YUROK MARRIAGES BY T. T. WATERMAN AND A. L. KROEBER UNIVESITY OF CALFORNIA PUBLICATIONS Iw AmraIcN AREczoLoGY AND ETHNOLOGY Volume 35, No. 1, pp. 1-14, 2 figures in text YUROK MARRIAGES BY T. T. WATERMAN AwD A. L. KROEBER IN 1909, ONE OF THE AUTHORS obtained numerous genealogies among the Yurok of the lower Klamath river in northwestern California, in connection with ethnogeographical studies subsequently published.' From the somewhat tangled mass of these genealogies, a list or census of recorded Yurok marriages was then compiled. This census we have analyzed, Philip Drucker giving us competent help, with the following results: J ~~~~~~~~~~~I Two forms of marriage exist among the Yurok, full-marriage and "half-marriage."2 In full marriage the man "pays" for his wife and takes her to live in his town and in his house. The children are his: even in divorce he is entitled to keep them if he refuses the refund of the marriage payment. When his daughter is married, or if his son is killed, the payment goes to him. In "half-marriage" the man pays less-nor- mally about half the rated value of his bride, goes to live with his wife in her father's house or adjacent to it in the same town, and is more or less under his father-in-law's direction. The children belong to the wife, that is, to her family, and their bride price or blood money goes to the woman's father or, if he is deceased, to his sons, the woman's brothers. Half-marriage is legitimate and carries no positive disapprobation; but it is presumptive indication of lack of wealth and therefore connotes relatively low social rating in a society which equates wealth and rank. Incidentally, Yurok half-marriage is of theoretical interest as an example of a mechanism by which a patrilineal culture might become converted to a matrilineal one, or the reverse. There is not the least suggestion that such a shift occurred in northwest California, or was impending; but the change is conceivable. The census count shows that 97 of 413 recorded marriages were of the half-type, or 23.4 per cent. If marriages involving a non-Yurok are omitted, the figures are 85 out of 356, or 23.9 per cent. These figures suggest that either the Yurok proletariat, or plebs, was relatively small, 'Waterman, Yurok Geography, UC-PAAE 16:177-314, 1920. The maps in this underlie the present paper. 2 Compare Kroeber, Handbook of Calif. Ind., BAE-B 78, chaps. 1-8, 1925; Water- man, just cited, especially 208-226, and passim; Goddard, Life and Culture of the Hupa, UC-PAAE 1:1-88, 1903. [1] 2University of California Publicationw in Am. Arch. and Ethn. or that only part of it half-married, the rest of this social class entering into the low-payment full-marriage. The latter interpretation seems the more likely, because really wealthy people are so well known, and are so often cited as such, that the frequency of an aristocracy was evidently more restricted than full-marriage. Half-marriage, it may therefore be concluded, was in the main a function of factors other than complete poverty. What these factors were can only be conjectured. Personal inclina- tions, it is likely, would be influential. If a young man formed an at- tachment to a girl and his father or paternal uncles objected to the alliance and refused to contribute payment on his behalf, he would normally have no choice but to renounce or half-marry her. That the latter is what would sometimes be done is expectable; particularly, per- haps, if the young man had been clandestinely living with her, and especially if she were pregnant, as in that situation withdrawal would subject him to fine for seduction, plus a second payment if he wished to claim the child. In fact, in such a situation, and more or less in gen- eral, a family of moderate means and attached to its possessions must often have been tempted to save the bride price by letting a youth drift into half-marriage. A father with more pride of lineage than avarice would presumably go far to establish his son in the most approved manner. But a father in whom personal greed outweighs other consider- ations, or one with several sons to provide for and no daughters to bring compensation, or an uncle with a dead brother's sons to sponsor as well as his own-all these might be willing to let some of their grandchildren descend somewhat lower in the scale of rank. It may be conjectured, however, that disagreement between father and son, precipitated by a premature love affair of the son, was often the cause of half-marriage by young men of well-to-do families. The Yurok say that such things hap- pened; and that sometimes, after children were born, the father would make additional payment, so as to render the marriage "full" and his grandchildren of suitable station in life as they grew up. The question arises, whether Yurok fathers were ever tempted to let their daughters half-marry so that they might have sons-in-law avail- able and under their control, and so that the family group might be strengthened in numbers as the children of both sons and daughters grew up in the father's ancestral town. Informants categorically deny this, at any rate for well-to-do families. They state that the desire