ethnography, 244. 538 Haeberlin-Gunther, Ethnographische Notizen, 35. Both authors, who had little experience with Northwest Coast culture, clearly misunderstood the impli- cations of their information. What may have been meant was that this was not a potlatch for the sole purpose of getting a return. Spier: Klamath Ethnography Whether the Chinook had an extended series of visits I do not know, but there is the same formal appearance of a gift-bearing messenger of the groom and the reciprocating visit of the bride 's party conducting the bride to her new home. Boas' informant set their gifts at slightly less than the bride price, but Franchere would have it that even more was repaid, depending on the means and munificence of the bride's parents. The Wishram case was just as clearly one of potlatching; that is, the gifts must be reciprocated in superior amount else her parents lost caste; they were given with a lavish hand through the medium of a spokesman, participants were paid for their services, and the goods distributed. among the members of the parties. The visits ran in this fashion: a messenger conveyed the groom's gifts, the groom's party proceeded to the bride's home where they were feasted, they distributed their further gifts and were recipients in turn. The same form was followed in transferring these later gifts. After matri- local residence of a few weeks, a series of reciprocal visits began which ended in permanent residence among the man's people. Two visits apiece terminated the affair. From this point southward we do not seem to be dealing with the same affair, rivalry in lavish giving, although the series of visits per- sists. The Klamath also sent an emissary, the bride's party appeared at the groom's house, and the exchange followed. After a month the couple,- with the groom's relatives, proceeded to the bride's former home, when an exchange again took place. Soon after the bride 's rela- tives repeated their visit. Beyond this point visit.ing in this fashion was not fixed but occurred as an expression of mutual solicitude. The Shasta procedure was much curtailed in contrast.. The girl was taken to the groom immedia.tely after her puberty ceremony. Some time la.ter her relatives visited to exchange gifts; after severa.l months they were recipients in turn. Below this on the Klamath river postmarital visit- ing did not appear as a fixed form. Something of the sort appears among the Huchnom. The kinsmen of the groom carried gifts to the bride's home, where the groom then stayed. "The return of the hus- band to his own people was made the occasion of gifts of food and beads to the pair, which they gave to the wife's parents, besides enter- taining their kin. A return visit to the wife's people followed, after which the couple usually founded their own home. This was always among the group or in the town. of the wife."5 Precisely the same visit.s are noted for. the Pomo, and in addition the procedure was repeated several times, particularly if the couple's original homes had 539 Kroeber, Handbook, 211. 1930.] 297 298 University of California Pubication in Am. Arch. and Ethn. [Vol. 30 been at a distance. This sort of formal visiting did not occur to the southward so far as I know. To return to Oregon, the Tillamook sent a messenger who brought back an equivalent gift. A week later the groom's party went to make the purchase, when they also received gifts. " Once every month this exchange of -visits was repeated. "540 Something of the sort held among the Takelma, for the son-in-law was obligated to visit his parents-in-law with presents from time to time. We must not believe that the potlatch sentiment was entirely extinguished in this direction. Farrand was impressed by the Alsea practice as suggesting "in certain ways the potlatch system of Van- couver island, for there was an apparent effort on the side of each family party to the contract to keep the other family in debt to it. "541 Coastal Oregon seems to be the limit to which such sentiment extended; it was certainly not the thought of the Klamath or the Californians. In the direction of the Plains, potlatching did not appear unless we so construe the fact that among the Nez Perce the gifts made by the bride's people were somewhat greater than the groom's. The visits of these people consisted in conducting the bride to the groom's home, with a return visit a month later. Crow and Hidatsa knew nothing of this formal visiting. The Assiniboin procedure however resembled that to the northwest. A messenger was dispatched with purchase presents, the girl's party went to the groom, and soon after the groom's people returned the visit. It thus appears that the marriage ceremonial partook of the nature of a potlatch in the northwest, but that beyond the Columbia river the practice had dwindled to an attenuated copy of the potlatch visiting without its sentiment of rivalry. One further point in this connection should be noted. The first gift-bearing party is sometimes the groom's, sometimes the bride's; the distribution of the practice is not random. Among the Kwakiutl, Klallam, Puget sound tribes, and Wishram, it is the groom's party who go to the bride. Among the Chinook, Klamath, Shasta, Hupa, Yurok, Nez Perce, and significantly the Assiniboin, the bride's party first approaches the groom's group. The data on the Tillamook are not clear and for others wanting. Huchnom and Pomo, Crow and Hidatsa, conform to the generality of Californian and Plains practices in that the groom carries his gifts to the bride's parents. 540 Boas, Notes on the Tillamook, 7. 541 Farrand, Notes on the Alsea, 243. Spier: Kkamath Ethnography Family Compacts To revert to the nature of the marriage contract, this was in some instances avowedly a compact between families. It takes two forms. When a man had paid for a wife and her family had made a settle- ment on him, one of them had a grievance if the marriage was dis- rupted by the death or desertion of the other. It appears again in the assumption that children are a natural fruit of the union; that the husband must be indemnified if there were none or the woman's family absolved from payments for every child born. These features can best be brought out by considering the levirate, sororate, divorce, and other relevant dat.a. It is exceedingly difficult to disengage the difference in meaning between the statements that the levirate, e.g., was compulsory or cus- tomary. So much must have depended on the precise personal cir- cumstances (the lack of appropriate relatives, the ages of the persons concerned, their sentiments, antipathies, etc.) that even where the custom was so regular as to be styled compulsory, it may nevertheless have never been more than a customary practice. The best that can be done is to discriminate those cases where the sentiment was an obligation to marry from those where it was merely good form. It should also be borne in mind that while the levirate was always of one kind, namely the marriage with a deceased husband's brother (or cousin), there being no polyandry in North America, the sororate included two different social situations. In one form there was the analogue of marrying a deceased wife's sister; in the other, of taking sisters as plural wives, in particular taking them to wife as they came of age. While the two forms occurred together, it was not so by necessity. They should not be confused. The sentiment of mutual obligation was perhaps strongest among the Yurok. If a wife died early in marriage and sometimes even in later life, a sister or kinswoman was due from her family; similarly she was obligated to marry his brother on her husband's death. A divorce was possible only if each party was willing to make full resti- tution. If the wife desired to take the children with her, it could only be done if the full marriage price was returned. Analogously "each child left with the husband reduced the repayment, and several can- celled it altogether. " Such contractual relations appear as a function of the regularity of purchase. How far that sentiment held may be 299 1930] 300 University of California Publicationm in Am. Arch. and Ethn. [Vol. 30 judged by the fact that the husband demanded payment for adultery or act-ions construed as adulterous, a practise unthinkable to the majority of Indians.542 The attitudes of the Hupa were quite analogous.543 Mutual obligation of the same sort is described for the Chinook. Leviratle and marriage with a, deceased wife's sister were a matter of course. The widow may be taken to wife by her husband's father or another relative in default of a brother. A man must be indemnified if his deceased brother's wife is carried away or he would kill the abductor. Similarly a man was indemnified with many slaves if his wife was abducted, else he had revenge. Children were as much part of the bargain as among the Yurok; if a child died the wife's father was obligated to pay a slave or a canoe as indemnity.544 The family contract did not differ in kind among the Kwakiutl, but had another emphasis. The members of a chieftain's tribe were jointly concerned in the purchase; they speak of "our bride"; the father-in-law says, "All the tribes came to marry my daughter." These participants were relatives of the groom, of near or remote degree. I cannot find any reference to the levirate and sororate among them, although it ought to have followed from this attitude. Here, even more emphatically than among the Yurok, children were an essential part of the compact. What is b-ought is "the right of membership in her [the wife's] clan for the future children of the couple. " Indeed the number of children fixes the rate of repayment by the father-in-law.545 The feature of the families as contractants appears in other cases but I doubt that the sentiment is so emphatically expressed as in those cited above. The levirate was well-nigh compulsory among the Wish- ram, and the sororate in polygamous unions followed, but not always. The widow married the husband 's brother or a cousin, unless he did not desire an additional wife. If she married any other, which was per- missible, he had to make payment to her dead husband's kin, not to hers, so that they would look on him as one of themselves. While I received no suggestion from my informants that children were part of the marriage bargain, there is an analogue pointing in this direction. A few months after the birth of each child, the husband's people, with the couple and the newborn., visited the wife's parents bea.ring gifts. 542 Kroeber, Handbook, 30 f. 543 Goddard, Life and Culture of the Hupa, 55, 56, 58. 544 Boas, Chinook Texts, 251, 252, 257. ra Boas, Social Organization and Secret Societies, 358 f. 4~ 1930] Spier: Kldmath Ethnography 301 The latter reciprocated by visiting with gifts for the husband's father. Here, too, indemnity was demanded for adultery or constructive adul- tery. Shasta data also indicate a contractual relation between the families; the levirate was compulsory and a sister must be furnished for a. deceased wife or an adulteress. So firm is the contract conceived that a man who carries off a widow without compensating her hus- band's family stands in danger of his life. Parallel to the Yurok case, the husband who made no payment for his wife had no redress for adultery. Perha.ps the contractual relation was as strong among the Puget sound people but the data are insufficient to support the view. A Nisqually widow was obliged to marry her husband's brother or cousin, and a widower married his wife's sister. But if the widow did not marry among his relatives, she usually returned to her own people with her children. Among Snohomish and Snuqualmi, if a woman abandoned her husband or was divorced for good cause, her people had to return the bride price. Klallam expression is much the same; a widow was expected to marry her brother-in-law or a near relative, a woman 's family to offer a sister or relative of hers on her death. Similarly a wife's parents were expected to offer repayment of the bride price if she left. her husband, but it was bad form for him to accept. it. The Klamath attitude is at some remove from this one of mutual obligations. The levirate was customary but not obligatory, and the sororate permissible but no fixed custom. The widow was under some obligation to her husband's brothers, who might object should she marry another, but this seems rather a matter of sentiment than right. The feeling was that she, a member of their family and even an inmate of their house, should remain one. The reaction to a wife's adultery is the common one of revenge on the lover, but with no thought that the husband's rights to the wife had been damaged. Throughout a wider area the levirate and sororate were both prac- ticed, but nowhere do we find them of the obligatory nature described above. It will suffice to list these cases. The levirate was customary among the Tillamook. (Perhaps if we knew more of this people this would transpire to be another obligatory case, considering their proximity to the Chinook.) Among the Achomawi it was perhaps regular ;546 the Sacramento Valley Maidu, usual; the Northwestern Maidu, general; occurred sometimes among Yuki and Pomo (northern, southea.stern, and eastern divisions, although of the latter Gifford says 546 Kroeber, Handbook, 313. 302 University of California Pu6lications in Am. Arch. and Ethn. [Vol. 30 they did not practice it) .547 Kroeber states that levirate and sororate were followed everywhere in California except among the Colorado river tribes.548 I have found specific statements to except from the levirate the Yuma, Akwa'ala, Cupenfo, and perhaps Luisefio.549 Further the Huchnom, in an area where it generally held, avoided the levirate. In the Great Basin it was customary among Paviotso, Moapa, Shivwits, and Wind River Shoshoni; in the Plains, Crow, Assiniboin, and Hidatsa.550 How frequently the practice was indulged in these cases is not clear. The Hidatsa, for example, held the levirate proper but marriages of this sort were uncommon. Curiously enough these practices seem to be absent on the North- west Coast precisely where one would expect contractual relations of the families to be strongest. I find no mention of them among the Tsimshian and Kwakiutl. Boas recorded only one Tsimshian tale where wives were sisters.5'l The Chilcotin will have nothing of the levirate, for the widow "is still regarded as his own sister" by the brother-in-law.552 On the other hand, the levirate was common among the interior Salish.553 Certain of the Plains tribes apparently lack it: Blackfoot554 and Gros Ventre. The sororate occurred throughout much the same areas. In Cali- fornia, I note specifically the Northeastern Maidu, Northern and Eastern Pomo, Northern Costanoan, and Southern Dieguenio; in the Basin, Paviotso, Southern Paiute, Uintah Ute, and Wind River Shoshoni; in the Plains, it was the regular practice, commonly taking the form of polygamous marriage with sisters, viz., Crow, Blackfoot, Assiniboin, Hidatsa, and Iowa.555 Again no mention appears in the account of the Gros Ventre. Levirate and sororate are customs so widespread in North America as to give us little concern here, were it not that they acquire special significance in connection with marriage by purchase. I have tried here to make out a case that in the coastal region from the strait of Juan de Fuca to northern California they have been absorbed in a 547 Clear Lake Pomo Society, 319. 