19. PRELIMINARY EXCAVATIONS AT THE THOMAS SITE, MARIN COUNTY Clement W. Meighan Introduction During April and May of 1949, test excavations were made at a large stratified shellmound on the shores of San Francisco Bay. 1 The site, designated in the records of the U.0. Archaeological Survey as 4-Mrn-115, is located on a point of land just north of San Rafael. The property is owned by Mr. W.H. Thomas of San Rafael, whose friendly cooperation in per- mitting the excavation is here gratefully acknowledged. The preliminary excavations were made to check the stratigraphic and cultural associations of the mound. At the time, it was hoped that excava- tion of the site would be continued on a larger scale. However, various factors have prevented continuance of the work, and it now appears desirable to place on record the sketchy data obtained from the preliminary tests. The artifact sample is regrettably small, consisting of only a handful of specimens, but it appears sufficient to outline the cultural position of the site. The excavation which has been done to date is of importance for two reasons: first, a C14 date was obtained which throws some light on the terminal period of the Middle Horizon on San Francisco Bay, and second, a burned house ruin was exposed which merits description because of the rarity of such features in Central California archaeology. The Midden Site Mrn-115 is a shell midden on the shores of San Francisco Bay, north of San Rafael in Marin County. The refuse heap, which is about 18 feet high at the center, was first described by N.C. Nelson in his survey of 1907.2 His description is entirely adequate today, as the site has been undisturbed in the past 45 years. Nelson comments on the site as follows: Situated in a small wooded canyon approaching the southeastern extremity of the marsh which continues from the point north . . . the site is only a few rods above the wagon road and lies also on the west bank of the gully. This occurrence is strictly a mound and, remarkably enough, is practically intact, due possibly to the fact that it is hidden from view in a heavy growth of timber. Several buckeye trees and large elderberry bushes grow on and immediately about the mound which is otherwise an open space. In form the mound -is oval with a fairly even contour excepting some indications of pits on the top. Dimensions through the base are -1 P about 95 by 150 feet, and the height is 15-16 feet. No doubt it has been smoothed and lowered by general weathering processes. The interior structure and composition can of course not be viewed without excavating, and the exterior differs in no particular from the usual -- unless possibly by a trace of less earth. The "evidence of pits" consists of 12 shallow concavities ranging from 5 to 14 feet in diameter and from 4 to 24 inches in depth, All of the pits are recorded on the map as house-pits (see map 1), although pit 1 is so much better defined than the others that it may be due to recent vandalism. Preliminary excavation in pit 7 revealed the remains of a burned house, described below. A longitudinal cross-section of the site is shown in figure 1. This is of interest because it shows the amount of alluvial deposition which has accumulated at the south edge of the mound. At the south edge, 5 to 6 feet of midden are below the present land surface,* This was checked by drilling pits through the alluvium with a soil auger to ascertain the actual margins of the site (see fig. 1 B, C). The internal composition of the site, based on excavations at the south end, is diagrammed in the profile in figure 2. The midden was composed primarily of much fragmented mussel shell, ash, and earth. Fire-fractured stones are scattered throughout, occurring in occasional concentrations as hearths, The major disturbance of the midden appears to be due to tree roots from the buckeye trees growing on the site; rodent disturbance is minimal, perhaps because the midden is quite wet throughout. Features Three fire hearths, aggregations