for "purity of good blood," that is, social status as founded on property and resulting in ethical conduct, would have prevailed. Confronted with the situation of a rich couple with several daughters and no sons, who would be alone in their house when their last girl full- married out, and who would have no grandchildren about, informants 2 Waterman-Kroeber: Yurok Marriages find the answer in an established pattern of Yurok society: the old man persuaded one of his sons-in-law to bring his wife and children to live with him, in return for which he made the son-in-law his personal heir, and in the face of this published arrangement, his brothers, and their sons, lost the claim to inheritance which otherwise they would have had. Or a father might insist that his only daughter should remain with him, but he would say publicly, on accepting the half-payment, "kitoksi- mek' netsnii'u (I shall respect my son-in-law)." This implied that he would not order him about, and that, when a granddaughter was to be married, her father, his son-in-law, would first pass on the payment. This payment then was handed to the grandfather; but he divided it, giving "half" (part?) to his son-in-law for the latter's father, that is, the bride's paternal grandfather. In another situation the girl had become a curing doctor before mar- riage-doctoring was lucrative among the Yurok-and her father flatly refused to give her in full-marriage, since her fees would then be re- tained in her husband's house. He did consent to her half-marriage. When, however, he took not half but three-fourths or more of the tendered full-marriage payment, some of the bridegroom's relatives who had contributed property in making up the payment were inclined to demur, but let themselves be persuaded to agree because of the bride's earning power. Here, too, the father-in-law uttered the formula of respect. A different situation is that of a father who feared to see his "treasure trunk getting empty"; in other words, he was stingy, and refused to help his son to full-marry the girl of his choice. This son was proud, and he worked four years-with rigorous sweat-house fasting and crying for wealth-to acquire in his own right the necessary property, and then with his own wealth he full-bought the girl. It seems clear that the concept and practice of fully and highly paid marriage as a thing of worth were deeply rooted in the Yurok system of social ideals, and would have operated powerfully to prevent the forma- tion of budding matrilineal foci in their society. Even when circum- stances interrupted the continuity of male lineage, the basic point of view was reaffirmed that the desirable and moral course was to pay fully for one's wife and to keep her and one's children at one's natal home. Only thus were the best standards upheld. The matter of residence might be adjusted, and so might the amount; but every compromise reaffirmed the standards. II It might be queried whether eighty years of life under American law had affected the relation of the two kinds of marriage, especially as the Yurok of today sometimes say that all their recent marriages are no 3 4University of California Publications in Am. Arch. and Ethn. more than half-marriages. They only seem to mean by this that, little or nothing being paid, the marriages are not full in the old sense of the word. The statement is inexact because true half-marriage involves resi- dence in the wife's house and control and ownership of the children by her people, characteristics which are certainly not regularly part of the civil or common-law marriages contracted by many of the more or less Americanized Yurok of the past generation. However, it does seem con- ceivable that the breakdown in native institutions which inevitably began soon after 1850 may have caused some change in practice. For this reason genealogical marriages have been segregated from the general ce*us data obtained in 1909. Nearly two hundred such marriages fall int6 generations within one lineage. These generations vary from two to five; the lineages concerned are from the towns of Rekwoi, Ho'paiu, Si'iL, NRgeL, Ko'otep, Pekwan, and Wahsek. Absolute dates are lack- ing, since the ages of persons are uncertain and generally were not recorded. The results are therefore only approximate: an individual three generations back from a living one might be an actual age-mate of an individual two generations back in another lineage. However, on the average, people reckoned as three generations back are presumably about a generation, or 25 to 35 years, older than those counted as two generations back. The results are as follows: TABLE 1 MAnRiAGEs AmoNG FiVE GENxRATIONS OF YUROK Generation Full-marriages Half-marriages M W Total M W Total Generation 1 (oldest) .............................. 5 0 5 1 0 1 Generation 2 .............................. 13 4 17 2 5 7 Generation 3 .............................. 27 18 45 2 12 14 Generation 4 .............................. 37 15 52 8 7 15 Generation 5 (latest) .............................. 15 16 31 3 8 11 Total for 5 generations .............................. 