548 Op. oit., 839. 549 Spier, Havasupai Ethnography, 254. 550 Matthews, 53. 551 Boas, Tsimshian Mythology, 535, 422. 552 Farrand, Chileotin, 18. 553 Boas, Salish Tribes, 222. 554 Wissler, Social Life, 12, 13. 555 Kroeber, op. oit., 469; Spier, Southern Diegueflo Customs, 311; Matthews, 53; Skinner, Societies of the Iowa, 739. Spier: Klamath Ethnography special configuration of social relations most aptly described as a set of mutual obligations constraining the families of marriage contractants. The Klamath, by reason of their geographical position marginal to these, show the effects of this to only a moderate degree. Postnuptial Residence Residence after marriage was prevailingly patrilocal in the north- west, the bride proceeding at once to her husband's home. This held from Alaska to northwestern California; Tsimshian, Bella Coola, Kwakiutl, mainland Cowichan, Lillooet,556 Klallam, Puget Sound Salish, Twana, Coos, Takelma, Hupa, and Yurok. The Yurok and Hupa cases indicate that this mode of residence is derived from the fact that brides were purchased; there if full payment could not be met, the husband had to be content with joining his wife who would not be given up by her people. I am not asserting that in general patri- local residence following promptly on marriage is due to purchase; in these northwestern tribes this seems to be the case. It is conceivable that they might have held that a man must furnish services before they would release the bride to him, but this was contrary to their feeling respecting wealth. Services were a poor substitute for mate- rial wealth. Even in those cases where temporary matrilocal residence did occur this sentiment was absent; it was simply customary practice. The Klallam sometimes resided with the wife's people for a short time; among the Wishram it was apparently the regular practice. The Tillamook had a curious alternation; first matrilocal, then patri- local, each for a few days, settling down to matrilocal residence, which however they might change later. The Klamath conformed to the northwestern practice. Residence was normally patrilocal, the bride being brought at once to her hus- band 's home; but, as with the Yurok and elsewhere, a man too poor to pay tendered services while living at his bride's home. The patri- local habit was by no means rigid; matrilocal residence might set in from the outset if it was desired. The relation of the Klamath to the northern tribes can best be brought out by pointing out their contrast to those to the south. Tem- porary matrilocal residence was the custom of a large part of north central California; Achomawi, Atsugewi, Maidu (northeastern, south- 556 Boas, Tsimshian Mythology, 421, 424, 441, 532, Third Report, 12, Social Organization and Secret Societies, 358 f., Lower Fraser River, 5. 1930] 303 304 University of California Publiccatiois in Am. Arch. and Ethn. [Vol. 30 ern, and the foothill groups), Eastern and Northern Pomo. There- after it was not of necessity patrilocal, for residence might be con- tinued with the woman's uinit. Outright matrilocal habits appeared among the Sacramento Valley Maidu (although the majority of mar- riages were there patrilocal from the outset), Huchnom, and Yokuts. Southern California was again an area of patrilocal habits. Great Basin practice conformed with that of north central California; tem- porary matrilocal residence for Paviotso, Southern Paiute (?), Uintah Ute, Lemhi and Wind River Shoshoni. Northern Plains tribes were generally patrilocal: Crow, Gros Ventre, Assiniboin. Matrilocal residence as well occurred among the last two. Hidatsa indulge in temporary matrilocal residence, after which there is no fixed rule. Summary We may now review the data detailed above. Sharply crystallized distinctions of rank and the importance of wealth in establishing them, wealth itself assuming real proportions, center in this area in the Kwakiutl and other northern tribes. Similar caste distinctions are traceable southward on the coast to the Alsea and Coos, and east- ward not beyond the Wishram. Social distinctions based on wealth to the almost complete exclusion of other factors extend beyond these limits; in attenuated form to the Sinkyone and Yuki and perhaps beyond to the south, to the Klamath and Shasta, to the Middle Colum- bia Salish of eastern Washington. Marriage by purchase had an even more extended distribution; from* the same center it is traceable southward perhaps to Huchnom and Pomo, to the Southern Maidu, and eastward into the northern Plains, if my interpretation of the Crow, Assiniboin, and Hidatsa cases is justified. In the northwest- ern portions of this territory marriage by purchase is a function of the caste-wealth system; in the outlying regions, as among the Maidu on the one side and Nez Perce, Crow, etc., on the east, it would seem that we witness the spread of the purchase factor with little of these underlying concepts. It must be borne in mind that nowhere in this area are the marriage payments justly described as purchase; there is always a reciprocation of the gift. The express point is that mar- riage is contracted only by the making of such gifts. Wealth was very real, huge in quantity, north of the Columbia, and a correspondingly high value set for a marriage settlement, at least among the upper classes. Valuables and marriage values dwin- Spier: Klamath Ethnography dle beyond this limit, with a notable recrudeseence of the latter on the lower Klamath river. Part of the mechanism of its exchange in marriage was a series of reciprocal visits, of potlatch character south- ward to the Columbia, a mere formal practice beyond. Among the concepts underlying the practice of bride purchase was the sense of mutual obligations of the contractants' families. This was avowed by the coastal tribes from British Columbia to northern California. It assumed the express forms of compulsory levirate and sororate and of the right to children. Further, the fam- ilies were obligated to restore the marriage settlements if the marriage terminated unfortunately. The man's financial interest in the wife he had bought was evidenced by his ability to mulct damages from an adulterer or one even thought to be such. The exogamy which appears in this area was not the prohibition of marrying within the community, but an exogamy of practice. To maintain caste, marriage must of necessity have frequently been with persons of the same rank and wealth outside the village or tribe. This was undoubtedly more the circumstance in the upper strata of society. It is obvious then that the distribution of this practice coin- cides closely with that of sharp distinctions of caste and wealth. Beyond those tribes whose attitudes may be so characterized, there was a fringe of peoples who had taken over the custom without its antecedents. Such local exogamy has been noted as far southward as the Hupa, the Sacramento Valley Maidu, and as far eastward as the Nez Perce. The habits of postnuptial residence seem to correspond closely with the sense of purchase. It was patrilocal, the bride being taken at once to her husband, in the coastal area southward to northern California. Some variability does indeed appear but this seems no more than the common circumstance of residence according to pref- erence. The Wishram instance of regular temporary matrilocal residence is an anomaly. The position of the Klamath with respect to this northwestern marriage complex was that of a distinctly marginal group. Pur- chase to maintain social standing was present but the values not fixed; the actual amount changing hands, relatively small. A series of for- mal gift-bearing visits was prescribed, in which the bride's family returned more than they received, but there was no intentional rivalry in giving. The first visit was that of conducting the bride to the groom as among all the southerly peoples. Residence was patrilocal 1930] 305 306 University of California Publications in Am. Arch. and Ethn. [Vol. 30 in conformity with northwestern views. The sense of obligation was weakly developed; the levirate was a mere custom, the sororate rare; birth of heirs played no part in their scheme of things. In brief, the Klamath derived the sense of social standing based on wealth from the northwestern center, and the mechanism of marriage purchase to effect it, but no more. I am not inclined to indulge in an historical reconstruction of the northwestern marriage complex. But if it be in order, it seems better to suggest that the ancient, undifferentiated custom was the practice of mutual giving, subsequently formalized with the growth of con- cepts of wealth and rank on the Northwest Coast. The alternative is to hold that these attitudes and practices diffused from that center only after they attained full flower there. If this were so we should expect to find the specific forms of Northwest Coast procedure at a greater distance. I am very doubtful that ideas and sentiments dif- fuse as such, though I may have given that impression in some of the summary statements above. It seems rather that particular practices are imitated, and if but imperfectly copied or made rational in their new context, then alone do we say that merely the underlying idea has been borrowed. In the present case social stratification based on wealth had spread over a wide area; wife purchase with it. I conceive that members of the imitating tribes observed the aggrandizement that went with wealth and took occasion among their fellows to insist on a similar, and in their eyes proper, respect for themselves and their wealth when opportunity offered. The spread of the marriage purchase cus- tom must have been one of the most effective mechanisms for the spread of the sense of rank. Boas has already made such a case for the Kwakiutl and their neighbors. By marriage with the northern tribes among whom privilege concepts were highly developed, the Kwakiutl obtained these privileges through their wives.557 The mechan- ism is clearly transparent in the Klallam preference for marriages to the north.558 We are not told this of other tribes, to be sure, but it appeals to me that marriage into foreign tribes where payment for their women was demanded would inevitably bring in its train the notion that these women were in some way superior to those at home. This view is not drafted from thin air; among the tribes of Puget sound, e.g., the women of the more northerly tribes consider them- 557 Boas, Social Organization of North-Pacific Coast; Social Organization of the Kwakiutl, 124. 558 Gunther, Klallam Ethnography, 242. 19]Sper: KMamath Ethnography selves the superiors of those among whom they are resident. Once the sentiment had become lodged in the group its fruition was made inev- itable by the imitation of others. With Kroeber, who has so expressed himself on the subject of the Yurok, 59 I do not believe that the levirate and sororate were born of this special development of purchase. Both customs have a nearly universal distribution in North America. It seems rather that such existing practices were drawn into the developing marriage complex of this area. SOCIAL STRATIIICATION The background for a study of social stratification has been laid in the preceding section. I will here deal with chieftainship, its relations to wealth and heredity, and related topics. Social stratification was but weakly developed among the Klamath. Social standing did not depend directly on heredity and there was apparently little difference in wealth from richest to poorest. Since wealth was inherited there was at least some stability to family stand- ing. The marriage purchase gave respectability to the union; it did not fix the standing of the couple by reason of the amount paid. Yet there was some differentiation based on wealth, a continuous gradation from the wealthiest to the slaves. The wealthiest individuals were called lo'ki, rich men. This was also the term by which chiefs were styled. My interpretation of the Indian's statements is that those prominent individuals, who exer- cised functions which would elsewhere have led to their being called chiefs, were at the same time the wealthiest. A chief became such primarily by successful leadership in war, coupled with some back- ground of spirit power, wealth, personal qualities, especially ability as an advisor, with some suggestion of hereditary interest. The chiefly functions emphasized were leadership in war and haranguing the people. The chief could not issue injunctions, he re- ceived no tribute, and was in no way responsible for the conduct or debts of his people. Just prior to reservation days there was intro- duced from Columbia river the practice of chiefs sitting as judges. While chiefs were set off from the mass, this did not imply any sense of caste stratification. Their wives and children were not known by a distinctive class term. 559 Handbook, 31. 