of fire-cracked stone with associated charcoal and ash, were found. One was in the burned house, one was at a depth of 76 inches, and the third was 19 inches deep. The latter also bad in association pieces of burned clay with grass impressions, and a few frag- ments of burned anal bone. The hearths were roughly circular in form, the one in the house being 18 inches in diameter and the other two close to 30 inches in diameter. The only other feature recorded was the burned houses The house was of very simple construction, but it is worthy of detailed description, since similar features have not been reported for this region. Excavation in house-pit 7 revealed a number of charred logs and sticks which were all in the same plane, 14 inches below the center of the pit. When exposed, it vas evident from the pattern of the logs that the wood remains were part of a house with a circular ground plan (see pl. 2). The wood consists of 24 pieces of redwood and oak (?), the individual pieces ranging from 2 to 16 inches in diameter and from 6 to 48 inches in length. Two of the pieces were in a near vertical position, sloping inward slightly and imbedded in the midden about a foot below the area of the house - 2 - floor. These were burned only at the upper end; they appear to be struc- tural supports which burned down to the ground level but were not uprooted .-in the fire. One of these may be a crude plank split from a redwood log (possibly by means of an antler wedge, one of which has been found in the site). Aside from this single specimen, the house supports were merely branches which were probably gathered in the near vicinity Adhering to the upper surface of the logs, but not occurring between them, was a layer of burned grass which averaged half an inch in thickness. This was evidently the material used for covering the log supports, although ,uo evidence was found which would indicate the method of attaching the grass to the outside of the structure. It will be noted from the diagram (pl. 2) that the center portion of the house ruin is missing completely while the marginal portions are well preserved, even to the grass outer covering. A likely explanation for this my be found in the fact that the California Indians coumonly heaped soil aroMu she edges of the house, sometimes covering almost all of it with earth.3 In this case, the center portion of the house, which was not earth-covered, apparently burned completely, and when the burning structure collapsed the earth which was heaped around the edges smothered the flames and preserved the lower parts of the wall virtually intact. Scattered in the central part of the house-pit were a number of pieces of burned clay. One side of the fragments was more burned than the other side; the less burned side showed numerous stick impressions, ranging from half an inch to two inches in width. These pieces probably represent clay which was plastered on the inside of the house around the smokehole, as a protection against the fire hazard of stray sparks. Two feet to the northeast of the concentration of baked clay pieces there was uncovered a small hearth. This was composed of one layer of 24 angular sandstone fragments, each of which was from 4 to 6 inches in diameter. The hearth was imbedded in a layer of compact yellow-gray ash which ex- tended one inch above and below the rocks This ash had packed into a firm spot which extended for a foot or so from the stone hearth. Upon reaching the edge of the ash deposit, however, the soft midden of earth and frag- mented shells exhibited no firm layer which could be interpreted as a house floor. The level of the floor must actually be that on which the burned remains are lying, but this level is in no way differentiated from the bulk of the deposit. The remains of burned baskets (probably four) were found 30 inches west of the hearth. The basketry had been preserved in a carbonized form but was in poor condition, having been crumpled by a collapsing house timber which lay on top of