97 53 150 16 32 48 The greater proportion of men recorded in full-marriages and of women recorded in half-marriages is expectable in genealogies, because these are the people who would stay in the home town and be remem- bered there, while the half-married men and full-married women would be scattered abroad and less likely to be remembered in a genealogy centering in a given locality. It is clear that the proportion of half-married individuals remains as nearly constant as could be expected from figures as limited as these: generations 1 and 2, 8 out of 30; 3, 14 out of 59; 4, 15 out of 67; 5, 11 out 4 Waterman-Kroeber: Yurok Marriages of 42-never very far from 1 out of 4. Since these data probably refer back in part to about the year 1800, it may be concluded that the part played in Yurok life by half-marriage was a relatively adjusted and stable one, that it was not a transient symptom of change, and that it was not appreciably affected by Caucasian contacts, except so far as the whole native system of marriage, law, wealth, and standards began to crumble by about 1880 or 1890. The average Yurok town, it has been computed, held about forty-five souls. All these might easily be the descendants of two or three brothers, or even of one man, a very few generations back. Often in the smaller towns they were all blood kin-plus their wives, the Yurok say. Neces- sarily, therefore, they married outside. It has several times been ob- served that here are nearly all the makings for true patrilineal clans of the localized type. The only addition necessary would be for the inhabitants of each town to include a recognition of their kinship as a generalization and to become averse in principle to the marriage of members of the same town group; in other words, to extend the actual kinship usual within the town into a fictitious kinship without excep- tions. This seems a slight step to take, especially with a clan name ready to hand in the town name. The step, however, was not taken. The Yurok say that a man might and sometimes did marry within his natal village, provided the girl was not traceably his kinswoman. The census bears them out. Seven and a half per cent of the recorded marriages were between people belonging to the same town. The figures for intra-town marriages are: 31 of 412 contracted with Yurok and aliens: 7.5 per cent 31 of 356 contracted among Yurok: 8.7 per cent 22 of 271 full-marriages among Yurok: 8.1 per cent 9 of 85 half-marriages among Yurok: 10.6 per cent. The series are too small to give much statistical reliability to the excess of 2.5 per cent of intra-town half-marriages over intra-town full-mar- riages; but we are inclined to accept them as significant because it is expectable that the chances of a couple's drifting into half-marriage through an unsanctioned love affair would be greater near home than at a distance. In any event, it is clear that the Yurok were concerned with actual kinship and not at all with fictitious group kinship. They might easily have had clans by making a rather slight addition to their social con- cepts. They did not make the addition, and remained without clans. One can believe, if one wishes, that they once had clans and lost them, or that they were on the way to acquiring them in a few more genera- 5 University of California Pusblications in Am.. Arch. and Ethn. tions. Such speculations, however, would chiefly reflect the speculator's wish so to believe. Actually the Yurok were as factual and particularistic in dealing with blood kinship as we are, and as devoid in their social thinking of the clan-group category. It is unlikely, too, that at the time of their contact with us they were in a transition out of or into a clan system. Their exact, realistic way of dealing with kinship is in accord with the precision of their economic-legal system, the intense, specific localization of their whole life, from fishing privileges and rituals of communal import to personal sentiments and attachments; in accord, in other words, with one of the most fundamental trends or patterns of Yurok culture. Since the average Yurok town comprised less than 2 per cent of the Yurok population,3 whereas home-town marriages run to 7 or 8 per cent, it may be inlferred that there was some positive tendency operating in favor of marriage at home. However, this gross reckoning might be questioned, because of the possibility that marriage into distant towns would be somewhat rare. As a matter of fact, as will appear below, some seven times out of ten a Yurok married a woman living within perhaps twelve or fifteen miles of his home, within a radius holding less than half of the whole Yurok population. A more reliable approach is therefore afforded by comparing intra-town marriages with neighboring-town marriages. A basis for such a reckoning is given by the eight groups or districts into which the Yurok are segregated below. These districts comprise, on the average, seven towns; that is, a man's own town and six others close by. The two adjacent districts would add, on the average, seven more towns on each side, or fourteen altogether. Within these twenty or so towns comparison seems reasonably fair. The results are, for 390 mar- riages of Yurok men: in home town, 31 or 7.9 per cent; in other towns in home district, 108 or 27.7 per cent, or 4.6 per cent per town; in adjacent districts,4 136, or 34.9 per cent, or 2.5 per cent per town. It will be seen that the frequency of intra-town marriages is nearly double that for intra-district, and that this again is nearly double the frequency, per town, in neighboring districts. These results are the more surprising in that a man must normally have had more relatives, among whom there were, for him, therefore, more unmarriageable women, in his own town, where his father and father's brothers and grandfather were born and where they raised their families if they were full-married, than in any one other town; surprising, too, because of the fact that small towns often consisted wholly of kinsmen. Evidently the Yurok were as free from an 3 According to Waterman (map 4), 57 towns; hence the average town contained 1/57 of the population, or 1.8 per cent. Similarly, Kroeber computes an average town population of 45, a total Yurok population of 2500, which again yields 1.8 per cent. 4 Including an alien tribe, in peripheral Yurok districts. 6 Waterman-Kroeber: Yurolc Marriages impulse favoring exogamy as from the principle of exogamy. Other things being equal, they were the more likely to marry a girl the nearer her house stood to theirs-always provided she was not remembered kin. IV This influence of proximity also appears from a count of the inter- marriages between all Yurok towns and the recorded marriages with aliens. The full numerical record is hardly worth reproducing, since there are more than three thousand possible inter-town combinations, but marriage records of only some eight hundred individuals. We have therefore grouped the towns into the same eight districts (fig. 1) as follows, in order north along the coast and then southeast upstream along the Klamath: Tsurai district: Tsurai only, on Trinidad bay; 1 town. To the south are the Wiyot. Opyuweg district: Opyuweg to Tsotskwi; Big Lagoon area; 7 towns. Orekw (Orick) district: Tsahpekw to Osegen; Stone Lagoon, mouth of Redwood creek, and north along coast; 6 towns. Rekwoi (Requa) district: WeLkwiiu-SekweL to Omen-hipur; mouth of Klamath and coast north to Wilson creek; 5 towns. Beyond to the north on the coast are the Tolowa. Turip-Erner district: Ho'piiu upstream to AyOL; 8 towns. Pekwan (Pecwan) district: Serper upstream to Nohtskum; 11 towns. Kepel district: Weihkem upstream to Otsap; 12 towns. Sa'a is the ritual center, but the name Kepel is better known to Americans and occurs on maps. Weitspus (Weitchpec) district: Wahsek upstream to Otsepor; 7 towns.' Beyond upstream are the Karok; and up the Trinity, which comes in in this district, the Hupa. With one exception each of these districts held a world-renewing cere- mony at a given spot in its territory, accompanied by one or both of the major display-dances, the White Deerskin and the Jumping dances. Each district forms also a fairly definable natural geographical unit, in respect of rapidity and breadth of the river, steepness of banks and height of terraces, occurrence of redwood and proportion of deciduous to coniferous timber, etc.; and similarly on the coast. For the Turip- Erner district alone no major ceremony is recorded. The myth of the origin of the Kepel fish-weir ceremony says that this was originally made at Turip and then carried upstream. This belief may rest on recollection of fact, or may be due merely to a noticeable resemblance of the Turip and Kepel-Sa'a terraces as the river sweeps around them, together with the fact that the farthest upstream of the redwoods which are thick 5 This makes 57 towns. The basis is Waterman's map 4. Kroeber's map, Handbook, p. 9, differs slightly with reference to the smaller towns; and so would any other map that might be compiled. Some minor settlements were abandoned before the Americans came, others soon after, and some were only established then. It seems impossible to get the exact situation for the whole Yurok tribe as they lived in any one given year. 7 8University of California Publioationa in Am. Arch. and Ethn. about Turip, stand opposite Kepel. The Turip-Erner district is anoma- lous in other respects than the lack of a great ceremony, as will be seen below (pp. 11-12 passim). Fig. 1. Yurok towns grouped by distriets. The grouped results are given in the adjoining table, which needs qualification by certain explanations before the meaning of its content is discussed. 8 Watermnan-Kroeber: Yurolc Marriages ?0 k C Co C US TIY .q .- X . i b 3 10 cq I- B= 1? . . o X * . .' b 'I~~~~~TNO Co~ ~ ~ ~~$4C vlnll,qC. W _ . .- U k11O 1 1& O g 1 eq W1 1 l dWpVALO O . _ _ co o: 4 a: M2|f at |t m~~~~ * zx.^.4 . . bo~ ~ ~~~ C 8 I I u dgTae WE s$a I~C Oo I -n IX1 9 UMqja r-I CO4 . Co C * _* salalo~~e eq w 0 C ; ce se _ 5bsnXdo > " t > ~~~~~~~* _ .* k*.