1930] 307 308 University of California Publications in Am. Arch. and Ethn. [Vol. 30 I Every evidence points to the recency of the rise of chieftainship among these people. It is doubtful that material differences in wealth and slave-holding existed prior to the opening of direct contact with the Columbia river tribes after 1800. Even then chiefs were of rela- tively minor importance. The shamans were the great men and the wealthiest, easily eclipsing the chiefs. As an instance I have cited the difficulty of obtaining the names of chiefs of pre-reservation days, whereas shamans are well remembered. It is clear that some of the local groups did not have chiefs. The relative dominance of the sha- man persisted until reservation days. Prior to this time chiefs were barely differentiated leaders of the Great Basin type. Basis of Chieftainship The Northwest Coast tribes stood in sharp contrast to this simple social scheme. There existed a caste system, a division into aristocracy, commoners, and slaves. While the relative ranking was continuous from the noblest aristocrat to the meanest commoner, there was never- theless a well defined consciousness of distinction between the grades. This aristocracy was based on lineage and wealth. Its highest mem- bers were the chiefs, ordina.rily the wealthiest persons. Yet a chief might lose wealth but retain his rank by reason of his birt.h. Within each community were several chiefs of the first rank whose authority extended hardly beyond their immediate retainers. Their advice was followed rather than their dictates obeyed. They were, however, among the Kwakiutl, for example, the recipients of gifts from their followers, not so much by reason of right as to bespeak their future aid and to enrich them to the further glory of the whole group.560 A definite caste system of this type is traceable southward from British Columbia through coastal Washington and Oregon at least as far as the Alsea. It has been noted, for example, for Klallam, the Puget sound tri.bes, Chinook, Wishram, and Alsea;. It did not appear in the interior among the upper Fraser river tribes, nor in eastern Washington as among the Okanagon and MIiddle Columbia Salish, nor in eastern Oregon among Klamath, Modoc, and Northern Paiute. At, the southern extremity of this area of distribution a somewhat different situation preva.iled. There was no rigid caste system, but differences of wealth were considerable and on these social standing 560 See Sapir, Social Organization of the West Coast Tribes, 360; Boas, Social Organization and Secret Societies, 338, Ethnology of the Kwakiutl, 1333 f., Social Organization of the Kwakiutl. 0Spier: Klamath Ethnography rested. This held at least for coastal Oregon and northwestern Cali- fornia (Coos, Yurok, Hupa) while inland among Takelma and Shasta it was true to a lesser degree. The Klamath feeling for stratification based on wealth is thus intelligible as an attenuated form. It, follows that chieftainship wa.s hereditary in the area in which ranking was fixed by birth as well as wealth, while it rested on wealth alone in the southern area. Merely to list the instances, it was heredi- tary among Klallam, Snohomish, Chinook, Wishram, Wasco, and Alsea (but see below). The richest men among the southern tribes were the equivalent: Coos, Yurok, Hupa, and Shasta, and, if we reverse this statement., the Klamath. It is thus intelligible why Sa.pir found among the Wishram that "being a chief did not necessitate being a particularly wealthy man. It often happened that the wealth- iest. people did not belong to the highest rank."56' Similarly that a wealthy Coos could flout the orders of a chief, and among the Alsea "it. was possible . . . . for a common man, by reason of extraordinary power or wealth, to rise to the dignity of a chief and thus raise his family in rank. '562 There was in none of these tribes a governmental organization involving hierarchy of chiefs. The point is rea.lly that what we here write of as tribes were rather dialectic and cultural units, not polit- ical entities. Political sepa.ra.tion was carried to a high degree through- out the Pacific slope, the local communities being almost without exception autonomous, or at most grouped. for common action within a quit.e restricted district. It followed from this that the number of chiefs was rarely or never fixed. Those who were entitled to leader- ship by virtue of birth or wealth were recognized as leaders. Thus it happened that while some communities had more than one of first rank, others had none. Yet other tribes did not always exhibit the haphazardness of the Klamath in this regard. Where the feeling was strong that every wealthy man was ent.itled to consideration, as among the Hupa, "each village had a head-man who was the richest there .... . The very fact of his wealth gave him the power of a chief and maintained him in that power."563 We may presume this true of the Yurok as well and probably the Karok. The Shasta "chief was the head of the richest family in the district," 561 Spier and Sapir, Wishram Ethnography. 562 St. Clair and Frachtenberg, Traditions of the Coos, 25; Parrand, Notes on the Alsea, 242. 563 Goddard, Life and Culture of the Hupa, 58. 1930.] 309 310 University of California Publiations in Am. Arch. and Ethn. [Vol. 30 and "one family was recognized as preeminent in each of the four geographie divisions into which the Shasta are grouped."564 The number of chiefs was definitely fixed only on the Oregon coast, where there were two for each local division or village. "There were two chiefs for every river, among the Tillamook as well as among the Alsea," one the principal chief.565 "The Coos Indians lived in small villages, each of which had two chiefs .... . Of the two village- chiefs, one usually ranked higher." There was in addition a tribal head chief.586 This is the multiple chief-wealthy-man situation of the whole coast reduced to definite form, and it is even conceivable that these people did not differ at all, particular cases where there were two chiefs having been cited as the general custom. It must be remembered that our accounts are of the slightest. It is possible however that there was a dual chieftainship prevailing on the Oregon coast. This might account for the curious fact that two chiefs of presumably equal rank are repeatedly referred to among the Rogue river Indians assembled at Table Rock about 1855 (possibly Takelma proper or Upland Takelma).567 Quite another basis for the recognition of chiefs existed in the northern Basin-Plateau and the Plains, namely, a war record coupled with generosity and other personal qualifications. The Klamath shared this sentiment as well as the northwestern recognition of wealth. The warlike character of chiefs extended as far west as the Middle Columbia Salish, including Nez Perce, Wind River Shoshoni,565 and such typical north Plains tribes as Crow and Assiniboin. In the southern parts of the Basin the characteristics of chiefs, their functions and powers, were vague. There were headmen or leaders for hunting, perhaps the only communal enterprises, among Paviotso, Washo, and Southern Paiute. These individuals were chosen by their people.569 In importance they were eclipsed by the shamans. 564 Kroeber, Handbook, 296. 565 Boas, Notes on the Tillamook, 4. 566 St. Clair and Frachtenberg, Traditions of the Coos, 25. 567 See Walling, History of Southern Oregon, 182. 568 Teit, The Middle Columbia Salish, 126; Lowie, Notes on Shoshonean Ethnography, 283. 569 Lowie, op. Cit., 284-285. Spier: Klamath Ethnography Chiefs' Functions The functions of chiefs both to the northwest and in central Cali- fornia differed sharply from those of the Klamath. On the Northwest Coast proper the authority of a chief was by no means absolute. The commoners were subject to his admonition rather than his will, his direct authority extending hardly beyond his immediate following. Such gifts as they made him were free-will offerings, large as they were, dictated solely by traditional sentiment. While these include almost all articles of food among the Kwakiutl, to the south, as among the Klallam and Snohomish, they were rather such things as were difficult to procure, such as the products of the hunt, and were no more than were given to any respected old man.570 The situation on the Columbia river and in coastal Washington and Oregon was different. These chiefs had real authority. Wishram, Quinault,571 and Coos chiefs were obeyed, their authority being chal- lenged only by the rich. Commoners among Wishram and Coos gave largely from their gains and in return could expect the aid of their chiefs when in distress. The chiefs sat as judges in cases of witcheraft and murder, assessing heavy fines and ordering the culprit killed if he defaulted. Such fines were not theirs but went to the family of the murdered man. They also paid such fines for their followers. What little is known of the Tillamook suggests an identical situation. This judicial function, with its meting of fines and punishment, made its appearance among the Klamath by way of the Dalles early in the middle of the last century. Northwestern California shows considerable similarity to western Oregon, but the authority of rich men was slight in comparison and they did not sit as judges. Among the Hupa, Shasta, and presumably their neighbors, they were obeyed because they gave of their store in times of scarcity and paid fines for followers who were in difficulties. The headmen in more southerly Californian groups exercised more formal authority, as the Yuki and Patwin headmen, who may have made decisions of war and directed town activities. In northeastern California (Maidu, Achomawi, Atsugewi) chiefs were much like 570 Boas, Ethnology of the Kwakiutl, 1333 f.; Gunther, Klallam Ethnogra- phy, 204; Haeberlin-Gunther, Ethnographische Notizen, 38. 571 Olson, Quinault field notes, 1926. 1930] 311 312 University of California Publications in Am. Arch. and Ethn. [Vol. 30 those of the Klamath; theirs was an advisory function and they led in war.572 A corollary of the chief rich man's position in California and Oregon was that he did not himself engage in war. By reason of his wealth he could terminate the conflict by paying the blood money necessary to balance. the score. This has been observed of the Tilla- mook, Shasta, Hupa, Kato, Pomo, and was perhaps general.573 There is some suggestion of this in Klamath custom; a chief was not ordi- narily a war leader but he certainly did not make good the blood- money payments. On the other hand this may have been nothing more than the common western attitude that a chief was not of neces- sity a war'leader. One characteristic feature of the cultures from Columbia river southward, which has been touched on before, was a well developed system of fines for injuries. This did not appear on the Northwest Coast nor in western Washington, but was by no means confined to northwestern California as has been implied. It is quite characteristic of a larger area: Wishram, Alsea, Coos, Takelma, Yurok, Karok, Hupa, Shasta, Pomo.574 Remuneration was often demanded for com- paratively slight injuries; thus an Alsea shinny player who was hit exceptionally hard had to be compensated. The eastern groups (Klamath, Maidu, etc.) had no such sentiment, but it is worthy of note that t.hey did compound a murder by a property settlement, a notion which may have been absent still farther east in the Basin. Speaker Chiefs' speakers occurred among many western tribes. There were none such among the Klamath, whose shamans had however assistants with compa.rable functions. These have been discussed with shamanism. The chief's speaker was eminently characteristic of the Northwest Coast, where his activity in the potlatch and other cere- monies has been often remarked. He not only repeated the words of his chief to the gathering but also served as his representative and advocate. In the Great Basin, on the other hand, the function of the speaker wa.s merely to repeat verbatim after an orator or story-teller. 572 Goddard, Life and Culture of the Hupa, 58; Kroeber, Handbook, 164, 296; McKern, Functional Families; Dixon, Northern Maidu, 2.23, Notes on the Achomawi, 215; Faye, Notes on the Southern Maidu, 41. 573 Kroeber, A Kato War; Loeb, Pomo Folkways, 202, 203, 236. 574 Frachtenberg, Alsea Texts, 199; Sapir, Notes on the Takelma, 270; Loeb, Pomo Folkways, 297. 1Spier: Klamath Ethnography3 The northwestern type of speaker was known among Klallam, Quin- ault, and Wishram, and proba.bly to all the intervening tribes.575 Their presence may also be suspected in coastal Oregon, for their functions would have been quite consistent with the social sentiment there. Some positiveness is lent to this by the presence of what seems to have been a chief 's speaker among the Rogue river tribes partici- pating in the treaty negotiations of September 10, 1855. It is recorded that the chief spoke to a man who shouted his words to the assembled warriors.576 Certainly the speaker was absent in northwestern Cali- fornia. On the other hand, Drake's description of the Coast Miwok in 1579 makes it certain th.at the speaker was a recognized function- ary further south.577 The Basin type of word for word repeat.er occurred over a wide area., having been recorded for the Nez Perce, the Sahaptins of Priest Rapids on the Columbia, Southern Maidu, Paviotso (Nort.hern Paiute), Kaibab Paiute (?), and Havasupai.578 Something of the sort might have been expected among the Klamath; they share, however, a substitute, the shaman 's speaker, with the northern California groups. Slavery Slave-raiding and slave-holding may not have developed among the Klamath until the coming of the horse and the opening of a slave traffic with the Columbia river tribes in the early nineteenth century. It is probable that prior to that time there was only the mild interest in holding captives found among their southern and eastern neigh- bors. Their slaves were captives of war. Slave-holding and the meas- ure of wealth in terms of slaves were characteristic of all north- western people. These were everywhere taken in war and sold. Slave- holding of this type prevailed not only in western Washington to the Columbia but southward through Oregon (Tillamook, Alsea, Takelma, Upland Takelma,579 and Shasta). Coos and others of western Oregon should undoubtedly be included. Slavery is not mentioned in Teit's account of the Middle Columbia Salish of eastern Washington, and does not appear consistent with their general mode of life.580 Directly 575 Gunther, Klallam Ethnography, 306 f.; Olson, Quinault field notes; Spier and Sapir, Wishram Ethnography. 576 Walling, History of Southern Oregon, 223. 577 Kroeber, Handbook, 276. 578 Spier, Havasupai Ethnography, 255. o79 Sapir, Notes on the Takelma, 252, 267. 