the fragments. There appear to be three twined baskets and one coiled basket represented; the baskets were nested one inside the other at the time of the fire, with the coiled basket on the out- side and a fine basket of Catlow twine on the inside. Further analysis and description of the basketry is given by M.A. Baumhoff in the Appendix. Although the house remains were in the upper levels of the site and the house-pit in which they were found was clearly visible as a surface feature, there are indications that the site was occupied for some- time - -3 A after the house was destroyed by fire. At least 12 inches of refuse had accumulated above the hearth. This cannot all be attributed to fill from the raised margins of the pit, for there was an undisturbed shell lens, 3 inches thick and 36 inches in diameter, 6 inches above the hearth. This fact throws some light on the persistence of house-pits, for this particu- lar pit, although small and shallow, was still clearly visible even after a foot of deposit had accumulated in it, and after the site had been aban- doned for more than 100 years. The house-pit was only 5 inches deep and 9 feet in diameter, indicating that considerable obscuring of the original depression had taken place. However, in view of the exposed nature of this surface feature, it seems remarkable that it should still be recognizable. The fact that an unoccupied house-pit can persist alongside other house-pits has an important bearing on population estimates. Such estimates are apparently not reliable when they are based on the number of house-pits visible on a site, since only a part of the visible house-pits may have been occupied at one time. At Mrn-115, for example, there are 12 house-pits. Estimating one family of five for each house, a population figure of 60 is derived. This is possible but would have led to serious crowding in the small space occupied by the 12 pits. The feeling of the writer, admittedly subjective, is that there could not have been more than about 30 persons occupying the mound at any one time, this figure is only half the amount calculated from the number of house-pits. To sum up the traits observable from the house ruin, the structure had a circular plan, conical or dome shape (probably the former), a pole founda- tion of redwood, grass covering, central fire hearth, and clay plastered over part of the interior (probably just over the fire area). This type of house is closely similar to that ethnographically described for Central California. Artifacts Less than 75 artifacts have been recovered from the site. The preliminary digging encountered no burials, and the artifacts represent only those which occurred loose in the midden mass. At this site, the artifact yield was only one per two cubic yards of midden, ranking it as of very low yield even in comparison to other shellmounds in the area. 4 The small number of artifacts recovered makes it impossible to delineate the cultural picture in any detail, However, as some pits were 12 feet deep there should be some indication of cultural change even in the few artifacts recovered. That this is the case is shown in Plate 2 -- artifacts on the upper part of the page (specimens A-L) were recovered in the upper 72 inches of the site; specimens M-V, on the lower half of the plate, were all below 72 inches in depth. A finer segregation of the site is not feasible at this time, but it is felt that there is a strati- graphy demonstrated which can be corroborated with reference to other bay area sites. The upper levels contain two artifact types diagnostic of the Late Horizon on San Francisco Bay. The first of these is the tanged projectile point type, which does not appear on the bay until Late Horizon times.5 Another indicator is the five-sided haliotis ornament (pl. 2 G). These - 4- occur in ethnographic collections in the U.C, Museum of Anthropology,6 and they are probably assignable to Phase 2 of the Late Horizon. The Mrn-115 specimen was found just beneath the surface on