*. n tAn IHd *N 4c *: C * . k. . . . . . C q Cqn4, : C o - 4 1 0 I E 04d 0 E 0 on?SEO 9 10 University of California Psblications in Am. Arch. and Ethn. The marriage records are obviously fuller for certain districts than for others. This is an accident of the opportunities for collecting. In 1909 most Yurok were still strongly averse to discussing dead people. A good many towns were already abandoned, their folk died out or scattered; others were fairly preserved; and so on. For some towns no genealogies or systematic census could even be attempted. People from such settle- ments enter the record only as they appear married into the genealogies of other towns. A rough check is available, however, in the number of Yurok houses enumerated by natives.R These total 327 plus, as against 356 Yurok- Yurok marriages. The houses listed for each district are shown in the first column of the table. Their number ought of course to agree at least roughly with the number of marriages, district per district, if the mar- riage data had been obtained uniformly. For five of the eight districts they do so agree. The discrepancies are in the two southern coast dis- tricts, Tsurai and Opyuweg, for which the marriage data were known to be specially deficient, and for which the number of houses is twice as great; and Turip-Erner, where the disproportion is two-and-a-half to one the other way. When first seen by us in the first decade of this century, this Turip-Erner district of nearly twenty miles from just above Requa to just below Klamath (Johnson's) was very sparsely settled by Yurok in comparison with the stretches above. The Indians explained that the great flood of the winter of 1861-62 had washed out most of these lower towns-the terraces of which lie relatively low-and that most of them had never been reoccupied. We were somewhat dubious of the full validity of this explanation because of the scarcity of house pits at nearly all the lower towns; also because some of the ter- races are too small, at least at present, to harbor many houses. Moreover, the dense redwood and other coniferous stands along the lower river would make for poor acorn-gathering and deer-hunting, and the slower river would provide fewer prolific salmon eddies. Yet the fact that more Yurok were brought into the marriage census from the Turip-Erner district than from any other except the Pekwan one, suggests that the Yurok may be largely right in saying that this stretch was once popu- lous. The large towns probably were Turip and Erner, because only these stand on flats or terraces of any size. One of the most important points established by this table has already been discussed: the Yurok propensity to marry near home. Of 390 men, 139 or 35.6 per cent married in their home district (of these, 31 or 7.9 per cent in their home town); 136 or 34.9 per cent in one of the two dis- tricts adjacent (124 with Yurok wives, 12 with aliens); 115 (93 with Yurok, 22 with aliens) or 29.4 per cent in all the remaining Yurok dis- 6 Waterman, Geography, as cited, 206. Waterman-Kroeber: Yurokl Marriages tricts or among peoples of alien speech. In round numbers: "at home," 36 per cent; near home, 35 per cent; farther off but within at most fifty or seventy-five miles, 29 per cent. The figures for women would be some- what but not seriously different. 0u **G . .wys 0 0UA' WIYOT \ "< Fig. 2. Prevalent directions in which wives are sought. The matter of adjacent districts presents several points of special interest. The distribution of Yurok settlements is of course linear, along river or coast; but, because river and coast meet at an acute angle, it may be described as taking the form of an inverted V on the map, with the Rekwoi district, in which the river debouches, forming the point. Northward along the coast are the Tolowa. Rekwoi therefore has three 11 12 University of California Publication8 in Am. Arch. and Ethn. adjacent districts-two on the coast and one upstream-in place of the usual two. The figures show that Rekwoi intermarried heavily with Turip-Erner upstream and with the Tolowa on the coast to the north- which the Yurok call "downstream"; but they intermarried very little with the coast Yurok of Orekw to the south, although the nearest town of this district, Osegen, lay only a few miles from the river mouth. In short, the stream of marriage flowed in a nearly straight line down the river and on along the coa-st. This, in reverse direction, is the route which Yurok myth makes personified Dentalium Shell take in his travels. The coast Yurok connect with the river Yurok by skipping Rekwoi, Orekw marrying inland-the ridges are still low there-to Turip-Erner. At least that is where Orekw district men tended to get their wives. The converse is not true, since Turip-Erner men went to Rekwoi more often than across the hills to the coast. There are some other indications of preferences (fig. 2). Karok and Hupa men sought more wives about Weitspus, the last Yurok district upstream, than the Weitspus men sought among these more upstream tribes (7 to 2). Weitspus however was in the same position toward Kepel: 7 to 3; and Kepel toward Pekwan, the next downstream: 20 to 12. Here however the impulse hesitates. Pekwan, the most populous district, gave as many wives to Turip-Erner as it took from there: 22 to 22. But Pekwan men absorbed 31 per cent of all Turip-Erner women and only 24 per cent of all Kepel women. The Erner-Rekwoi relation of 9 to 7 is perhaps too close to be