1930] 313 314 University of CaUifornia Publications in Am. Arch. and Ethn. [Vol. 30 southward of the Klamath there was no slavery; this is certain for Northern and Southern Maidu,581 and very probably Achomawi, Atsugewi, and Northern Paiute. This seems generally true of Cali- fornia with the exception of the northwest (Yurok and probably their neighbors), where there was debt slavery, but where slaves were not measurable articles of wealth. Sunmary The general position of the Klamath in relation to surrounding tribes was in earlier days the simple social structure of the Basin peoples. Since the opening of the last century they have borrowed from the Columbia tribes some sense of social stratification based on wealth, slave-holding, and trading, enhanced importance for their chiefs coincident with a change in the basis of selection from war to wealth, and quite recently judicial functions for the chiefs. There was thus a shift in social dominance from shaman to chief. GIRILS' PUBElRTY RITES AND WOMEN'S OBSERVANCES It is characteristic of western America that adolescence ceremonies were held for girls rather than boys. This is not to ignore the wide extension of the gua.rdian spirit quest, more regularly performed by boys than girls, nor the boys' initiation ceremonies of southern Cali- fornia and the Southwest, but the boys' activities were less commonly ritualized save in the southwestern sector. The boys' quest among the Klamath has been discussed in connection with shamanism. The girls' rites in the west can best be understood with reference to a background of women's observances throughout the continent. Certain features of women's behavior had a well-nigh universal dis- tribution. The general conception was that woman during her period- ical illnesses was a source of contamination to, others or a peril to herself. A series of restrictions ensued; she was secluded or her movements restricted during the period. For her own well-being she must fast or at least refrain from fresh meat or fish, but the latter proscription was also rationalized. as avoiding damage to, the food supply, especially fishing. She must not touch hands to face or hair, using instead a scratching stick, must not wash nor arrange her hair, and must dispose of her garments before resuming her normal life. 580 Teit The Middle Columbia Salish, esp. 122. 581 Dixon, Northern Maidu, 206; Faye, Notes on the Southern Maidu, 44. Spier: Klamath Ethnography Restrictions to women's behavior at this time have a very different implication from elsewhere in the world, for on the whole in North America the Indian woman's social position was on a plane of equality with that of the man. At least women were not excluded from social and ceremonial participation merely because they were women, although they were commonly excluded during their illnesses. On the contrary, in almost all parts of the continent women did partici- pate in ceremonies with men and in analogous ceremonies of their own. Such restrictions held throughout the major part of the continent; California, Northwest Coast, Basin, among the non-Pueblo South- western tribes, in the greater part of the Plains, and Eastern Wood- lands. On the other hand very meager restrictions or none at all have been noted in the northern Plains (Gros Ventre, Blackfoot, Crow, Arapaho), in the Arctic (Copper and Central Eskimo), pos- sibly among the Pueblos generally (Zufni), and again among the Wishram.582 Quite uniformly seclusion was in a hut apart, a scratcher was used, and fresh meat tabooed. Lowie has discussed the distribution of these features in the Basin and southern California, Kroeber in Cali- fornia at large.583 They are however not confined to these areas. I will give here the distribution of these features so far as literature is available; our authorities are remarkably reticent on the whole subject. Hut and the period of seclusion: Tlingit (1 or 3 days), Snohomish, Nisqually (5), Middle Columbia Salish, Nez Perce, Kutenai ?, Chi- nook ? (5 or 4), sometimes Klamath (5), Achomawi-Atsugewi (3 or 4), Sacramento Valley Maidu (3 or 4), Northeastern Maidu (4 or 5), Pomo ?, Northern and Southern Ute, Lemhi and Wind River Sho- shoni, Pima (4), Assiniboin, Dakota, Cheyenne, Omaha and Ponca (4), Menomini, Winnebago, Chickasaw (3) .5 To these should be 582 Kroeber, Ethnology of the Gros Ventre, 181; Wissler, Blackfoot Social Life, 29; Lowie, Crow Social Life, 220; Kroeber, The Arapaho, 15; Jenness, The Copper Eskimo, 158; Boas, Central Eskimo, 596; Parsons, in American Indian Life, 157 f.; Spier and Sapir, Wishram Ethnography. 583 Lowie, Cultural Connection, 145; Kroeber, Handbook, 861 f. 584 Krause, Die Tlinkit-Indianer, 218; Haeberlin-Gunther, Ethnographische Notizen, 46, 47; Teit, Middle Columbia Salish, 114; Spinden, Nez Perce, 198; Chamberlain, 186; Boas, Chinook Texts, 247; Dixon, Notes on Achomawi-Atsu- gewi, 217; Northern Maidu, 233, 238; Kroeber, Handbook, 254, but cf. Loeb, Pomo Folkways, 273 f.; Lowie, Notes on Shoshonean Ethnography, 272 f.; Northern Shoshone, 214; Russell, Pima Indians, 183; Lowie, The Assiniboine, 39; Kroeber, The Arapaho, 15; Grinnell, Cheyenne Indians, I, 130; J. 0. Dorsey, Omaha Sociology, 267; Skinner, Menomini Social Life, 52; Radin, The Winne- bago Tribe, 136; Speck, Notes on Chickasaw Ethnology, 56. Modoc women "abstained from the use of all kinds of meat for five days in each moon" (Meacham, Wi-ne-ma, 127). 1930] 315 316 University of Cah;rfornia Publications in Am. Arch. and Ethn. [Vol. 3U added cases where the menstrual hut was prescribed' during the adolescence rites but not later (or at least is not so recorded): Tahltan, Tsimshian (10 days), Wishram (5), sometimes Klamath (5), Shasta '(10), Coast Central Pomo (4), Eastern Pomo (5), Paviotso ? '(5), Northern Saulteau'x, Plains Cree. Seclusion within the house was also insisted on throughout the Northwest Coast and northern Cali- fornia, in the northern sector in a screened-off corner: Ten'a (about a week), Klallam, Northern' Maidu of the foothills, and Pomo ( ?).585 The distribution of the menstrual hut and this substitute was so gen- eral (save in the northern Plains, Arctic, Pueblos, and Wishram) that I think we may infer its presence in the majority of cases where it, has not been recorded. Scratcher: Klamath, Northeastern Maidu, Pomo, Havasupai, Dakot.a, Plains Cree; to which we may add the Icases of its use in the adolescence rites: Tahltan, Nootka, Klallam, Snohomish, Sacramento Valley MIa.idu, Shasta, Yurok, Hupa, Yuki, Luisefio, Pima, White Mountain Apache, Winnebago, AMenomini, probably Ojibway, Otta.wa, Sauk-Fox, and Kickapoo. This again is so general a distribution that its presence may be suspected whe-rever -restrictions were imposed. It was however definitely absent among the Klallam. Fresh meat or fish tabooed: Kutenai (food taboos), Snohomish, Chinook, Klamath, Mlodoc (no meat), Achomawi-Atsugewi, Maidu (no meat or fish), Huchnom (deer meat), Pomo, Paviotso, Northern and Southern Ute, Lemhi and Wind River Shoshoni, Southern Paiute, Havasupai, Pima, Central Eskimo (raw meat). The girls' rites usually involved a general fast, which, while it might be relieved by taking some food, uniformly mainta.ined the meat taboo. The additional cases noted in this connection are Tahltan, Tsimshian, Bella Coola, Kwakiutl, Nootka., Klallam, Nisqually, Tillamook, Shasta, Yurok, Hupa, Wiyot, Sinkyone, Yuki, Luisefio, White Mountain Apache, Cheyenne (boiled meat), Winnebago, Northern Saulteaux. The Klal- lam woman also fasted on the first day of her periods. The meat taboo was perhaps the single universal restriction in the areas where taboos were at all observed, for it even occurred. among the Eskimo. The distribut.ions outlined are not quite complete even when the absence of data is noted. Lowie has pointed out that in some areas childbirth must take place outside the home, the menstrual lodge being sometimes used, and the same precautions against eating meat and using the fingers to scratch being observed. 585 Reed and Parsons, in American Indian Life, 346; Gunther, Klallam Ethnography, 241; Dixon, Northern Maidu, 235; Loe, Pomo Folkways, 273 f. Spier: Klamath Ethnography 3 It will be observed that the Klama.th conformed to the general usage in that meat and fresh fish were avoided and the scratcher used, but a woman need not leave her home unless she was a member of a shaman-'s household. There are certain other taboos which were by no means universal but occurred over considerable a.reas in the same form. For example, salt was taboo to menstruants in southern and cent.ral California (Mohave, Dieguefio, Luisefio, and much to the north, Pomo and Yuki). It was in the na.ture of the case that those regulatory prescriptions that surround women throughout their lives held with equal force at the first period. It was not inevitable however that the conduct of the girl at her first period should be further regimented, or that restrictions be applied to the first period and none other (as among the Wishram). As I not.ed at the outset of this discussion, ritualiza- tion of the first. period was characteristic of the extreme west alone. The rites had everywhere a composite character as far as purpose went, although it was very much the same throughout their whole range, but differently weighted in the north and south. I think we can safely state that the general purposes of the rites were to announce the nubility of the girl and thus qualify her as a candidate for mar- riage, and to serve her well-being at this physiological crisis. In the north the inevitable wealth display served a.s well to fix her social status. In the central section the industry and modesty she was forced to display were intended to fix her habits for life. While this looks much like a magical procedure, it may well have been a more direct way of achieving that end. It is indeed probable that the acts of this period of emotional stress should retain a high degree of con- sciousness throughout her life or at least until her habits were sta- bilized. In the south her physiological well-being was accentua.ted to the extent that other ritualistic possibilities, such as dancing, were dwarfed. Apart from those features described above tha.t ha.ve a general distribution there wyere rites of a specific character which fall into a number of distinct geographic types: 1. In the Northwest the adolescent girl was secluded in a closed-off section of her home, squatting in a constrained position, her head covered by a hat or mat. At the expiration of this period of seclu- sion a potlatch, a feast with distribution of property, was given in her honor by her parents. A series of songs were sung, heredit.ary 1930] 317 318 University of California Publicationm in Am. Arch. and Ethn. [Vol. 30 in her family and satirical in nature or relating to love affairs. The personal restrictions continued for a period upward to a year. Observances of this type have been noted from Alaska to the strait of Juan de Fuca: Malemut and Unalit Eskimo, Ten'a, Tahltan, Tlingit, Tsimshian, Bella Coola, Kwakiutl, Nootka, Klallam, Sno- homish, Nisqually, and Quinault. Variations will be discussed later.586 2. In Oregon and northern California there was a dance in which the girl participated each night of her period of seclusion. During the day she lay quietly at home or in a, hut, alternating this with periods of strenuous activity, gathering firewood or uselessly piling logs. At dawn she ran toward the east. She wore a feather visor or her head was covered to hold her gaze averted whenever she emerged. Two dance forms occurred together; a circling ring and a line abreast trotting to and fro. Their accompaniment was a deer-hoof rattle; the songs related to men and women or to food products. The dura- tion of the rite was quite consistently five or ten days. In Oregon at least this was the occasion for a vision quest. Throughout the region the performance was repeated at one or more subsequent illnesses. The Klamath observances belong here. Wishram, Tillamook, Coos, Takelma, Klamath, Achomawi, Atsu- gewi, Northern Maidu, Shasta, Karok, Tolowa (?), Hupa, Wiyot, Sinkyone, Yuki, Huchnom.587 3. In central and southern California and in adjacent Arizona the rite was simple: the girl was placed on a heated bed, sometimes danced over or sung to. The period was somewhat variable but most commonly four days. Again her eyes were covered or averted. Added to her personal restrictions was the proscription of salt. These restrictions continued in force as in the north, but not for so long, a month or more. Pomo, Miwok, Tiubatulabal, Juaneino, Luise-no, Northern Diegueino, Cahuilla, Mohave, Havasupai.588 586 Nelson, Eskimo about Bering Strait, 291; Emmons, Tahltan, 104; Krause, Die Tlinkit-Indianer, 217; Boas, Tsimshian Mythology, 450, 531, Third Report, 12; Ethnology of Kwakiutl, 699 f.; Sapir, Girl's Puberty Ceremony; Gunther, Klallam Ethnography, 239; Haeberlin-Gunther, Ethnographische Notizen, 46; Olson, Quinault field notes, 1926. 587 Spier and Sapir, Wishram Ethnography; Boas, Notes on Tillamook, 6; St. Clair and Frachtenberg, Traditions of the Coos, 26; Sapir, Notes on Takelma, 273; Dixon, Notes on Achomawi and Atsugewi, 216; Dixon, Northern Maidu, 232 f.; Kroeber, Handbook, 106, 119, 127, 148, 195, 205, 299; Goddard, Life and Culture of Hupa, 53. 588 Loeb, Pomo Folkways, 271 f.; Kroeber, op. cit., 452, 609, 641, 673, 707, 716, 748; Spier, Havasupai Ethnography, 325. Spier: Klamath Ethnography 4. Some of the non-Pueblo Southwestern tribes had a dance at this time, occupying four nights. "The essential features of this cere- mony are numerous songs and prayers uttered by the priest hired for the occasion, dancing by the girl or girls for whom it is held, a foot race by the girl, and the painting of the girl and of the spectators, who expect good fortune as a result. "589 White Mountain Apache, Navaho, Jicarilla, and perhaps Pima should be included here.590 In contrast to this definite ceremonializing of the first crisis in western America, all of the groups east of the Rockies and those in the Great Basin were one in ignoring this opportunity for ritualiza- tion. They confined themselves to prescribing only such restrictions as applied to women throughout their span of life. Some indeed lacked restrictions of any sort. At this writing I have access only to a scattering of accounts but they may be taken as random and representative. In the Great Basin these included Paviotso (Northern Paiute), Shivwits, Northern and Southern Ute, and northward the Kutenai; in the Plains, Plains Cree, Assiniboin, Dakota, Cheyenne; in the Eastern Woodlands area, Winnebago, Menomini, Ojibway, Ottawa, Sauk-Fox, Kickapoo, Northern Saulteaux; in the Southeast, Chicka- saw.59' I have pointed out above that there are areas where restric- tions to women in their periods were absent or almost nil; the North- ern Plains, Arctic, and Pueblos. There were there no first rites, of course. In pointing out these four types of ceremony it is not implied that there was no difference among the tribes holding to any one of them. It will be sufficient for my present purpose, the analysis of Klamath customs, to discuss the variations of the second type. Wherever information is available it appears that the dance was held each night of the girl's seclusion. The exception is the Wishram, whose dance followed that period. In this the Wishram conform to the habit of the Northwest Coast tribes. 