the rim of house- pit 7. Presence of this type, plus the absence of historical material or historical records of the site, suggests that the occupation ended about 1800. It is possible that the occupants of this village were taken to one of the Spanish missions. The group of artifacts from below the 72 inch line includes large projectile points, charmstones, bone pins, an antler wedge, and an awl made from the vestigial outer metatarsal of a deer. All of these could be fitted into the complex of artifacts from the McClure facies of the Middle Horizon, 7 and until further work is done the level is tentatively assigned to that horizon. However, in view of the limited areal defini- tion of Bay phases, it is possible that the level is of different affili- ation, perhaps even corresponding to Phase I of the Late Horizon. Yet it could not under any circumstances be assigned to the Early Central California Horizon, for even this small sample shows numerous sharp differences from the Early Horizon. Neither of the charmstone types occurs in the Early Horizon, nor does the antler wedge. Carbon 14 Date The Mrn-115 excavation was open and appeared to contain Middle Hori- zon material at the time the first C-14 samples were being run, and a sample was accordingly taken from the bottom of the excavation in the hopes of getting a Middle Horizon date which would be useful in cross-dating other Bay sites. The site contained considerable charcoal in the lower levels, but it was distributed more or less evenly in small pieces, a quarter of an inch or so in diameter. It was therefore necessary to collect a number of scattered fragments to get a large enough sample for a C-l4 test. Some charcoal came from a depth greater than 108 inches in trench C; the bulk of the sample was collected from the wall of Trench D, 5 feet further into the site, from the lowermost 18 inches of deposit (depth 114-132 inches). Two determinations were made on the charcoal sample and the dates were first published in 1950. 9 This was Arnold and Libby's sample 186, dated as follows: 633 + 200 Average 720 + 130 years old. 911 + 180 Using the average figure, the level is seen to have been occupied some time between 590 and 850 years ago. If the range for both samples is con- sidered, the level could have been occupied any time between 433 and 1091 years ago. As it now stands, the dating information is rather vague, and probably not too much faith can be placed in any specific date for the level. However, accepting the figure of 720 years for purposes of discussion, the date is of considerable interest. This means that the ten foot level was occupied about 1200 A.D. The evidence indicates that the site was abandoned about 1800, and one can therefore deduce the rate of accumulation of the midden as a foot every 60 years. The sample was taken at the edge of the mound, and as the center of the site must be close to 18 or 20 feet - 5 - deep, the age of the mound would be roughly 1000 years, the bottom level presumably dating from about 500 A.D. This assumes a constant rate of deposition, which of course may not be correct. Publication of the C-14 date has led to some confusion among archaeolo- gists. Heizer has commented on the date as follows: The unfortunate use of the word "Archaic" in Libby's pub- lished description is referable to Heizer who used the term in an attempt to define the general type of culture; "Archaic" is not used in Central California culture terminology. 