UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Esilao : a pit house village in the Fraser Canyon, British Columbia Mitchell, Donald Hector 1963

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1963_A8 M4 E8.pdf [ 22.02MB ]
JSON: 831-1.0107144.json
JSON-LD: 831-1.0107144-ld.json
RDF/XML (Pretty): 831-1.0107144-rdf.xml
RDF/JSON: 831-1.0107144-rdf.json
Turtle: 831-1.0107144-turtle.txt
N-Triples: 831-1.0107144-rdf-ntriples.txt
Original Record: 831-1.0107144-source.json
Full Text

Full Text

BSIL40 — A PIT HOUSE VILLAGE IN THE FRASER CAIION,. BRITISH COLUMBIA by POMLD HECTOR MITCHELL Bi., The University of British Columbia, 1955 B. Comm., The University of British Columbia-, 1958 I THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF Master of Arts in the Department of Anthropology and Sociology ¥e accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA April, 1963  ii  ABSTRACT This thesis undertakes a comparative study of two adjacent, yet linguistically distinct Fraser River Canyon groups — the Tait and the Lower Thompson — through an examination of ethnographic and archaeological data. The archaeological examination is based on results provided by an excavation at the recent Tait pit house village of Esilao in the Canyon near Yale, British Columbia. There were two related objectives. The purpose of the archaeological study was to test whether there was a discernible overlapping of ethnographic and archaeological data. Secondly, the Canyon culture was to be examined to determine whether it showed a greater alignment with the coast or with the interior. The results of ethnographic study show considerable uniformity of Canyon culture and pronounced interior affinities. The archaeological investigation reveals much overlap between ethnographic and archaeological data and indicates that the Esilao villa ge assemblage had a definite interior alignment, thus lending support to the ethnographic findings *  In presenting this thesis In partial fulfilment of the requirements for an advanced degree at the University of British Columbia,. I agree that the Library sha.ll make it freely available for reference a.hd study.. I further agree that permission, for extensive copying of this thesis for scholarly purposes may be granted by the Head of my3Department or by his representatives»  It is understood that copying, or publi-  cation of this thesis for financial gain shall, not be allowed'' without my written permission.  Department of  Anthropology and Sociology  The University of British Columbia,. Vancouver 8, Canada. Date  . '  -April. 18, 1963  iii I&JBIE OF CONTENTS  ABSTRACT  Page ii  PREFACE  xii  INTRODUCTION  1  LOCATION AND EOTIRONMENT ECOLOGY OF THE FRASER CAMION  5 ,  7  HISTORY OF THE FRASER CANYON ETHNOGRAPHY OF THE FRASER CANYON INDIANS  10 ;  14  THE CANYON INDIANS  16  Material Culture  19  Structures The Pit House The Plank House The Grave House The Sweat House The Puberty Hut The Storehouse The Mat Lodge  19 19 22 22 22 23 23 23  Carved Grave Figures  23  Transportation Canoes Snowshoes  24 24 24  Clothing  24  Personal Adornment  25  Household Utensils . Birchbark Containers Basketry Containers Wooden Vessels Stone Vessels  26 26 26 27 27  Tools and Weapons Bows Arrows Spears  27 27 27 28  "it  Tools and Weapons, continued Clubs Wood-working Tools Fire-making Apparatus Smoking  Page 28 28 28 28  Subsistence Fishing Dip-netting Bag-netting Nets and Eqpe Harpooning Hook and Line Weirs Preservation  29  .  