589 Goddard, Indians of the Southwest, 173. 590 Goddard, White Mountain Apache Myths, 123; Goddard, in American Indian Life, 147 f.; Reichard, Navajo, .135-9; Goddard, Jicarilla Apache Texts, 266; Russell, Pima Indians, 182; Parsons, Notes on the Pima, 464. 591 Lowie, Shoshonean Ethnography, 272 f.; Chamberlain, 186; Skinner, Plains Cree, 69; Lowie, Assiniboine, 39; Radin, Winnebago Tribe, 136; Skinner, Menomini Social Life, 52; Eastern Cree and Northern Saulteaux, 152; Speck, Chickasaw Ethnology, 57. 1930] 319 320 University of California Publications in 'AM. Arch. and Ethn. [Vol. 30 The place of seclusion shows an interesting cleavage between coastal and interior tribes. The former confined the.girl to her home (Tillamook?, Coos?, Yurok, Hupa, Wiyot) ; the latter to a hut (Wish- ram, Shasta, Sacramento Valley and Northeastern Maidu). The exceptional case here is the Northern Maidu of the foothills who used the dwelling. The Klamath, Achomawi, and Atsugewi conformed to the interior practice of excluding the girl from her home, sending her into the brush, and at least the Klamath specified that a hut was erected for, her if shelter was needed. It will be ob-served that the seclusion within the home in coastal Oregon and northwestern Cali- fornia looks-like an extension of Northwest Coast practice. The duration of the period was quite uniformly five days or its multiple -ten (five days: Takelma, Klamath, Achomawi-Atsugewi, Foothill Maidu, Sacramento Valley Maidu [5 days or a week] ; ten days: Coos, Shasta, Karok, Yurok, Hupa). The Wishram seclusion was five days, followed by a five-day dance period. The Wiyot period is stated as five or ten days. Here again an antithesis is suggested between interior and coast, but with the line drawn further inland. The fixing of the number of days at five or ten seems to me clearly an application of the pattern number five common to this area. This view is emphasized by the fact that elsewhere the number was differ- ent, most commonly four days. On the Northwest Coast, e.g., the number was quite variable, most commonly four days, and those cases there where a five-day period was prescribed lay toward the southern end of the area (Chinook, most commonly five days; possibly Nis- qually, where the duration was of the same order, five to nine days, Quinault, ten days). This was true again in more southerly Califor- nia and the Southwest where the number was most commonly four, with the Eastern Pomo in the north holding to five. Finally we note that in the Basin the period was five days for Paviotso, ten for North- ern Ute. Kroeber was puzzled by the scarcity of the number four in the girls' ceremonies within the boundaries of the present California.592 The majority of his cases were from northern California. But the situation is at once clear if we extend comparisons beyond its bor- ders; five (and ten) was common to Oregon, northern California, and the northern Basin, four to the remainder of California (with certain exceptions in the south) and the Southwest. This is what it should be when we consider the distribution of pattern numbers in the west. Five was the pattern number of the 592 Handbook, 863. Spier: Kltmath Ethuography southern Northwest Coast, Oregon, California as far south as the Sinkyone (and perhaps the Costanoan), and the Basin. The number was variable in central and southern California, as Kroeber has pointed out, but was as regularly four on the Colorado river as in th.e Southwest.593 The curious featu.ure is that five as the pattern number did not regularly make its appea.rance in the adolescence rites; it is seemingly missing from them on the northern margin of its distribu- tion and again in the southern Basin. What we have to recognize is that the pattern number does not operate automatically at every pos- sible opportunity; furt.her, that there ma.y be an appreciable lag before a specified number of days, things, etc., is made to conform to the standard number. Thus the Ghost dance of 1890 derived by the Walapa.i from the Southern Paiute, whose pattern number was five, appea.red among them as a five-day ceremony although their pattern was four. Characteristic of this local area was some device to hold the adoles- cent girl's ga.ze averted when she left her seclusion. It was a visor of blue-jay feathers among the Takelma, Shasta, and Karok, an eye- shade of skin among the Northern Pomo. Elsewhere her head must be covered in some fashion; it was so prescribed by Hupa, Wiyot, Sinkyone, Coast Central Pomo, Tiibatulabal. The visor appeals to me as the local analogue of the large hat. or mat prescribed by North- west. Coast tribes: Ten'a, Tlingit, Tsimshian, Bella Coola, Kwakiutl, Nisqually, and Tillamook. The Nootka and Achomawi-Atsugewi head- bands may also be analogues. In. some instances the girl's head must be covered with a basket even as she rests: Takelma, Shasta, Yuki, Luiseiio. This is, to be sure, only a specific form based on a more wide- spread view that the adolescent's glance was malignant a.nd must be cast. down. This notion prevailed as far south as the Colorado river (Yurok, East.ern Pomo, and Mohave in addition to those just cited), bu.t was notably absent in the Basin and Southwest, and may not have been nearly so strongly emphasized on the Northwest Coast. It. cropped up again in a far distant area., the southern Great Lakes (Winnebago, Menomini, and probably Ojibway, Ottawa, Sauk-Fox, and Kickapoo), where the menstruant might. not look up or out of the lodge. The otherwise unintelligible prescription that the Klamath maiden must dance with downcast eyes-for I did not find that there was any belief that her gaze was malignant-.is thus explicable as a bald imitation of neighboring types of behavior. 593 Handbook, 875. 1930] 321 322 University of California Publications in Am. Arch. and Ethn. [Vol. 30 It was equally characteristic of this group of rites that the rest during the day, which was dictated as proper, was interrupted by strenuous activity, which took two forms, gathering firewood and racing at dawn. Gathering wood was the Achomawi-Atsugewi, Shasta, Karok, and Yurok custom; the Northeastern Maidu girl went off piling up logs. This was not the Klamath habit, unless I misunder- stood my informants, yet it is analogous to the activities of their adolescent boys. It does appear again among the adjacent Paviotso. There was an analogue among the Mohave, the equally useless picking of leaves. The race at dawn, or where it is specified in the accounts, toward the dawn, was common to the Hupa, Karok, Northeastern Maidu, Achomawi-Atsugewi, Klamath, and Takelma (at midday?). This particular form of enforced activity was apparently absent both on the Northwest Coast and in central to southern California, but ap- peared again in the Southwest (Havasupai, White Mountain and Jicarilla Apache, Navaho). One feature of the Klamath rite, the bands worn about the joints by the girl, appeared again in almost precisely the same form among the Chinook and somewhat differently perhaps among the Nisqually and Kwakiutl. Further, these were disposed of in the same specific fashion, by placing them in trees. This suggests that their use was more general than these data indicate. As to the girl's posture during her rest this is not always explicitly described, but the implication is that she lay rather than sat. We observe however that the Tillamook posture was like that uniform in the north, namely sitting. This is noted again for Sacramento Valley Maidu and Wiyot, although in these cases the statements may be only a fashion of speech rather than a precise statement. The Klamath conformed to the general practice of the area of having the girl lie. The dance itself was held at night, each night of the period, throughout this whole area from the Wishram to the Huchnom. The notable exception was the Yurok who held no dance. The precise sequence of events was somewhat variable, but the dance form was not, and the dance rattle was uniform. Likewise, unlike in the regions north and south, the girl herself danced. Two dance forms occurred together, a ring of dancers and a line trotting to and fro. Their distributions were as follows: ring, Takelma, Shasta, Karok, Foothill and Northeastern Maidu, with perhaps the Hupa ring of stationary men as a doubtful analogue; to and fro, Klamath, Acho- mawi-Atsugewi, Northeastern Maidu, Karok, the Shasta form being Spier: Klamath Ethnography a line of men moving sideways. It was only in connection with the latter form of dance that the girl performed; to be specific it is known that she danced among Klamath, Achomawi-Atsugewi, North- eastern Maidu, Shasta, Karok, and, though we do not know the dance form, Sinkyone and Yuki. Inasmuch as it is specified in the case of the last two that the girl was supported as she danced, we may well suspect that they had the same dance form. It should not be assumed that dancing during the puberty rite was confined to this area. Dancing may have occurred on the North- west Coast in connection with the potlatch more commonly than recorded (Nootka and Ten'a?). There is nothing to suggest that it was the same form of performance, however. In southern California the dancing was about the girl as she lay on a hot bed (Luise-no, Juaneiio, Miwok) and was apparently quite different in character and intent. Dancing was again characteristic of the Southwestern ceremonies (White Mountain and Jicarilla Apache, Pima), and again the White Mountain and Pima forms were lines moving to and fro, with the girl engaged with them in the case of the former. Even the Jicarilla girl herself danced. These specific features are suspiciously like those of northern California and Oregon. Perhaps we do not yet know all that can be learned of this rite in the Basin. To revert to the dance rattle; it was prevailingly the deer-hoof jingler type, although handled somewhat varyingly. This has been noted for Klamath, Achomawi-Atsugewi, Northeastern Maidu, Shasta, Karok, and Tolowa, while the Hupa, Sinkyone, and again the North- eastern Maidu used the clapper rattle. This type of deer-hoof rattle was distinctively Californian and generally distributed in the state. So far as I know it does not occur north of the Klamath. The clapper rattle was on the other hand a central Californian type, as Kroeber has pointed out.594 As he further remarks, rattles were used in north- western California only in connection with adolescence ceremonies, but this was also true of the sector to the north and east apparently (Shasta, Achomawi-Atsugewi, certainly Klamath,595 and perhaps Takelma and Coos). There is little information on the songs, but the little existing suggests that these were of two types only; songs about foodstuffs and about love affairs. Thus it is noted that the Sacramento Valley Maidu 594 Handbook, 823, 862. 595 The Klamath acquired the use of the clapper rattle as well only with the Ghost dance of 1870. 1930] 323 324 University of California Publipations in Am. Arch. and Ethn. [Vol. 30 sang about foodstuffs gathered by .women, the Yuki about acorns, also, a woman's food product. The Klamath sang about men and women; similar- love songs have been noted for the Klallam and -those of the Nootka.were satirically sexual. This temptation to invoke the sexual implication of the occasion took another expression; a period of license during the dance among the Northeastern Maidu and the Yuki. It is highly probable that this attitude was widespread. How generally this crisis was made the occasion for a vision quest by the girl is not clear. This may have been a trait of Oregon and Washington alone. To be sure, the guardian spirit concept was but little developed in California, at least among more typical tribes. The girl was sent out at night by Nisqually, Tillamook, Coos, and per- haps the Klamath, to seek a spirit in the manner stereotyped for boys, wandering on the mountains and bathing in lonely pools. The whole performance was repeated at the following monthly periods in this a.rea. The number of subsequent performances brought the total to five among the Klamath, conformable with their pattern number, to three among Achomawi-Atsugewi, Shasta, and Karok, to two among Chinook, Northeastern Maidu, and Sinkyone, while it was repeated at monthly intervals throughout the winter among the Huch- nom. While the number of performances was variable, the habit of their repetit.ion was confined to a fairly consolidated area. The Mohave custom of imposing restrictions at successive periods for a reduced number of days at each period is not directly comparable. It is rather analogous to the continued restriction of a month's duration at first menst.rua.tion imposed by Shivwits, Luiseflo, and Coast Central Pomo. There is no indication of such continuous restrictions in the Oregon-north Californian area, but this was characteristic again of the Northwest Coast tribes who imposed them for a period upward to a year (a year, Ten'a, Tlingit, Tsimshian, Bella Coola; six months to a year, Tahltan; four to ten months, Nootka; six to eight months, Nisqually; five months, Quinault; fifty or one hundred days, Chinook). The place of the Klamath in all this is fairly clear. They have very little concern with women's periodicity and its possible social or cere- monial significance. Women are not excluded from home at their periods save in the case of those living with sha.mans. The rigorous segregation that occurs on all sides of them, throughout the Pacific coast, the Plateau, and Basin, made practically no impression on the Klamath. It is surely not accidental that it is the near-by Wishram who also care nothing for such restriction. Such restrictions as do Spier: Klamath Ethnography occur among the Klamat.h are the well-nigh universal taboo of fresh fish and meat, and the use of the scratcher. On the other hand, the Klamath share a rite at first menstruation with the Pacific coast peoples. Their rite is of the local type known throughout western Oregon and northern California, and is only slightly distinguished by peculiar detail. They follow the interior practice of sending the girl away from the dwelling and maintaining the dance for five days. They link with some of their neighbors and the northwest Californians in using the deer-h.oof rattle only in this rite. Possibly they share with the tribes to the north some feeling that this is an appropriate time for the girl to secure spirit power. For