10 Since this note appeared, the site has onff more been erroneously referred to the Early Central California Horizon. In view of the great dis- crepancy between the dates for Mrn-115 and the Early Horizon (the latter dating over 4000 years 12), it is important to place on record the fact that Mrn-115 does not bear any similarity to sites of the Early Horizon, and it was never believed that the site was so affiliated. Conclusions Preliminary excavations in a Marnn County shellmound have yielded evidence of two culture horizons, the earlier of which was functioning in the thirteenth century. The mound appears to have been occupied until about 1800. Site Mrn-115 is one of the last large shell heaps around San Francisco Bay to remain in its original condition. Because of this and the site's depth and stratification, Mrn-115 holds promise for clarifying many problems of bay area archaeology. It is hoped that a full-scale excavation will be carried out here before the site is destroyed in the inevitable advance of construction work in this region. C.W. Meighan Archaeologist, U.C.A.S. a 6 - NOTES 1. The excavation was under the sponsorship of the U.C. Archaeological Survey, which supplied excavation equipment and part of the travel expense, The project was aided by Franklin Fenenga, then Archaeolo- gist of the Survey, and by many students who volunteered their time and effort to help excavate the site. Students who worked at the site include: Charles Crary, David A. Fredrickson, Robert E. Greengo, James Kellar, Donald W. Lathrap, Dorothy Rainier Libby, Ray Mrotek, Arnold R. Pilling, J.P. Redwine, F.A. Riddell, and R.J. Squier. A special acknowledgment is due to Donald F. McGeein, who called the attention of the U.C.A.S. to the site and assisted throughout the excavation. Mr. McGeein also prepared the illustrations for this report. J. Arthur Freed identified several of the mammal bone speci- mens. '2. See site record book, Marin County, site 115. U.C.A.S. files, Berkeley. 3. Barrett, 1916, pl. 2, fig. 2, illustrates a brush house of this type for the Pono; see also Dixon, 1905, pp. 172, 173; pi. 45, for a similar house of bark used by the Northern Maidu. 4. Meighan, 1950, table 2. 5. Beardsley, 1948, p. 17. 6. Gifford, 1947, p. 23. 7. Beardsley, ms. 8. Heizer, 1949. 9. Arnold and Libby, 1950. 10. Heizer, in Johnson, 1951, p. 25. 11. Cressman, 1951, p. 305. 12. Johnson, 1951, p. 13. -. 7- REFERENCES Arnold, J.R., and W.F. Libby 1950. Radiocarbon Dates. Nuclear Studies. ( University of Chicago, Institute for Chicago, Barrett, Samuel A. 1916. Pomo Buildings, in Holmes Anniversary Volume. Washington. Beardsley, 1948. RoK. Culture Sequences in Central California Archaeology. American Antiquity, vol. 14, no. 1, pp. 1-28. Menasha. n.d. Temporal and Areal Relationships in Central California Archaeology. Ph.D. thesis (1947). University of California, Berkeley. Cressman, L.S. 1951. Western Prehistory in the Light of Carbon 14 Dating. South- western Journal of Anthropology, vol. 7, no. 3, pp. 289-313. Albuquerque. Dixon, RIB. 1905. The Northern Maidu, American Museum of Natural History, Bulletin; vol. 17, part 3. New York. Gifford, E.W. 1947. Californian Shell Artifacts. University of California- Anthropological Records, Vol. 9, no. 1. Berkeley. Heizer, R.F. 1949. The Archaeology of Central California; I: The Early Horizon. University of California-Anthropological Records, vol. 12, no. 1. Berkeley. Johnson, Frederick (editor) 1951. Radiocarbon Dating. Society for American Archaeology, Memoir No. 8. Salt Lake City. Meighan, Clement W. 1950 Observations on the Efficiency of Shovel Archaeology. Reports of the University of California Archaeological Survey, No. 7. Berkeley. - 8 - APPENDIX CARBONIZED BASKETRY FROM HE THOMAS SITE Martin A. Baumhoff Coiled Basketry There are 16 pieces of coiled basketry from the site, undoubtedly from the same basket. The foundation is a single rod and is sewn with a single noninterlocking stitch (Pl. 2C). This is not particularly fine coiling com- pared to the Porno coiled ware, both the coils and the stitches running about 50 per 10 cm. The rods are just over 1 mm. in diameter while the splints are about 1.5 mm. in width. On coming to the end of a splint, the fag end was apparently secured by weaving it over and under the last few stitches. The literature on Coast Miwok material culture is very sketchy so comparisons here must be with the adjoining Pomo to the north. The Mrn-115 coiled basketry is closely similar to that of the Por of the historic period. Mason (1902, p. 250) says, The