29 29 30 30 31 31 31 32  Hunting  33  Vegetable Foods  33  Social Structure Social Units The Extended Family The Village The Tribe  33 33 33 34 34  Kinship  34  Social Stratification  35  Rank  Slavery  The Society in Operation  35  35 35  Life Cycle of the Individual Birth and Infancy Puberty Marriage Death  35 35 36 36 37  Political Activities Leadership Concept of Property  37 37 38  Religious and Ceremonial Activities The Guardian-spirit Conplex  38 38  A  Religious and Ceremonial Activities, continued The First Salmon Ceremony The Potlatch Inter-group Relations Trade Warfare Marriage . SUMMARY AND ANALYSIS' OF ETHNOGRAPHIC MATERIAL . ARCHAEOLOGICAL INVESTIGATION AND AlALlSlS Site Description Rock-lined Pits House-pit Number 1 Excavation Procedure  Page 39 39 AO 40 40 41 43 46 46 48 48 49  Structural Data  51  Stratigraphy  51  Structural Features Plan Earth Walls Floor Hearths Notched-log Ladder Posts and Rafters  53 53 54 54 55 55 55  Artifacts Stone Industry  60  Chipped Stone Projectile Points Drills Scraper-Knives Chipped End-Scrapers Choppers Boulder Chip Scraper-Knives  60 60 69 69 70 70.  Ground Stone Ground Slate Points Ground Slate Knives Crude Slate Knife Celts Carving Chisels Pipes  72 72 73 76 76 77 7g  vi  Ground Stone, continued Saws File Abrasive Stones and Whetstones Spindle Whorl  Pa ge 95 97 98 99  Personal Ornament Stone Beads Stone Pendants *  100 100 103  Miscellaneous Objects Hammerstones Hand Maul Stone Vessel Forked .Steatite Ochre Problematical Objects  105 105 106 109 111 113 113  "  Wood Industry Bark Rolls Wooden Haft  115 115 117  Trade Goods  117  Glass Glass Beads  118 118  Copper Tube Beads Ring Bracelet Jingle Pendants Sheets  122 122 124 124 125 125 128  Brass Pendant  129 129  Iron Points  130 130  SUMMRY Al® ANALYSIS OF ARCHAEOLOGICAL INVESTIGATION Upper and Lower Grey Deposit Assemblages Dating of the Occupation layers Inter-area1 Comparisons  131 131 132 133  vii  CONCLUSIONS The Comparative Study Ethnographic and Archaeological Overlapping Coastal and Interior Affinities Problems for Future Investigation BIBLIOGRAPHY:  PLATES  ;  Page 136 136 137 140 . 141 143  147  viii  ;  I'  '  LIST OF TABLES  PRECIPITATION, SNOWFALL AMD' iEMPERATUBE FOR THREE ERASER CANYON STATIONS  Page  II - DISTRIBUTION OF SELECTED GULTUKE TRAITS  III INTERIOR DIMENSIONS OF FNEXCAVATED HOUSE-PITS  44  <  47  IV UPRIGHT. BRACE POST HOLE DIMENSIONS'  56  V DIMENSIONS OF PROJECTILE POINTS FROM UPPER GREY  62  DEPOSIT  VI DIMENSIONS OF PROJECTiliE POINTS - FROM. LOWER GREY: - DEPOSIT \ . VII , RELATIVE MPORTMCE OF POINT TYPES VIII. COMPARISON OF STEMMED AND UNSMMED POINTS FOR SElffiN SITES OR SURVEY ARMS  IX DIMENSIONS OF DRILLS X DIMENSIONS OF CHIPPED SCRiPER-MIVlS: FROM. UPPER GREY DEPOSIT. XI QUANTITIES AND DIMENSIONS OF BOULDER CHIP SCRAPERII®' ' ,  XII. /DIMENSIONS OF GROUND SLATE POINTS XIII GROUND SLATE KNIFE FRAGMENTS  5  ,  63 64 67  68 69 71 72 74  XIV BOWLFMGMENTS-.FROM UPPER GREY DEPOSIT  ' 80  XV STEM FRAGMENTS FROM UPPER GREY DEPOSIT  82:  XVI STEM END FRAGMENTS FROM UPPER GREY DEPOSIT  83  XVII BOWL FRAGMENTS FROM LOWER GREY DEPOSIT  85  XVIII ,STM:MGMENTS FROM LOWER GREY DEPOSIT  87  XIX STEM END FRAGMENTS FROM LOWER GREY DEPOSIT XX CLASSIFICATION OF PIPE FRAGMENTS FROM UPPER AND LOI'IER, GREY. DEPOSITS  87 92  •viae: .  