the rest, the rite is generally like that of western Oregon and northern California, even to such specific details as having the girl recline at length during the day, running at dawn, and in the general practice of repeating the whole affair for some months subsequently. Other general features appear only in diminished or attenuated form: only one of the two dance forms, the line abreast., is known, and, while the gaze of girls of this area. is averted by a visor, the Klamath demand only that she voluntarily keep her eyes cast down. 1930] 325 326 University of California Publications in Am. Arch. and Ethn. [Vol. 30 APPENDIX THE MODOC NOVICE DANCE596 The medicine-man of the Indian is not selected in the same manner as are the medicine-men of the white race. He does not choose the profession as an easy way to make a living; he knows too well the fearful responsibility of his position. He has seen the medicine-man too often slain on account of his failure to cure. Hence, the position is one not sought, but as they believe is forced upon them by the Great Spirit. The indications of the selection appear at various ages; sometimes in childhood, sometimes in middle life, and again in old age. The first intimation of being tow-ed is by bleeding at the mouth, or by becoming crazed, with jerkings and twitchings of the muscles. When these manifestations appear, the tribe, or band, is notified and a council is called, on which occasions the older doctors officiate. It is one of great impor- tance to all the people, inasmuch as the medicine-man has so much power. The candidate is very solicitous, because of the fearful responsibility involved in assuming the office. He dare not resist the spirit, as doing so would imperil his life forever, present and future. Hence he comes to the examining council with trepidation..... When the council is convened the preparations are completed by the older doctors making prayers, and offering sacrifices, killing white-haired dogs and hanging the skin upon a pole, which is raised above the council. On the top of the pole are placed the tail-feathers of the great Medicine Hawk, which is called the "king hawk," because of his mastery over all other birds. The skins of the white weasel, the otter, the mink, and black-fox, all of which are said to be "sacred," are also hung upon the pole. These sacred skins are supplied by the older doctors, or the friends of the candidate, and they are his after the ceremony is over, if he shall be found to be a doctor of the right kind, for with these people, some of them are called of the good spirit and some of the bad spirit. The authority under which they act, is a matter of great con- cern to all parties and more especially to the candidate; himself; for should he prove to be called of the evil spirit, his ministry is of short duration, ending his life as soon as the authority is clearly recognized, unless he escapes. It may be easily understood why "The bringing out" of a medicine-man is a great event. That my reader may better understand the ceremony, I herewith present a full report of the "bringing out" of young Ut-ti-na, a half Modoc and half Pitt River Indian, who had found himself among the former, when he was called to the ministry, as related by an eye witness. The candidate was not more than eighteen years of age. He was, however, a full-grown, well-developed man, of more than ordinary stature and physical strength. This fact made his "coming out"s a matter of vast importance. For several days young Ut-ti-na had been bleeding at the mouth and nose, and otherwise showing by unmistakable signs that he was " tow-ed, " by some great spirit; that is to say, he was "marked for a medicine-man."> When this fact became known, the chief, Captain Jack, called the medicine-council. The medicine-pole of the tribe was brought out. The "watchmen" of the band appeared with the usual insignia of office a cap made of fawn-skin. The chief 59e Meacham, Wi-ne-ma, 129-136. Spier: Klamath Ethnography gave the order, and the pole, surmounted by the feathers, skins, and the flag of the tribe was raised. The Modocs have long had a national flag. It con- sisted of two colors-red, and white, in separate parts. In peace, both colors are on the pole, when at war the red alone. The pole was raised, stakes were driven into the ground at short intervals, and a rope stretched round, forming a circle twenty feet in diameter. The men now formed round the rope, facing the pole; next, the women behind the men, (all standing). When all was in readiness the "curley-haired doctor" of Lava-bed fame, led the candidate into the circle. He was closely wrapped in a blanket, nothing but his face being visible. He was placed in position facing to the pole. The Doctor now retired to midway between the centre and the rope, and stood a few moments in silence, suddenly raising his hands and covering his face one of the watchmen fired a signal gun. The Doctor saying "we are here to test the power of the young Ut-ti-na who has been called to serve a Great Spirit." The orator of the tribe, rehearsed this speech in a loud voice. The response of "aye, aye," from the men forming the circle signified approval. The Doctor made choice of nine men to assist him in the services, the remainder sat upon the ground. The Doctor now called upon the Great Spirit to assist him in the work to be performed. Each sentence was repeated by the orator. Stopping suddenly he broke out in a medicine-song, in which the whole assem- bly joined. Clapping his hands the song was stopped as abruptly as it was begun. Another prayer was made and repeated, calling upon the Great Spirit. Then another song and another prayer. The Doctor went round the circle in a slow dance, mumbling in a low voice. When he had made the circuit three times, the candidate dropped the blanket and started round behind him, joining in the songs and prayers. The dance and song went on for half an hour, when the Doctor again clapped his hands, signifying that a spirit had come to him. It was the part of his assistants to guess the spirit. The first said, "a dove"; the Doctor who was standing with his face upturned and covered with his hands gave no response. "A fawn," said another; no motion came from the Doctor; so on through the line of assistants, all eyes watching the Doctor's hands. Had any one of them guessed the spirit seen by the Doctor, he would have clapped his hands as a signal that they were right. If they guessed any one of the sacred birds, or animals, such as are harmless and non-resisting, and it had been the spirit seen by the Doctor, it would have been to them proof that the candidate was called of good spirits, and he would have been put to other proofs. If of the vicious birds or animals, it would have indicated that he was called of bad spirits and was a dangerous man. No one having named the right spirit, the dance and prayer and songs were renewed, and continued until the Doctor clapped his hands. Then each assist- ant guessed again, some one named the "wood-rat," the Doctor clapped his hands. Another song and prayer was begun, asking that some other spirit appear. Again the Doctor clapped his hands and an assistant said, a wolf," the Doctor gave the signal of assent. Thus the ceremony went on until ten dances had been performed, each with the same exactness and propriety; the candidate meanwhile dancing and praying with great fervor. He knew that his life was at the mercy of the Doctor, and dependent entirely upon his integ- rity. In this instance the candidate had been a favorite with the people and was in reality a nephew of the officiating Doctor, but he knew the Doctor would not swerve from what he believed to be the truth; indeed he dared not show the least partiality to his kinsman, for he would be held responsible for the young Ut-ti-na's success in the practice of medicine. When the ten dances 1930] 327 328 University of California Publications in Am. Arch. and Ethn. [Vol. 30 had been celebrated, and the decision based upon the spirits as seen by the Doctor had declared that Ut-ti-na was called of "bad spirits" he was crest- fallen, and hurried away..... Under favorable report the ceremonies would have been continued for five days and nights with short intermissions. At the end of that time the young doctor would have gone alone to the Sacred Mountains, where with praying, fasting, and bathing he would have sought wisdom from the great spirits who had called him to this high office, and after remaining seven days, performing numerous feats of physical strength, piling up rocks and logs while constantly murmuring prayers he would have returned to his tribe and assumed the official insignia of his office, beneath the medicine-pole which he would have brought with him from the mountains. This pole would have been of mountain fir, as no other would have answered the purpose. It would have been carefully stripped of twigs and bark, polished very smoothly, and painted in several colors, white, red, blue, and sometimes black, with curious figures to suit his own fancy. Then, ornamenting it with the skins and feathers whieh were used upon his trial, the feathers of the sacred hawk being above the others, the pole would have been raised beside his own lodge. He would then have begun the month of probation, during which time he would have eaten no meat of any kind, dancing several hours each day round his own medicine-pole, sometimes supported by the women and the ora- tor. At the end of that time the people would come again together, and with great circumstance he would have been admitted to the profession as a regular medicine-man. This ceremony would have been conducted under the direction of the doctor who sat in judgment upon his election in the first place. The people would have made it the occasion for feasting and dancing..... Transmitted July, 1929. Spier: Klam'ath Ethnography BIBLIOGRAPHY ANGULO, JAIME DE 1926. The Background of the Religious Feeling in a Primitive Tribe. AA, n.s., 28:352-360. 1928. La Psychologie religieuse des Achomawi. A, 23:141-166, 561-589. BARRETT, S. A. 1910. The Material Culture of the Klamath Lake and Modoe Indians of Northeastern California and Southern Oregon. UC-PAAE, 5:239- 292. BECKWITH, PAUL 1889. Notes on Customs of the Dakotahs. SI-AR for 1886, pt. 1:245-257. BENEDICT, RUTH FULTON 1922. The Vision in Plains Culture. AA, n.s., 24:1-23. 1923. The Concept of the Guardian Spirit in North America. AAA-M, no. 29. BoAs, FRANZ 1888. The Central Eskimo. BAE-R, 6:339-669. 1891. Third Report on the Indians of British Columbia. Brit. Assoc. Adv. Sci., R, 1891: 408-449 [reprint pp. 2-43]. 1893. The Doctrine of Souls and of Disease among the Chinook Indians. JAFL, 6:39-43. 1894a. Chinook Texts. BAE-B, 20. 1894b. The Indian Tribes of the Lower Fraser River. Brit. Assoc. Adv. Sci., R, 1894:453-563 [reprint pp. 1-11]. 1897. The Social Organization and Secret Societies of the Kwakiutl Indians. USNM-R for 1895:311-738. 1906a. The Salish Tribes of the Interior of British Columbia. Ann. Arch. R, 1905 (Toronto): 219-225. 1906b. The Tribes of the North Pacific Coast. Ann. Arch. R, 1905 (Toronto): 235-249. 1916. Tsimshian Mythology. BAE-R, 31:29-1037. 1920. The Social Organization of the Kwakiutl. AA, n.s., 22:111-126. 1921. Ethnology of the Kwakiutl. BAE-R, 35, 2 pts.: 43-1473. 1923. Notes on the Tillamook. UC-PAAE, 20:3-16. 1924. The Social Organization of the Tribes of the North-Pacific Coast. AA, n.s., 26:323-332. BOURKE, JOHN G. 1892. The Medicine-Men of the Apache. BAE-R, 9:451-603. CHAMBERLAIN, ALEXANDER F. 1906. The Kootenay Indians. Ann. Arch. R, 1905 (Toronto): 178-187. CLARKE, WILLIAM J. 1885. Rock Piles and Ancient Dams in the Klamath Valley. Am. Anti- quarian, 7:40-41. COMMISSIONER OF INDiAN AFFAIRS 1854. Report, 262. 1930] 329 330 University of California Publications in Am. Arch. and Ethn. [Vol. 30 COVILLE, FREDERICK V. 1897. Notes on the Plants Used by the Klamath Indians of Oregon. Contr. U. S. Nat. Herb., 5:87-108 (no. 2). 1904. Wokas, A Primitive Food of the Klamath Indians. USNM-R for 1902:725-739. CUTLIN, STEWART 1907. Games of the North American Indians. BAE-R, 24:1-809. DIXON, ROLAND B. 1904. Some Shamans of Northern California. JAFL, 17:23-27. 1905. The Northern Maidu. AMNHI-B, 17: pt. 3. '1907. The Shasta. AMNH-B, 17:381-498. 1908. Notes on the Achomawi and Atsugewi Indians of Northern Cali- fornia. AA, n.s., 10:208-220. 1910. The Chimariko Indians and Language. UC-PAAE, 5:293-380. DoRSEY, GEORGE A. 1901. Certain Gambling Games of the Klamath Indians. AA, n.s., 3:14-27. DORSEY, J. OWEN 1884. Omaha Sociology. BAE-R, 3:205-370. Du BOIS, CNONSTANCE GODDARD 1908. The Religion of the Luisefio Indians of Southern California. UC-PAAE, 8:69-186. EELLS, MYRON 1884. Census of the Clallam and Twana Indians of Washington Territory. Am. Antiquarian, 6:35-38. 1889. The Twana, Chemakum, and Klallam Indians, of Washington Terri- tory. SI-R, 1886-7, pt. 1:605-681. ELLIOTT, T. C. 1910. The Peter Skene Ogden Journals. Oregon Historical Society, Quarterly, 11:201-222. EM0oNS, GEORGE T. 1911. The Tahltan Indians. UPM-AP, 4: no. 1. FARRAD, LIVINGSTON 1898. The Chilcotin. Brit. Assoc. Adv. Sci., R, 1898:645-648 [reprint pp. 18-21]. 1901. Notes on the Alsea Indians of Oregon. AA, n.s., 3:239-247. 1921. Notes on the Nez Perc6 Indians. AA, n.s., 23:244-246. FAYE, PAUL LOUIS 1923. Notes on the Southern Maidu. UC-PAAE, 20:35-53. FRACETENBERG, LEO J. 1920a. Alsea Texts and Myths. BAE-B, 67. 1920b. Esehatology of the Quileute Indians. AA, n.s., 22:330-340. 1921. The Ceremonial Societies of the Quileute Indians. AA, n.s., 23: 320-352. 1922. Coos and Siuslawan (Lower Umpqua) in Franz Boas, Handbook of American Indian Languages. BAE-B, 40:297-630, pt. 2. FRANCHtRE, GABRIEL 1854. Narrative of a Voyage to the Northwest Coast of America (New York). FREELAND, L. S. 1923. Pomo Doctors and Poisoners. UC-PAAE, 20:57-73. Spier: Klamath Ethnography FRn1ONT, JOHN CHARLES 1846. Narrative of the Exploring Expedition to the Rocky Mountains in the year 1842 and to Oregon and Northern California in the years 1843-44 (New York). 1887. Memoirs of My Life (Chicago and New York), vol. 1. GATSCHET, ALBERT SAsMUEL 1878-79. Mythologic Text in the Klamath Language of Southern Oregon. Am. Antiquarian, 1:161-166. 1890. The Klamath Indians of Southwestern Oregon. CNAE, 2:2 pts. Cited as Gatschet. 1894. Songs of the Modoc Indians. AA, 0.8., 7:26-31. GIBBS, GEORGE 1877. Tribes of Western Washington and Northwestern Oregon. CNAE, 1:157-219. GIFFORD, EDWARD WINSLOW 1916. Miwok Moieties. UC-PAAE, 12:139-194. 1922. Californian Kinship Terminologies. UC-PAA2E, 18:1-285. Cited as Kinship. 