stitch passes around the stem in progress and is caught under the one in the preceding coil . . . The foundation is the stem of the plant in its natural state; the sewing is with splints of the same material, having the glistening surface outward. As this is somewhat unyielding it is difficult to crowd the stitches together, and so the foundation is visible between. This is about what we have, although it is hard to say whether the materials are the same for the warps and wefts. Twined Basketry Essentially there are two kinds of twined basketry from Mrn-115, the first of which is a fine flexible ware whose pitch is up to the right. The wefts are of a fine grass, usually slightly less than 1 mm. in diameter. The warps are composed of two strands of even finer grass which are twisted together in a clockwise direction (see pl. 2 D). The fineness can be Judged by the fact that there are about 50 warps and 50 wefts per 10 cmo. One piece of this type has been sewn through with a thin (willow?) splint about 3.5 mm. in width. This seems to have been either a mend or a stitch for holding some kind of attachment. The distribution of this basketry is interesting.. It occurs in the caves of southeastern Oregon (Creseman, 1942); the Humboldt Valley of west- central Nevada (Loud and Harrington, 1929, pl. 31D; Heizer and Krieger, mA.; " 9 - Robson and Baumhoff, Ms.); Tommy Tucker Cave of northeastern California (Fenenga and Riddell, 1949, fig. 56G); among the Klamath Indians of north- eastern California (Barrett, 1910, p. 254; Spier, 1930, fig. 14). The pitch of stitch is up to the right in the specimens from M4rn-115, whereas it is usually down to the right for these others. Cressman (1942, p. 42) points out the similarity of technique between this ware (sometimes called Catlow Twine) and the soft Basket Maker bags. In this connection it is interesting to note a very flexible twined textile discovered by L.L. Loud at the Glen Cove site (4-Sol-236) on the north bank of Carquinez Straits (San Francisco Bay). It has a two strand warp with a clockwise twist that averages a bit less than 1 mm. in diameter and a weft about half that diameter, both being made of a very fine grass. The pitch of stitch is down to the right. The warps are alternately single and in pairs so that they go two, one, two, one, etc. This "cloth" is shingled with very small rectangular Olivella beads and is apparently the same weaving technique as described above. The common elements of the specimens mentioned above are twining on a two-strand twisted warp, mostly with a down to the right pitch, and a great deal of flexibility which is mainly because of the cordage-like warp. Although the specimens are not typologically identical, the area of dis- tribution is fairly continuous and it may well be that they are part of the same tradit ion. The remaining twined ware lran Mrn-115 is somewhat heavier, and being woven on a solid warp it is stiffer. There are three varieties of this, the first being a single piece (UCMA 1-127796L) of simple twining with a down to the right pitch. It is the only piece with such a pitch and because it is so small (2.4 by 2.2 cm.) it may be that it is just a piece of decoration from one of the other baskets. This piece has a flat weft 3.5 rnm. wide and a stick warp 2.8 mm. in diameter. It has about 30 wefts and 30 warps per 10 cm. The second type of sti.ff twined basketry has a pitch which is up to the right and has a fairly tight weave with about 40 wefts and 28 warps per 10 cm. The warps are flat sticks about 3 mm. wide and the wefts about 2.7 mn. wide (see pl. 2A). The flattened warps may be due to the weight of house posts falling on them, but this is not likely, since they give a rather shingled ap- pearance as if they had been slightly twisted in the weaving process. There was only one piece (UCMA 1-127796M), but it was larger than the down to the right piece mentioned above so it probably represents the complete basket. The final type was the most numerous, having 17 