significant; but beyond, Rekwoi-Tolowa again shows a ratio of 9 to 3. Any one of these pairs of figures might be rejected on account of the role of accident in such small values, but the con- sistency of their one-sidedness gives them force. Only as between the two middle and most populous Yurok districts, Pekwan and Turip-Erner, is there balance. Elsewhere, from Karok and Hupa through the Yurok territory to the Tolowa, the inclination is for men to go downstream for their wives: 52 times down against 27 upstream.7 The meaning of this drift is not clear. For the coast, the figures are too small to be of much value; except that it is clear that Orekw men, as already mentioned, tended toward Turip-Erner more than the reverse. v The census reports 34 Yurok men who married alien wives. All but 2 of these made full marriages. Evidently those who half-married abroad had their children reared there and mostly were lost from the genealo- 7 The "adjacent" districts have therefore been construed as follows in the com- pilation of the foregoing figures: for Rekwoi, Turip-Erner and Tolowa; for Orekw, Opyuweg and Turip-Erner; for Turip-Erner, Rekwoi and Pekwan. (For Weitspus, Kepel and Karok-not Hupa; for Tsurai, Opyuweg and Wiyot.) Waterman-Kroeber: Yurole Marriages gies obtained among the Yurok. On the other hand, of the 22 Yurok women recorded as married by foreigners, as many as 10 were half- married, that is, remained at home. Here the factor making for remem- brance was the reverse. In effect, the two sexes nearly balance each other: 21.4 per cent of the Yurok-alien unions were half-marriages, as against the 23.9 per cent of intra-Yurok ones. The table shows that the foreign tribes fall into two groups in respect of their Yurok marriages, apart from the generic tendency to seek wives "downstream." The Tolowa and Karok sought wives not only in the immediately adjacent Yurok districts, but also to some degree in farther ones; and the Yurok reciprocated correspondingly. The Hupa and Chilula, on the contrary, exchanged wives and husbands with the Yurok almost exclusively in the Weitspus district. This difference seems to be connected with the Tolowa and Karok being on the upstream-down- stream line, as the Yurok construe the world, but the Hupa and Chilula living in a "side-stream" or "up-hill" direction. Intercourse and rela- tions evidently flowed most freely along the main thoroughfare of the Klamath and its coastwise "continuation." On the whole, language seems to have been a lesser barrier to inter- marriage than distance. The Rekwoi Yurok married almost as often with the Tolowa, and the Weitspus Yurok with their Karok and Hupa neigh- bors, as with adjacent Yurok, and more often than with each other or with distant Yurok such as those of the southern coast. It is not con- tended that language was no barrier; but it evidently was a relatively minor one. For this reason the gross figures for Yurok marrying aliens versus Yurok-56 as against 712-cannot be used. The whole Karok or Hupa or Wiyot "tribe" does not come seriously into question for Yurok mar- riage habit. It is mainly the nearest district of these foreign peoples that is involved. VI Our findings may be briefed as follows: Yurok "half-marriage," that is, legitimate but less esteemed marriage with half-payment, residence in the wife's house, and ownership of the children by her kin, occurred in slightly less than one marriage out of four. This type of marriage is of interest as an essentially matrilineal institution in a society definitely committed by its standards to patri- linear descent and control. It was tolerated as an adjustment. It was not frequent enough to have been the regular marriage form of the non- wealthy majority of the population; and it occurred under some circum- stances among the wealthy. It was not caused by imitation of aliens, since the Yurok and their neighbors knew of no matrilineal peoples in their world. 13 14 University of California Pusblications in Am. Arch. and Ethn. There is no indication of a change in frequency of half-marriage between 1800 and 1900. A successful and stable adjustment of the two coexisting types of marriage is therefore indicated. The Yurok married freely where they chose, except with known blood kin. In their choices, they were influenced by proximity. Relative to the available population, they married most often in their home town, next most often in towns in their home district, then in adjacent districts, and least frequently at a greater distance. Alien speech seems to have been only a slight bar to intermarriage. There is an unexplained tendency of men to seek their wives down- stream. This holds not only as between most of the Yurok districts, but for Karok and Hupa in respect to Yurok and for Yurok in respect to Tolowa. It is clear that the Yurok town was not a relict or incipient clan, as far back as our knowledge carries us. Even though it was small enough to be composed wholly of a single group of kinsmen and their spouses, this was looked upon as an accident of no institutional significance. The pattern in the terms of which the Yurok thought was one of geography and actual kinship, not clanship.