1923. Western Mono Myths. JAFL, 36:301-367. 1926. Clear Lake Pomo Society. UC-PAAE, 18:287-390. 1927. Southern Maidu Religious Ceremonies. AA, n.s., 29:214-257. GrFFORD, E. W., and LOWIE, R. H. 1928. Notes on the Akwa 'ala Indians of Lower California. UC-PAAE, 23:339-352. GODDARD, PLiNY EARLE 1903. Life and Culture of the Hupa. UC-PAAE, 1:1-88. 1911. Jicarilla Apache Texts. AMNH-AP, & 1919. Myths and Tales from the White Mountain Apache. AMNH-AP, 24: pt. 2. 1924. Indians of the Northwest Coast. AMNH-Handbook Series, 10. 1927. Indians of the Southwest. AMNH-Handbook Series, 2 (ed. 3). GOLDFRANK, ESTHER SCHIFF 1927. The Social and Ceremonial Organizations of Cochiti. AAA-M, 33. GRINNELL, GEORGE BIRD 1923. The Cheyenne Indians (2 vols., New Haven). GUNTHER, ERNA 1927. Klallam Ethnography. UW-PA, 1:171-314. 1928. A Further Analysis of the First Salmon Ceremony. UW-PA, 2: no. 5. HAEBERLIN, H. K., TEIT, JAM-ES A., and ROBERTS, HELEN H., under the direction of FRANZ BoAS 1928. Coiled Basketry in British Columbia and Surrounding Region. BAE-R, 41:119-484. HAEBERLIN, HERMANN, und GtYNTHER, ERNA 1924. Ethnographische Notizen fiber die Indiannerstiimme des Puget-Sundes. ZE, 1924:1-74. HoDGE, FREDERICK WEBB, ed. 1907. Handbook of American Indians North of Mexico. BAE-B, 30, 2 vols. HOOPER, LUCILE 1920. The Cahuilla Indians. UC-PAAE, 16:315-380. 193G] 331 332 University of California Publications in Am. Arch. and Ethn. [Vol. 30 HOUGH, WALTER 1888. Fire Making Apparatus in the U. S. National Museum. USNM-R, 1888:531-587. ERDLICKA, ALES 1905. Head Deformation among the Klamath. AA, n.s., 7:360-351. JENNESS, D. 1922. The Life of the Copper Eskimos. Canad. Arctic Exped., 1913-18, R, 12. KNIFFEN, FRED B. 1928. Achomawi Geography. UC-PAAE, 23:297-332. KiRAUSE, AUREL 1885. Die Tlinkit-Indianer (Jena). KRoEBER, A. L. 1902a. The Arapaho. AMNH-B, 18: pt. 1. 1902b. Preliminary Sketch of the Mohave. AA, n.s., 4:276-285. 1907. The Religion of the Indians of California. UC-PAAE, 4:319-356. 1908. Ethnology of the Gros Ventre. AMNH-AP, 1: pt. 4. 1920. California Culture Provinces. UC-PAAE, 17:151-169. 1925. Handbook of the Indians of California. BAE-B, 78. 1928. A Kato War. Festschrift, Publication d'Hommage offerte au P. W. Schmidt (Wien), 394-400. LEWis, ALBERT BUELL 1906. Tribes of the Columbia Valley and the Coast of Washington and Oregon. AAA-M, 1:147-209. LOEB, EDWIN M. 1926. Pomo Folkways. UC-PAAE, 19:149-405. LOWIE, ROBERT H. 1909a. The Northern Shoshone. AMNH-AP, 2:165-306. 1909b. The Assiniboine. Ibid., 4: pt. 1. 1912. Social Life of the Crow Indians. Ibid., 9: pt. 2. 1917. Notes on the Social Organization and Customs of the Mandan, Hidatsa, and Crow Indians. Ibid., 21: pt. 1. 1923. The Cultural Connection of Californian and Plateau Shoshonean Tribes. UC-PAAE, 20:145-156. 1924. Notes on Shoshonean Ethnography. AMNH-AP, 20: pt. 3. MARSDEN, W. L. 1923. The Northern Paiute Language of Oregon. UC-PAAE, 20:173-191. MAsoN, OTIS TUFTON 1904. Aboriginal American Basketry. USNM-R for 1902:171-548. MATTHEWS, WASHINGTON 1877. Ethnology and Philology of the Eidatsa Indians. U. S. Geol. and Geog. Surv. [Hayden], Misc. Publ. 7. MCCLINTOCK, WALTER 1910. The Old North Trail, or Life, Legends and Religion of the Blackfoot Indians (London). McKERN, W. C. 1922. Functional Families of the Patwin. UC-PAAE, 13:235-258. MEACHAm, A. B. 1875. Wigwam and War-Path or the Royal Chief in Chains (ed. 2; Boston). 1876. Wi-ne-ma (The Woman Chief) and her People (Hartford, Conn.). Spier: Klamath Ethnography MERRIAM, C. HART 1926. The Classification and Distribution of the Pit River Tribes of California. SI-MO, 78: no. 3. MERRILL, RUTH EARL 1923. Plants Used in Basketry by the California Indians. UC-PAAE, 20: 213-242. MOONEY, JAMES 1928. The Aboriginal Population of America North of Mexico. SI-MC, 80: no. 7. NATCHES, GILBERT 1923. Northern Paiute Verbs. UC-PAAE, 20:243-259. NELSON, E. W. 1899. The Eskimo about Bering Strait. BAE-R, 18: pt. 1. OLSON, RONALD L. 1925, 1926. Quinault field notes. MS. PARSONS, ELSIE CLEWS, ed. 1925. American Indian Life (New York). POWERS, STEPHEN 1877. Tribes of California. CNAE, 3. RADIN, PAUL 1923. The Winnebago Tribe. BAE-R, 37:35-550. REICHARD, GLADYS A. 1928. Social Life of the Navajo Indians, with some Attention to Minor Ceremonies. CU-CA, 7. RIVERS, W. H. R., and HADDON, A. C. 1902. A Method of Recording String Figures and Tricks. Man, 2:146-153. RUSSELL, FRANK 1908. The Pima Indians. BAE-R, 26:3-389. SPIR, EDWARD 1907a. Notes on the Takelma Indians of Southwestern Oregon. AA, n.s., 9:251-275. 1907b. Religious Ideas of the Takelma Indians of Southwestern Oregon. JAFL, 20:33-49. 1909a. Takelma Texts. UPM-AP, 2: no. 1. 1909b. Wishram Texts. AES-P, 2. 1913. A Girl 's Puberty Ceremony Among the Nootka Indians. Royal Society of Canada, Trans., ser. 3, 7:67-80. 1915. The Social Organization of the West Coast Tribes. Ibid., 9:355-374. 1921. Vancouver Island Indians, in Hastings, James, ed. Encyclopaedia of Religion and Ethics (Edinburgh), 12:591-595. SKINNER, ALANSON 19,11. Notes on the Eastern Cree and Northern Saulteaux. AMNK-AP, 9:1-177. 1913. Social Life and Ceremonial Bundles of the Menomini Indianu. AMNH-AP, 13:1-165. 1914. Notes on the Plains Cree. AA, n.s., 16:68-87. 1915. Societies of the Iowa. AMNE-AP, 11:679-740. SPECK, FRANK G. 1907. Notes on Chickasaw Ethnology and Folk-Lore. JAFL, 21:50-58. 1930] 333 334 University of California Publications in Am. Arch. and Ethn. [Vol. 30 SPIER, LESLIE 1923. Southern Dieguenlo Customs. UC-PAAE, 20:295-358. 1925. The Distribution of Kinship Systems in North America. UW-PA, 1: no. 2. 1927a. The Ghost Dance of 1870 among the Klamath of Oregon. Ibid., 2: no. 2. 1927b. Tribal Distribution in Southwestern Oregon. Oregon Historical Society, Quarterly, 28:1-8. 1928. Havasupai Ethnography. AMNH-AP, 29: part 3. SPIER, LESLIE, and SAPIR, EDWARD 1930. Wishram Ethnography. UTW-PA, 3: no. 3. SPINDEN, HERBERT J. 1908. The Nez Perce Indians. AAA-M, 2: pt. 3. ST. CLA., HARRY HULL, and FRcaHTENBERG, LEO J. 1909. Traditions of the Coos Indians of Oregon. JAFL, 22:25-41. SWAN, JAMES C. 1870. The Indians of Cape Flattery. SI-CK1, 16:,1-105. SwANToN, JoHN R. 1904. Social Conditions, Beliefs, and Linguistic Relationships of the Tlingit Indians. BAE-R, 26:391-485. TAYLOR, S. H. 1921. Oregon Bound, 1853. Letters of S. H. Taylor to the Watertown [Wisconsin] Chronicle. Oregon Historical Society, Quarterly, 22: 117-160. TEIT, JAMES A. 1900. The Thompson Indians of British Columbia. AMNH-M, 2: pt. 4. 1906. The Lillooet Indians. Ibid., 4: pt. 5. 1928. The Middle Columbia Salish. UW-PA, 2: no. 4. WALLNG, A. G. 1884. History of Southern Oregon (Portland). WATERMAN, T. T. 1910. The Religious Practices of the Dieguefio Indians, UC-PAAE, 8: 271-358. WISSLER, CLARK 1911. The Social Life of the Blaekfoot Indians. AMNH-AP, 7: pt. 1. 1914. The Influence of the Horse in the Development of Plains Culture. AA, n.s., 16:1-25. INDEX Abies concolor, 173. Achtllea millefolium L., 155. Achlomawi, 24, 25, 39, 40, 66, 216, 224- 324. Adultery, 50, 51. See also Marriage. Adzes, 172. Agrostis, 163. Alnus tenuifolia Nutt., 191. Altair, 221. Apocynum cannabinutm, 175. Arctic, 173. Arctostaphylos nevadensis Gray, 87. Artemisia tridentata Nutt., 131, 172. Arrows, poisoned, 32. Ashland, 8. Athl'ameth, 2. Atsugewi, 40, 66, 216, 224-324. Axes, 172. Bags, 173, 227. Balsa'morrhiza, 162. Bannock, 4, 29, 224-324. Basin Shoshoneans, 35, 216, 224-324. Basketry, 6, 177-193, 213, 227, 230, 234. Barrenness, 49, 55, 57. Barter, 25. Bassariscus, 204, 275. Beaver, 6. Beds, 204. Beliefs. See Omens, Taboos. Big Dipper, 221. Birth, 55-59. Blankets, 46, 175, 204, 232, 293. Buffalo, 42. Buttons, 7. Calendar, 217, 229, 240. Calochortus, 164. Canoe, 31, 72, 83, 127, 144, 147, 169- 171, 227, 230, 232, 293. Carex, 174. Cascade,, 9, 24, 29, 37, 156, 159, 160, 194, 224-324. Cassiopeia, 221. Cayuse, 3, 9, 25, 224-324. Cercocarpus ledifolius, 176. Charm, 83. See also Omens. Chief, 23, 27, 31, 32, 33, 35-39, 51, 68, 94, 113, 213, 235, 236, 282, 307- 313. Children, 86. Chinook, 224-324. Chondrophora na.useosa (Pursh) Brit- ton, 131. Chrysothamnus bloomaeri (Gray), 131. Circuta maculata, 193. Clairvoyance, 273. Coos, 4, 224-324. Comb, 214, 228-232. Constellations, 221. Cradles, 57, 58, 59, 189, 229, 230. Crater lake, 9, 28, 162, 165, 216. Cremation, 13-15, 18-20, 47, 55, 71, 99, 112, 141, 142, 168, 229. Culture hero, 4, 14, 20, 23, 60, 73, 82, 102, 142. Curing, 87, 101, 104, 108, 110, 122, 127, 128, 129, 133, 148, 206, 240. See also Medicine, Shamanism. Dalles, The, 11, 24, 27, 39-41. Deformation, head, 59, 214, 228. Digging stick, 176, 227. Dogs, 6, 7, 19, 42, 122, 133, 158, 287. Dream, 71, 93, 94, 126, 250, 252-254. See also Shamanism. Dress, 30, 69, 74, 92, 110, 111, 173, 207-217, 228, 230, 271, 278. Drill. See Fire drill. Drums, 35. Equisetum hyeinale, 195. Eriogonum stellatum Benth., 131. Etiquette, 91-92. Evernia vulpina, 191. Face painting, 30, 33, 77, 216. Fasting, 56, 73, 77, 93, 96, 115. Figurines, stone, 118. Fishing, 7, 14, 20, 21, 140, 147-155, 189, 227, 232, 238. Fire drill, 172, 228. Field Museum, 69, 110, 111, 118, 158, 182, 184, 185, 189, 207, 212, 213. Five-day period, 56, 72, 86, 114, 115, 229, 320. Fomalhaut, 221. [335] Index Food, 6, 18, 112, 127, 144-147, 160- 169, 225-227, 270. Flour, 8, 204. Fr6emont, 6, 7, 31, 163, 197. Games, 42, 76-86, 229, 230. Gemini, 221. Genealogy, 65. Ghosts, 28, 101. Ghost dances, 34, 41, 89, 93, 101, 141, 207, 208, 232, 234. Great Basin, 36, 89, 94, 158, 224-324. Hair, 213, 328. Hairbrush, 214. Head deformation, 59, 214, 228. Heracleum lanatum Michx., 131. Horse, 4, 6, 7, 24, 25, 31, 37, 40, 41, 42, 43, 46, 52, 60, 68, 86, 95, 97, 99, 100, 127, 225, 235, 293. Houses, 197-205, 227, 232. Hunting, 155-160. Hupa, 43, 224-324. Huts, 6, 7. See also Tents. Insult, 62, 102. Iris misgouriensis, Nutt., 128. Jacksonville, 8. Junous baltiou Willd., 174, 183, 187. Kalapuya, 224-324. Karok, 43, 91, 224-324. Kmaii' kuimps, 14, 20, 23, 60, 73, 82, 102, 139, 141, 142, 143, 148. Klickitat, 212, 224-324. Klallam, 224-324. Kniff en, 25. Knives, 173. Knotted string records, 70, 112, 220. Eunzia tridentata, 131, 194. Kwa,kiutl, 224-324. Labor, 144, 160. Lele'ks, 8, 26, 27, 29, 33, 35-38, 40, 44, 46, 49, 51, 52, 54, 60, 71, 95. Levirate, 49, 302. Librooedrus decurrens Torr., 206. Link river, 7. Linum lewisi4 Pursh., 174, 175, 210. Lutuami, 1, 224-324. Mackenzie, 94. Magic, 115-117, 269, 270. Maidu, Southern, 66. Ma'klaks, 1. Maricopa, 58. Marriage, 41, 43-51, 279-292. Mats, 173, 174, 228. Medford, 8. Medicine, 30, 87, 118, 123, 128, 129, 131, 147, 273. Mentzelia, 163. Metal disks, 7. Mexico, 173. Milky Way, 221. Modoc, 1, 2, 8, 24, 25, 32, 34, 41, 42, 47, 59, 73, 76, 81, 110, 117, 120, 122, 130, 139, 155, 156, 166, 175, 178, 190, 191, 194, 224-324. Mohave, 58. Molala, 3, 4, 9, 24, 29, 41, 43, 53, 59, 71, 107, 158, 160, 194, 214, 215, 224-324. Mono, 58. Mount Pitt, 4, 9, 104, 119. Mount Scott, 52, 97. Mount Shasta, 2, 120. Mythology, 15, 18, 21, 32, 37, 55, 74, 75, 76, 94, 102, 104, 141, 143, 220, 238. Murder, 26. Museum of Anthropology, 118. Music, 88-90, 229. Names, 59-61. New Year, 91. Nez Perc, 3, 5, 39, 224-324. Niootiaiza attenuata Wats., 87. Northern Paiute, 24, 28, 29, 32, 40, 59, 66, 81, 86, 90, 156, 158, 188, 190, 216, 224-324. Nose, 214. Obsidian, 14, 32, 55, 72, 76, 114. Ogden, Peter, 5, 42, 197, 235. Omens, 138, 139, 168, 274, 315. See also Taboos. Oregon wars, 8.L Orion, 219, 221. Otter, mythical, 15, 209, 293. Painting, face, 30, 33, 77, 216. Paiute. See Northern Paiute. Palmer, 9. Panioularia, 163. Paviotso, 39, 58, 66, 208. Payute Indians, 2. Pelican bay, 130, 165, 193. Peucedanum, 164. Philadelphus lewisii Pursh., 195. Phra.gmites, 166, 191. Phragmites phragmites, 182, 195. [336] Index Pictographs, 19, 142. Pillows, 204. Pinus lambertiana Dougl., 169. Pinus murrayana Balf., 176. Pinus ponderosa, 169. Pipe, 87, 88, 127, 229, 231, 232. Pit river, 2, 8, 25, 33, 39, 40, 42, 59, 89, 122, 178, 190, 193, 224-324. Pitch, 74, 75, 173, 217. Pit, Mount, 4. Plains Indians, 4, 21, 33, 42, 43, 111, 196, 207, 208, 210, 211, 213, 214, 216. Plants, 166, 193-196. Plateau, 94, 213, 224-324. Pleiades, 221. Poisoned arrows, 32. Polaris, 221. Polygonum, 163. Pompom religion, 89. Population, 1, 5, 23. Populus baligamifera, 175. Populus tremuloides Michx., 183. Prayers, 91, 141, 168, 169. Puberty dance, 45, 68, 314. Puget sound tribes, 220, 222, 224-324. Punishment, 39. Pups. See Dogs. Purification, 33, 70, 73, 75, 93. Quarreling, 50. Rattle, 35, 68, 229, 232, 236, 323. Renier, Henry, 25, 29. Residence, 48, 53, 61, 284, 303. Revenge, 28. Rhamnus purshiana DC., 131. Rheumatism, 71, 129, 156. Ribes cereum, 195. Ropes, 175. Rumex, 163. Sahaptin, 3, 25, 42, 224-324. Salish, 3, 224-324. Salt, 8, 160. Soirpus lacustris ocoidentalis Wats., 174, 179. Scratcher, 315, 316, 325. Sewing, 173. Shaker sect, 101. Shamanism, 15, 18, 27, 28, 30, 36, 37, 45, 55, 60, 70, 73, 86, 87, 90, 93 if., 208, 235, 239-245. Shasta, 24, 33, 40, 41, 42, 51, 69, 87, 217, 222, 224-324. Shoes, 6. Shoshoni, 29. Singers, 68. Sisymbrium, 163. Sium, 163. Slaves, 22, 24, 25, 32, 34, 39-43, 236, 313, 314. Smohallah cult, 89. Smoking, 86-88. Snake, 2, 3, 4. Songs, 131-138, 238, 323. Sororate, 49, 302. Southern Maidu. See Maidu. Spirits, 28. See also Ghosts. Spoons, 176, 228, 232. Sprague river, 143. Stars, 221. Stepbrothers, 44. String records, knotted, 70, 112, 220. Suicides, 71. Sweat-lodges, 205, 206, 227. Takelma, Upland, 4, 24, 26, 28, 29, 31, 32, 35, 40, 58, 71, 87, 89, 155, 177, 217, 224-324. Taboos, 33, 36, 48, 55, 69, 75, 111, 112, 139, 140, 148, 168, 314, 315, 316, 325. Tale, 4, 32, 73, 75, 76, 94, 102, 104, 109, 119, 120, 141, 143, 206, 220. Tanning, 173, 228, 232. Tattooing, 41, 217, 228, 230. Teeth, 56. Tenino, 4. Tents, 5. Tipis, 3, 232. Tobacco, 232. Tools, 171-177. Transvestites, 51. Travel, 72. Tricks, 115-117, 269, 270. Triglochin, 163. Twins, 56. Typha latifolia, 179. Upland Takelma. See Takelma. Ursa Major, 221. Urtica breweri, 182, 210. Utes, 164, 224-324. ralerian edulis Nutt., 164, 168. Wappatoo roots, 10. [337] Index War, 24-35, 122. Warm Springs, 3, 9, 24, 27, 37, 40, 41, 42, 89, 209, 224-324. Wasco, 3, 4, 24, 25, 118, 224-324. Washo, 58, 66, 224-324. Weapons, 31, 193-196, 228, 230. Widowhood, 50. Wishram, 3, 24, 25, 39, 55, 56, 90, 113, 222, 224-324. Witches, 121. Work. See Labor. Yakima, 224-324. Yreka, 41. Yuki, 42. Yurok, 1, 22, 43, 91, 201, 203; 215, 217, 224-324. Zygadenus veenenoosus, 128, 164. [338] I ??l.?II,. ,__ ..': I .. 1. , ; I11 ?I I ,._'ll,, ",-III -i, I Iii, , '' ?_ "I -, ";,-? -I.,.II. I,,???.,I!;I. ,, 11 ?I.!._;l I III.I,,,'.. ': ? , ,,,t-.I I II.,pI ..II , ,1, I.I-I.,;'?