pieces. This is a twined ware with 40-48 wefts and 28-30 warps per 10 cm. The round warps are 2 mm. in diameter and the wefts are about 2 mm. wide. It is similar to the stiff twined ware described above, but is quite different in aspect (see pl. 2B). So far as this twined ware, with the exception of the one piece (UCMA 1-127796L) has an up to the right pitch, it differs from Pomo twining, in which, according to Barrett (1908, p. 147), "All the tightly woven baskets are made with a downward (down to the right) turn of the woof -strands ..." Except for this difference, the stiff ware from Mrn-115 is quite the same as Pomo basketry in technique of manufacture. Martin A. Baumhoff Assistant Archaeologist, - 10 - 4U.C.A.SA REFERENCES FOR APPENDIX Barrett, S.A. 1908. Porno Indian Basketry. U.C. Publications in American Archaeology and Ethnology, vol. 7, pp. 134-308. Berkeley. 1910. The Material Culture of the Klamath Lake and Modoc Indians of Northeastern California and Southern Oregon. U.C. Publications in American Archaeology and Ethnology, vol. 5, pp. 239-292. Berkeley. Cressman, L.S. 1Q42. Archaeological Researches in the Northern Great Basin. Carnegie Institution of Washington, Publication 538. Washington. Fenenga, F. 1949. and F.A. Riddell Excavation of Tommy Tucker Cave. American Antiquity, vol. 14, pp. 203-214. Menasha. Heizer, R.F. ins. Loud, L . L . ai 1929 Mason, O.T. 1902. and A.D. Krieger The Archaeology of Humboldt Cave, Nevada. the University of California. Berkeley. Manuscript at nd M.R. Harringtc Lovelock Cave. Ethnology, vol. )n U.C. Publications in American Archaeology and 25, pp. 1-183. Berkeley. Aboriginal American Basketry. Annual Report of the Smith- sonian Institution for 1902. Washington. Morris, E.H. and R.F. Burgh 1941. Anasazi Basketry. publication 533. Carnegie Institution of Washington, Washington. Robson, J.C. Ms. Spier, L. 1930. and M.A. Baumhoff Excavation of site 26-Pe-8. California, Berkeley. Manuscript at the University of Klamath Ethnography. U.C. Publications in American Archaeology and Ethnology, vol. 30, pp. 1-338. Berkeley. - 11 - EXPLANATION OF ILUISTRATIONS Map 1 Contour map of Mrn-115, showing location of house-pits. Plate 1 House ruin in Pit 7. Plate 2 A-D E F G H J K L M N p Q R S T U V Basketry fragments from house ruin. Scapula saw (1-127906, Pit D-2, 21") Bone awl (1-127894, Pit D-3, 32") Haliotis pendant (surface, house-pit 7) Cut bird bone tube (Pit 1S, 17") Unfinished projectile point (?) of green chert. Edges chipped to a point on one end (1-127848, surface) Obsidian projectile point (1-127864, Pit A-3, 9") Triangular sandstone slab with bevelled edges (Pit IN, 21") Blue steatite charmstone (1-127880, Pit C-3,. 80"; asphaltum around neck showing string impressions) Coarse brown sandstone charmstone (1-127869, Pit C-1, 96") Obsidian projectile point (1-127861, Pit CO',3, 82") Obsidian projectile point (1-127918, Pit D-1, 90") Obsidian projectile point (1-127860, Pit C-3, 72") Deer metatarsal awl (1-127916, Pit D-1 73") Antler wedge (1-127912, Pit D-1, 112"; fragmentary and burned specimen) Bone hair pin (1-127849, Pit C-3, 80") Bone hair pin (1-127871, Pit C-2, 86") Figure 1 A Cross-section of mound along long axis. B Detail of south edge of site, showing alleviation over margin of mound. C Plan of test holes augured to determine profile in fig. B. Figure 2 Profiles showing internal construction of mound. Figure 3 All specimens are illustrated actual size. A Obsidian saw or blade (1-127872, Pit C-3, 41"; crude chipping) B Flake scraper (1-127951, Pit C-2, 41") C Cut maimmal bone frag. (1-127856, surface; distal end of coyote femur (Canis latrans ]) D Molded clay object, probably a cast of a shell which was burned while full of clay (1-127946, Pit C-3, 60-72") E Obsidian nodule, broken in half but not worked (1-127857, Pit iS, 3") F Worked bone fragment, possibly the tip of a bone hair pin (1-127957, Pit C-1, 49") G Bone awl frag. (1-127952, Pit C-2, 20") H Worked bone frag. (1-127881, Pit D-2, 41"1) J Ulna flaker (1-127851, no loc.) K Obsidian projectile point or drill (1-127852, backdirt) L Blue steatite labret (?) frag. (1-127891, backdirt) - 12 - Figure 4 All specimens are illustrated actual size. A Scapula saw fragment (pit C-2, 54") B Scapula saw fragment (?) (1-127883, Pit D-2, 24-36") C Grooved object of sandstone (Pit A-1, 6") D Pestle fragment, sandstone (Pit C-3, 411") E Hammerstone (1-127950, Pit 1S, 20") Figure 5 All specimens are illustrated actual size. A Bone awl tip (1-127949, Pit C-3, 88") B Bone awl tip (1-127868, Pit C-1, 67") C Obsidian flake scraper (1-127867, Pit C-2, 81") D Cut piece of deer antler (1-127855, Pit 1S, 54") E Sandstone pestle fragment (1-127879, tr. C, 95") F Spatulate bone object (1-127853, Pit C-2, 108-120") G Worked bone fragment, possibly a piece of a sweat- scraper (1-127948, Pit C-1, 67") H Bone awl fragment (1-127913, Pit D-2, 72") J Chert scraper (?) (1-127954, Pit C-3, 90") K Chert scraper (1-127955, Pit C-2, 67") - 13 - TABLE 1. ARTIFACTS NOT DISCUSSED IN EXPLANATION OF PLATES 1 No. 1-127850 1-127852A 1-127853 1-127863 Item Mica fragment Base of obsidian Projectile point Lump of red ocher Obsidian proj. pt. Pit: C3 D L 108 3.6 W Th 2.2 0.1 Surface 1.4* 1.6 o.4 Remarks Fragmentary; may have been part of a pendant but there is no perf. Same type as pl. 2K (with rounded base). Same type as p1. 2K (with rounded base). 1-127865 Chert scraper 1-127874 Base of Chinese porcelain bowl 1-127875 Vesicular lava mortar frag. Surface A 14 A2 10 Surface Surface 2 cm. 2.2* diameter 1.5 0.5 4.3 3.5 1.1 19th century, un- doubtedly intrusive. Globular, rounded rim, original height ca. 15 cm. 1-127876 1-127877 1-127878 1-127890 1-127900 Obsidian blade tip Chert scraper Chinaware fragment Tip of bone pin Worked bone frag. Surface 3.7* Surface 4.2 Surface ca. 2.5 Backdirt 3.8* Test hole 2, 241" 5.8 2.9 2.8 cm. 0.8 d. 1.1 o.6 1.2 1.3 1.3 1.)1 square 0.14 Probably intrusive. 0.5 Medial section of awl? 1-127914 1-127917 1-127953 Bone awl frag. Bone awl frag. Bone awl D1 no. loc. C2 47 13.2 4.2 6.3 0.4 0.7 o.8 Tip missing 1 Depths in inches, artifact measurements in cm. * indicates measure. ment is incomplete due to fragmentary nature of the specimen. - 14 - \\ I Il "I" I- -" 1 4 1 I NI 3 i 9 1 1 \ 12 I I1 I? / I z ./ 1/ *\s. AP 1. . . . = . . I 4u Datum B THOMAS SITE I 4 - tarn - 115 Scale: 1/161 1 ' ( A= house pits Nf Mag. north 'I ~-#~ I. j PLATE I. 4-Mrn-115 House Ruin, Pit T. ____Limit of house-pit(surfoce). Li-mit of excavation Burned logs . l I N. . - 11 - . - . - . - . - . - . - . - . -- 0 -.W=I 0% H / III l1/ iN Ii i-r. , I~ jI i d A A C D 4 W i\-- -- /m I C c~. 1 2 1Cm ar) 7// 1 .7/If . I I Cm. V PLATE 2 a1) Po a1) .9 io q-@5ezz-**ee-Xe~e-ac*@I ... . . . I.. . . .. .. .. ..C . 0 4 *r4 c.) 3 :j r- (N CYN .. . . . .. . . . . . . . ;- G O- - - Q 0 . . . . . .. . . ... .. . a1) a1) Eh 4>-N C) - - - - - - - O r- CN m -1- e- S- E 4- 4- 0 X0 CD% r- I i- a 0 cn I CO H -I at) 1 Q co V I g: ' 0 1 4-) 0 1 a) (a 0 4- r- 0 a) a) C3- I I i i 71I i. I I II i i I I I i i I I I i II ko ad H 2 El? V) t-3 2? !X> r?-4 I I "It F. I !--I C It t F-Hr H CD H3 or) I :r wt'- To z -. I r~- ItE H C) ci o e c: H jH 3-- coDi c0s CD) H ri 0 H- Urq 0 PH r5 (0 1-i 0 HK- I, w P. H' (0 N) -3 t I' I- H HCD H It- ft It' H -I) HIt O B Cf) En r- 6 F-Hl H- Hz t It cO 1 :r. Ml I Fig.. 3 Artif acts from Thomas Site CC Hc FI.. K1L G1Tiil G J |A ,J / v'-- - 1/ \ \ A %-) Fig. 4 Artifacts from Thomas Site N , I : t i, !s -t - K,i, I.P. ''I j?) // -? I K '7 7 . 2' ('N II \\ ? ?1 -- "XI K ?? NK?zL. I.I l\-, B C - ii /1 111\ - -. K ,- It' / - . V 'I' . 1t? 1k '?.,, ,,,, ..- A I, - -'-- Ii 4' '1 'U -? ''" *' '"-' - 'A:__ Yb "- I . . 'I ,/ - , I .. . i - i . I -, -, ,? '-- t . I I N (. ,r C I 1$ - -. i `7 I \ (; . .' i , f . . \ el, ? - I\ -1 A -/ - . 4C-' / I / I I i/ if I / Ij 2 * . :3 &. 67 C -1,- -., I E #/ .'- .'-" - ,... ;f? ?6 - C 4,IJ -4'-- '7/ ,? - / ?--2 ? F' ? 1i ii I. i .e I .i 31f A A I B Fig. 5 from Thomas Site Artifacts I E D l I F i~ I !' I i IIl i I ii I II 1X!., I J K