- I ,- II.o?, "",?-,,,",",,I ;i%,?I II?; I,I ,I ,,II?,,,,? , ,,I .II'i ?,.. k,II-,"II?o_, ,. ?- 1? "I,? ,??,, "-,? ),r'. ?,?,.? .I,I??,.,I11 ,. ",,'I,?,?." ?.I,I, I .,?,,'s,II?,.," ,Ij??,j-,,,..,,?l "??, ,)I ?..IIII-.,,,.,II-,I?. ,: ,?f, ,?,,,".i,,,, I.I./I.I.I,I- II,,,;.: ",?-_1 ?l ,-.,",,,??I ".'?.' , ?1- .'. I .,"I",I,L, ,1,?II Ih?".. . I.I.III .II,IIn?.'.,,. 1,"-1 1, -1II, 141?i. ?;,1_? I, ?1, .11 .11",I,.,",, ,II III,Ill? I i,1?,,-,., ?.I?, 4,,?,,, , ,?, ?, ?`,.I,ii,, I.? ;? ;? ??? .?,.?.?..?III.I; ?-,, -, ,I ,I''. .?,,?I-. 11 ?? 11??,??.1.,.11,,"? ?,'.??.,,, 1.,",,I,4;..,.?. ..?I: ?'.?.?l..I I, ,..".?,..I? , `;J ?,11;I?.1..-, .,?.II1, ,,?....,:,.?I, 1,.,,,,,.:,,?,I.III ,,:.." ,I 1, ?,.I. 11.I.,IL,,, ..II1?'. -_ ?l..?,UI_ ,? ff.,?I?II?,,,.I,I If,?lI?. I-II?..III,I.I1.?.I?;.. .?Molll?l ?I I?1, (,?1, sI?. I; .??, ?, ,,,IklpT-1, 4 1" ,, ?" , .":II; ;Oraw : ..-1, : - ?, ? ?, IIII., -.1I: ?, INO , C.&=c I .:, V?,. ," III..I.?.,,?!,, ? "-,.?.- 11?'I,;,'i - ,D.. ?1?,1. 'Ws .?, , 1. "I'l ??,.,- ?? ??.USITY I..,,,''? ,??._':-I, ,?':,, II:, '".I,1?, I I..-.?I,j'' .4"'. 1-?? .1..1.,-".," I.-? ",? ?.I-,??? 1. i,?i, ",3?l,,?."., I " -1I??Ill.?__?',,?:.-I'If ,?, '' 11-II ?_? ?1-1 I 1,I", -., .- "I ?:, ,??'!I,-,?,, ?,''I ?,_ !- 1?__l 11 " 1,-? " _ -.w, ".?11.. "? " ,?? e, .1 ?, ., i': 4. " ,, II",? '1?"? .I., ,'' ",?,f.- 1, .; II .? ,?I:,! .. ". DEP '. ''?,,v1. ''i ,7 ?I1? ?l ,ft -ftiivi?-!, I" ; ?',. t. -I.: 11 I I.-II1? I I _.? ?,. ,? ??I- it), - ANZAA _??:?,-i,??`i.,< .I ?? ,.?l ?, ?, Id,%'. ?'tIPOLO, j, ? I , 1. , ?,,,,` "', ? -,.?, II.II, -?.,Al T' r':I,I-,:, - I ..I`I. ? ?1;1,, , i '11, ,,, I_ 11qI-' _,?,I,_.,,,Q:-''y ? 1, "', .", II.: , ,,, ". " I.-I -? "" , ,,.. ''.-,.1 1, ,j? ?? , ,1), I _;,? .. ,?l ... I ,,, 1! , ,, -.? ;_ 11, i.. I I I `, ,. 1,II," ,,? -, '. I11.:,? I., ,? ,.??. ,-, _?',II" ,_? ,,,-,"?. , ,?l ,III,,II,.,. ?:,? I,IIII'';?r.,,,""II? ,,,?,I,4?1. , ?? ,,I.'f? ,,I.,It,,I ,_ ?,,"? ,)".I,? f. ??III..?,Z?tI".?..I?',II,_t ? -?. ,? " I ",, I.jI., ? .?02 "(I .O--- " fiii' ??tho.?. .,.vtAIO;.I-?, .,, I., lj'? I'? ,??.i` ,UI ? :, "F , ? -i , ;f rr?lb? -9 - 6tIjI I ,I, ThpFjpUb,UcaUwX9 410,,II I*A ?.I.???? .I." -gd?66, I,rith W6 .6 ibloii,, III11 111.I_'! , I . "?,? W#1 ., (., II,;.. ?I .,"I?V_ 'L l6i)?16, , ?,, ? _?,,?_ ??? .1.I?, I I '' o ??, ?,: ? '?:",.,?- sl?- , ". ,, I?_, Ii16, A411 Ur SOIX-. ?,"-''j 6t tlt6 1,I?. -'. ?.. ". I?' ?;, ? :'Alr6t im di 'Illt?e ? IMPRttment, -of". 0:,iii,, t jn4 - -,'b I" " V, -, ,of"?-l -, ?,. ,;,'-,III I'', 11 ,,, I11,.I...., ."I . ..?4-:',?, 'i- -, I1-Iti,-.' - ?_..4.-I I, -,? 1?.- ;,? I ,'?_'- ,. - 1?I , `14.,.4?I,, ???' -... 40aI I -=. im " - . ", . ?-,' , ", ? II,4.ute-ova ?Mell ImI-6t6a',i4, ota - 'id'. . r ", bXopolol "a, I .;til ''. ?;, .??,WI .. ?'! ? ?',III10 "";-4 41 , I J???i ov, _& %II,'-,: ,ml ?, e, ,,?]-? _?., .1II I "',-I, ?, - .",ll??I-' `? . rI I?I'll?. .Il,I I- 11 '., I ..I. IIkar ?, for",l?I--1? .,:,, ?1: ?- ?6'1?1.6 '"A', ftAed:''lu,lh,o',',Og, II'.-,;h?,tlio,I ?l ?l -. .or uchaeolog,, ??, liia?'41t'lllli?i ii? I , i, : I,.I i ,Y-, ,6tkiio . ?I -?, . ?l..,,? '. ,., ti 1. 1.11., I I .. .I ..II, .?? I ".?.,",, i ;? 1? 1? 11;?l 04", - ,,: be, Aftt ft aiid-mf, I ,f ,!?? I .- II, I?."I ?.,I.""'I -I I . ,.?of h -on-II,?66'161 0?,,.0L? .f .fl 11 'U-.., r, ici -, "a -1,, ,"I -". ?? ,.t;e_',-'kubl'!c4,4` 19`01,thiip?;' l!.jij4 ?4 'Welid, ?-,-- 'Dj*1 ?'w Im .., -" ,,?. 1?! :1 FI I..", "I.-1:1 P?, ? ? I J,?l 1, I ?_? ,? 'I; ?, i'? fI VI i?,? ,_ - ?, ?. ..1III.-II??i .; , ;o; ,? ''I- ,a] ? ,?,_ "idI.cft 6? II I'"XIOB, 1? lbz - 'I O,I-? T='Ili , " ;P," I''I II-4ith 4gci , tb IIee cI 1. 1.?--tPo,lrequo#. -' 'i '40 a, i? 4 ??e6, i T-. : ? ? " ? ??IO I-I .11I1". -? ? ? -?'I_ lu , ?,,? I'bf,'I_:-,. ?,?.j ? 0A lmlt*l,", I, J?FtLI'S i.& ?, ?"..., .1 Ift . ?I?I ..?1Jul_K72 1 1 11 ..-RA ,,? / ?..I I,??1?I-j ,,l ,.? 17?- II-? I?t ?I11, ??,-, ?JL t, -???? "ij ., "'., ? ,.. I-- ?l I ,, ..?, T's-RISI-,.0 IIT , ",I BI.,I??tl ;;?11 , 7,,"4la ,II,-.. I,?l Lt, ': ?_ I?.. I I--1'?f '' ?. ?II.-,, "i I ,. .I?.1 '', ,,,, I r', P Slap''??-- II?e, 4A,?-OAIL.,j?j,.t -I I1.,,I I I ? L - , ??-? ,,I'?- ,I . i0essidd t* ih6--I?4?$4.&, P. ,,,, ,?-?? , ", ,i: b?04 tll-o?.- P -?',,,I.I III ...I?, I.1.III",I, ..: ?P? I'll''.I? -1, .1?f,il ? ? :?,` -', -"A???.. ? ?P, - ,-??:,?'l ,.??"- I-I,I ?-' -I, '' , '',,_,,,-I-, 11I,-,--I": ?"I II --? ..II._ , ;?- I -... I ' , v-.I .?I?- II __ - ?, i- I .,, obl - t6m, 1. ?c 1. ?, ? , III II??? I.. 17,,.1 ihi4j"' ]al: be ?wno Altmiii ,,, k;_,; ?, ? " ?, .1. I1? 1? ,?,., I-I If'?'.Of% , 1'1?.??L Ionq,., o"11,aP _* ''T 7.-?: '.. -U?aii ,,t?144i 0440 !? :, 1, ?ab I-EA pio?i -? -1, ??, ., I ., ., ? ,I,,1..I--1 , I I ,Ili _? 1- II-I *rII III I''I .4XD6 ?, -,,z. .,;?..,?, ?._:i??'Ill,, ?- .11:?,? I. I,mp I ?,I, t3LI?. ,q,a.iA ?l .1? ,,sp-rPOS, Ui ?,xv-',%u' 's I.X,c, ?, - ,'? , ... ''; , - ?i;, "?111.I11 '' ,P ,,,, ).Ily tL '11,;-11i12,? 1 _,I 1?'. 1, ".I" , "-;? ,?`. - ,,, ''I',II,, LL,.,h*%A.? 10, 1? -1I? ?I,I-??r I . ""-?:-?I ,?".1-1 11 , ,-, -', :- ?-"...", arlit, ( - ' 'i-Q,.14?6n ., I1-. :. ,? -"d'';,?,'i;'''?'?-`?i-- ?.- '?:j`4 "a ",II 1...I ?I1, _..1.I.",?"_. ""',,-? ??i? I..0"I I I?? ? -n,, t --- -bi, 4u&.) ibiul'4 ?? ?"IIIiI,7,? 77 ?. ?I,?, -`-?, 1, -,;i;-'- -.-,"",-?.??,? ?_ I ?l - '' ?-_?'"-.. ?? ?; II?? " ,- -,4, ,I ,,.1, ?.;I- ,I" ?.I-I,- - ?l ?l 1. -II. I?- ?. -, 1, : i-, ': ?,,?.I,,- ?' ?.1I, ,,::`?,,II,,?, ", I?, ;, i..'' -, ? I., 1, ?, - ,,,. "'i ?"I I I?, ? ""; ":, I ., -?i ,??. ?! _,".. .. ', ??',? ?? I" ?l,.,? I ?!?."I,,??-, .1 "', - ?;,,,? ?- --'.--- .., -,.- 1.?'-,,. ? I..,:, ..." ..-m- 1,?%I, ??, ? ?l "I??!` ? ?!k' . - , '' ?` , -II'll.. . ?',. . ?-,. -.1 , C, , .1,. ,, &,. .Ut' roeber'..'a4d1o -?* '. ?_,? II%'? `A= PJl " ,l AA'C ktik ;60 P .i?b*,,? t6 . ? .I I ? - I II? PI I- .1 ? ?4-I..'' ,- ;.- ?. ? II?l " -,t.7,''I I - 1. ,? I--_1??, .'t,,.?. ?ll:)i,:? ."- - I.;?? ljj::??;- V,144?"?' 7, t',I1. ,? I;I;-,, .,) -,,? It&-A. Iw). ,; ,,?I ,?, ?I 1 d"a,& ig.,?,',." 'Prf ' ' _' 'I" it'a, . 19' t6i -'ii ",;,"I'o - ?4; I,,,.I , I WIP ixsilv O to .;..1 ,? '' , ?,'_,"1? , , 1*4? "I ?11 ''I -; ?4', .L, ,?, 1- ?_??lI, - ,. ,, I.1 I.. I, ??,?', Io? 4._.,:I,II.,`?I":11 I II. ?. ..,? I ?i. ? ",I,..?( , ell?_e,%Ie,141, 1, I?1.-?..,,,vl? 0.1 %, '' 1; I'! vlolwn -:Q?%.f. -t? , Iz doll" 1v ? ,?-i -io ", h- 40 11,,, - II _ '' ? ,_ - _P?-I I _., I ?III"11 1: ?.? ,.. '. ?-II -I I).. 1. 1.I,?1, '"'t?- 1;1. ,_- , " ?'_' ?'_' .1I,? ?C: . I IIII ,I1? ,I- ?I , uillim ?.,94 ?-2k aiid -", -w ?' .00. - 4 ?'Ae ,jjv6luv?sl` ?l ? ?- I ,,1, $4. ' lpnin .-.,ft ix . I.I.- ?, 1..,_? '. I - ,? " i, I ,2" ?,-, ?? 11 ?_ w I?_ I I_"?'I" ?,g ? " ''--.1?io, . Al?? 1.1w 1?.., , -III?II .. 1. 'I",_,I? ?'' ,_,111.iIII , ?, ?,.",?1. t "",?,-'it 1,I 2121 ?, III:,??.I..I?;,e."I " -?!.I? ?.-I.I I, . II. .?, 1,.'Of., ?,,-,I.. , I 't 061 I, kol Irlitels I , 86'1amgi u:,-'?;-?:- I, I1.-`; ?,?, '' I- , , , "A " ,O , "' "Id W1- I ,- - 11 11.,- ,,?II.,,I?.0, II . -II ?? t, -.,---,?;l ?,,,? I'i? , '',I,? -,I,,.- ?, '! ??? , , 1,:,?. :?. ''I?, ?, , ,,, 1,,?., .., . ,i, -, ,,:",I?_,I.,I-II11 .I.I?-Fb ??".. _?!. ? ?, ..,,-.,. - ,:',.?,,.,? II??II: I1..,, ?.- o,,.?,"I . .?..1?11'?_;,?-1,?:, f?I1?-,,,, - -, ?, ?, - ?'? ,11 ? f, ,?"-_ .?,;,,. ,, ,, ,4- ? ? .I.I?-I ??, il,.,;? I -I." ???,?II". I I. ,, ,,??-,,,, ,?, I "? (I I ,"?? 11 11 il !,,,?I.I"- , I." ,-: I?? q :? II I,,,I??,,-:- - ,-? , ",- I-- ,."_,"- I"-1 ,,k, ''-,,.. -?.?- I.-?, ,I,..?- .?, ,i, 'A .,. ,, ".1 -.1 ;111 I 9 , ,ill ? I, ,,.? : ?_ ,:?-, _?.'' 11 11?'201"; ,.,;":,, WI) I.1 ?-'?., -1, I , ... ;., I.,_ , ,.,- . ", ", -''.", ? 1, ,:,. . "' 1.?I, ..-,- ?' - ?,-'- ..e ,IV' ?' . ,. .., ...l[ft+, ., .,?.'no-. ?,"'i,? II-VOL,oWb ' A" .Vo , tv '01%fte,I2-,?'?? ?,?; ??11 !, 1, ??;-,,,1, ,,#%, olo. 1, - ? iqII-i, , -'- _.. . , ,,,,,, I.I.) ? .1 I ?, ?.r?,,,I.-I,-.1,?. ?? ,?.I-ll??,, - -1?I ;I ,,?,,?"I??. I? "II ? I??__rI ?," .I? 1121.4, r"_ ",`J4',tft-t-?-,-_, 3D,eqppllb6am',,. --- ?.,- -----?, IIII .I %-,? 1: ,J ? ?_I:l II ?i?.?. Ir,, ?,? II.- _1_7., -? .I... I j,. AAI,'.-,?,? ,"', ,, ",I, ,?,,'I '. ?..? .'IrI? "??!",II,?? .,,-, J: ?? . &tjd''*jgjjjj 4?_ I,?. ?.Il ; , ". . O I'l l . - - , , ? , ? r'l" ?' P -= , ,, ? " ?] 1, ..- 77 1,,4_?I', ;1:,I'e, .??','_ ,,,,? ? , - .. - ".,,,?' .',, ..11 ? ,I"': -?,,,'',?,.?,.,?''II1?,?. ., U6.1ir .. '. '?'., ?I , .,I? .1. 1 ,1 ,-.1: ?.,;- . i1- .?, .: ,,, . 'l ..I. ?- ;?.. ? ?:? ,; i .I ??r? ?I".?,-1.II? '' X.1 "I", I.; ?. ??.,??_ .` I.,II,'?, "I ?.l ,-".1? ?' ,; -1. i,; 6VI,I'' 11 ? '.I P! 'b " "T& ro #I , , , '' ill'.,: li'll. "? , ,orble In, , , , i?? ? ??"` ' " jrjj?:j , 00 eft.I? ?:i? ,. .:-Yl'.L ."If ?r,,?,; ,4. ''I', p " ,"#* `1. ,? A.,.1?1 ? ? 4_.?, , ""',,-? ''.I. II?I-?'IV'"t,-?I, ,.1 i. iI .. .I..?..7 ?. I1 -,? -. 1, : 611?-I.? .., 11 ?, ", ...? - TA' "II ,,?,?, I , ??- gum In'' t-,.-I ,,?: ?1; I: ,;- ?? -L-?tx6ng. , Pli ??.,:,'' ,?- " '. . ,,"Ws _'J-?4Z:P. ,, L-,I ?11 .. I 1, -,-II -.I I?I?-I,II"?I?'. .ft, _;. . ,.,.?.. i- -?,;, ?l., I,,'' "Li.- -'? "M , '--, I. ?I.?. I? ?, ?I /-?YI ?X 'I , ?? " ??,,., ,,, In ?.tq.?_ ?. 1,; , ",-, Ili ?I I , I.?-e f1glorbIII I.. ,1?I1,:11 2, ?": .1 '?r, ,,? A. . t:;,, I.r.4iqOlu?01A 1; ,;: ,,JW ?.?_?: ..1-I .II??I?,.- .1.1. - 1?_,b,'.,., ?, 7" li 1,?.1,,1-? 7 ,,, ? , " rol ,? ?? ..,;rl"I1".-, .. I I9,. . .I.I I? ?I?.?;., ? - ?'. ' '?,? 44"-?.,?,?,I,",",,,, -? -"'.,-". , ,I.I,., ver.?. I-,,? .j -I .llr,?,? 1,,?-- I't ' r194 ..& -, 4'.. ,,_.,,?. - ?'. -toI ,?.? Sosf,-l."AAd- ?66? Iptoinbd ?' eI,f, .12, ,? I "I' ?I ?. ? ,!?, 1.1? ..I ,.0-11 .;? ..I II, ,_ ,-.? 1, "._."."I?_1 ..-I ?-_,.1 I -.? I ",-,,.,I?_ I.,? .." .." . ?.I , .- "I-, -AI-1?? " U, I1'.,?L_I?i -I 'r-? " `,? ? & V_ I .1., 1,?,I" " '' ...,1.?, . ,?I- ?:,? ,, "I,I .A,r; '"1. bnil "O ,OU'?- - "I, ,'' '.-'.I1"O-?..Ip.!, 1..,'.: II. I 1:I,?,,Iel - .00 oh$ tiI :.?'-11.1 .3?lim II , - -- , - L.1L.1?'"?11.,I..?:?I?I? .I?,?il -1 -1. ?l,; II?.??,:,,?, II .?,.? ,,,, / -,'; _? ... '',?I'jII- II..rb- . ?,''; 1; ., IIII.):? Y.? I.1 : "I,?, ,, -I-ca2u? Won" ?Illli??-.06.,193- ,'!!, ",4 ", . ,,, i -,.1 ?'7? , , Ifiii,,! .0'A t6l?oi' 3ts''ix.?, ?,I_11 ''IV. ..I,, 11 .11 I ,?,.?! ?I,. III, .I ."a,*,??loa,l " -- 11I1. 1..?.i. III,,,?:'',,??.I?I... ?, I.? .1?.II.I- r,?".1? : , II II?I. I,,L ??;,? ?..- If.' ".,., ,-I..---;.?_ .."..I-,? ,c 'l, , ?. "'' 2"Al,' 47? Aj?Liei,?6. ?,Doc4w ." .,,.8 1, .- ;;?11 ??. ??, ?;Ioe",?.,. I,P--, 7 t ? I,- ?, F.?!_.,?.t..!?. ;I;I.4=, , "?..?t'?'II'D ., --, ?,?.,-?,_- ":,", ,: .-1? ?.? i)-m ? -1'-!??`_'?, ,,.. ,.jI? I,",strol. ?,_, I;l '' , ,?'k, I1?II,,)au ...11 .1.-my"""WIMAIA. lp,,I:,.-??_?-.I III ?'_ , ,,, ? ?%.,.,?II -?16644 .?,I, ,Al .. ,- -'. ., ?, I '.;'t?,,,, *u6oil ? I?. 1; ., ,, ?.,". I'Iiol.?, , -? 9 ?Illt;ft '0- ? 1. 1? ?...lI? I -11I? -- ???l ,,,I',-,?;:: ?,_*:,,i) I I4, "' " -.? '"?i,:, ?" ,? ?-: '' ? I1:1I"136 d I,4," j- _""'.," ? -I IiI II ?'06`r-?? 1?# ' "A" II ? JelII'", il':l' , r " ?' _PP? .19 ?`?'gtlr b -',',??S','A,-: -:? , ?? III?I-, ,,"" ,?-- %, ?l III_,,,,, ll? 1, ?, - .-! ''i - " ??1. I ',?",.,''I .9, ?-? . ? P ?l " 11,.,,;?, - , ?1: ll?, i ." , 1? .1 . ??? , , , , " ,,, fi .II?III.?l?, , " - .? ?4 ;.,'-iI?41vI , ?.11, . ,"? '' i---' ,.1?I1- ,.,. -II &"Th II? ., - -,"r ' -''Oo '. j?do'ohli, I ? 1?` : i?"-_ ,; ki6 '',; ,I?l 66,:O?,?.,-- lw?&.I, ',Pl? ?', "" ? _,II , 2 2 t' t,i '' ,t ,;,' '/ 1 ''_ ' t 2 ! f - f, t ,U ' ..-' ' . NIV T CAIP t: ' OBNIA PUBVL A Ot(S-(C ." - - ' Vola 24.l. 1. The Uble 1tter, Ceeti froni Naz-a by A4 L Gayton and A. I.. % X - " " . . K rebe r~ P.. . 1 -46, p lat es 1 -4 2 1 2; lig ur in * t e t Pb b rw , a1 r, 1 27 ;. . . 6 , '. ' W: ... . ' ' - s 2.: P etr o g l y p h e o f O a .k l l fr* a n d A 4 joA i n tg - S t a t e s b y J u l i an'H . S te w a" d r " ; !; t | ~~P p . 4 - 2 3 8 f ro n t ispe v ( i n c o losr ) a i p la t es 2 2 9 , 9l Z f i g g r es I nt e x t , {1 , +t K1 '; 49t m a p s. S ep emer, ^ s z - ;''"i' -,_' ' -,___* ,_ r' -__^S ' - '- -' - - -J 2 '.50 5 7: 0 ,0r 5..:okuts and:d WestrnV Mono ?ettery-M akln F, b A. II.................... Gato- -lp. 285: - X,,. . 7 .-'X.' -' .;: : D '' 2, 5 1^ , p Ia t e:. 9 5K.4 )2 2 f gurs in t e xt,': 3,map . 0 S ep t e m ber''* ', 1 ,.w .{ .'; - . . 40. .,. , : ~4. The V3alley Nisenan, b:. A. 3L. eelet.: Pp.- 253-9q0z-Deme, 19.29 ~.. .50 > f s ' ' '' . ' + ,r ' .5 . T h e , B ' e a r R i v e r D ll e t of A t h a p z s c a n l b y P l i n y GB a l e G o d d a r d . p .! * A, ;0;0 . - ;291.44. December*, 19t9 .--* ?- _.**._.e -~ - - -. . 4 ; -- '' ~~6. P eruavian .C iubro.us B ow l,, by Isabel T. Welly,< Plp. 325-341,' ;figure':' in,,- Vl txt Ari 190-C--..2 9 - -4~~~7. The arqer's Ar of the mdw u f -Notwestern Oalifornia,--by Isabel--T.:-; _ , ', ' i -s, 'd*t ' ,t; K eS I ty. P> p* 3 4 6 p la t ets- 0 31 1 9, 7 fig ures < in,,te,x t-. Ai gus, 1930f . . 40- -iber 190-e-- - -9. Yuki Basy etry, by Isabe l T . Kell ey. Pp. -4 - . s p 1a s 1 17, 5 fure s- --. . ; : i r)183, plates 1 -48 .g2e fig e s n t e. eFe brarxt .3. 92 - -..0 1 ; - ~ 2 . lM e n ta w e i; Re l i o s C u lt, bi B d w j M . L o b P p . 1 8 5 - 4, p la t es 6 9 .7S 3 . b.. + \ (. F e b r a y, 1 5 2 9 . - -tLW#. t, -. ? Xt..8 0 i*z -* - "~~~~8 T ri,,; &sbal' Initltions an S|ecret .Qlte, by Bd ,i: M M'S. ob. Ip ................................ -49-288, ', .. S t ; 1 m ap PFZe b tu y -392-... - w- .50*? S; L~~~~ ~~ A r h a ol g o. t ;; ,-. I,* X ; X G , 90 h e. No r thrn " S a n J o aq i n V alle y,b y W . E g be rt S c h e fl .o kr\ " *a ,nd . l. W imer l .... ...................., V ol. 26 . Abo rgin a Soiety n S :outhe rn G oi n by W Aillim DuncaS tr o g b A.+ * >: . A358 pp., 7 maps .... . Mvy; 1929;.y . . .. - -*-.-. . -- -4 .50 . --- D 0 ; ^ ::S8tpple~ et -A ut4or and Title I nd4e Z Univer sity of Califoa, P ub-: . .-, : lic ai n i n pericaa Ar chae.ol egrah 4 thology, - V ql tues: 1l2,. .1 9 03-1929. 1-6, pp ., June , 1 92,,9 . ... .,- ,,, , . Vo.- 2-7.:. A GrTammar of thie Wppo LI gage, -by - aulladi n. viiS+ 3A94 pp . November, :1929. .,.... ..... 2.8 --;VoL 28. 1. Chiu nash Preluiatory,- I) kn L . Ol0on. Pp. 1-21, 0$ fi:g ues in tet lI: 0 m: ap . Janua ry 19>- ; r-f :. -:Textile Pe riods in A ncient ;Pe ru, by Lila M. O'N eal andV A . L.0 oeber.- P . ;23 -56, paTt e 148, 33 fi e s in text F arch, 13 -. .6* ' ' i:~~.: '''--3* The Ghos't Dance ;of i870in So uthi-'Centtral' C aiornia 'by 'A.- H.Gayton.'--2': : :P p 57 -8 2 2 f i rs i te . Mac 19 30 ...........--3. Vol ; o29. 1;. 4rchaeel-ogy-;of Sth>DalleaDeshute,sRegion, b3r W Dunacn Stron, W. . Eg< 0 t r Rbert Schen ek, and Julia n; 1. D Steward. r P p i + 1- 1t pla tes 1-28; 22: :I ,-:fures in it + , 1-map. Nov mb.er ; ' 1930-.I2 .00 o. -0 . ;:KfLi,-a mkath fthnpgoa fhy, - by s lieS :S . Pp. x+-13y22 f gr in te xt. er 1930. ..cbe 19 30 -- -- - ------- i > ^./: o