XXI DIMENSIONS OF SANS FROM UPPER GREI DEPOSIT XXII DIMENSIONS OF ABMDERS FROM UPPER GREI DEPOSIT  Page 96 98  DIMENSIONS OF WHETSTONES. FROM LOWER GEEI DEPOSIT  99  XXI? "DIMENSIONS OF STONE BEADS FROM UPPER GREI DEPOSIT  100  XXV DIMENSIONS OF STONE BEADS FROM LOWER GREI DEPOSIT  100  XXVI DIMENSIONS OF HAMMERSTONES FROM UPPER GREI DEPOSIT  107  XXVII DIMENSIONS OF HAMMERSTONES FROM LOWER GREI DEPOSIT  108  mil  XXVIII TECHNIQUES OF WORKING STEATITE XXIX DIMENSION OF BARK ROLLS FROM UPPER GREI DEPOSIT XXX DIMENSIONS OF BARK ROLLS FROM LOWER GREI DEPOSIT  xxxi: DIMENSIONS AND QUANTITIES OF SPHERICAL BLUE GLASS TRADE .HEADS.. .  XXXII. DIMENSIONS OF MISCELLANEOUS GLASS TRADE BEADS.  111 . 116 117 118 120  XXXIII 'DIMENSIONS OF COPPER TUBES FROM UPPER GREI DEPOSIT  123  XXXIV DIMENSIONS OF COPPER TUBES FROM LOWER GREI DEPOSIT  124  XXXV DIMENSIONS OF SHEET COPPER PIECES FROM UPPER GREI DEPOSIT ,  129  XXXVI DISTRIBUTION OF TRAITS REVEALED BI THE ARCHAEOLOGICAL INVESTIGATION, AT ESIIAO XXXVII INTEGRATION OF ETHNOGRAPHICAL Al® ARCHAEOLOGICAL INVESTIGATION .. ..  135 138  LIST OF TEXT FIGURES  FIGURE Is Map of Lower Canyon Area  Following page  16  FIGURE -~2t Plan and Elevation of Pit House (after Teit, 1900:193)  Following page  21  FIGURE 3S Map of Esilao Village  Following page  4.6  FIGURE W. DjRi5s Pit House 1 Contour Map  Following page  47  FIGURE 5 s DjRi5: . Pit -House 1 Profile of S5 - S60 at W10 Looking East  Following page  52  FIGURE 6 s D;jRi5: Pit House 1 Profile of ¥ IS - E17 at S25 Looking North  Following page  52  FIGURE 7s Schematic Representation of Rim Profile  Following page  53  FIGURE 8: DjRi5: Pit House 1 Features  Following page  57  Page  72  FIGURE 10s Slate Knives  Following pa ge  73  FIGURE lis Slate Knife Ends  Following page  7A  Page  76  Following page  92  FIGURE 9: Slate Point No. 5275  FIGURE 12s Celt No, 2608 FIGURE 13 s. Pipe Forms FIGURE 14.: Hand Maul Fragment  Page 106  FIGURE 15s Reconstruction of Stone Vessel  Page 109  FIGURE 16s Copper Pendants  Page 126  xi LIST OF PLATES  PLATE I:  Chipped Points and Drill from Upper Grey Deposit  PLATE II:  Chipped Points from Lower Grey Deposit  Page 147 148  PLATE III: Miscellaneous Chipped and Ground Artifacts  149  PLATE IV:  Pipe Fragments from Lower Grey Deposit  150  PLATE V:  Pipe Fragments from Upper Grey Deposit  151  PLATE VI:  Miscellaneous Objects from Upper Grey Deposit  152  PLATE V I I S a w s and Abraders  153  PLATE ¥111: Hand Maul, Hammerstones and Stone Vessel  154  PLATE K :  Pendants, Beads and Problematical Objects  155  PLATE X:  Metal Objects from Upper Grey Deposit  156  Since 1959? intensive archaeological investigations have been carried out along the southern portion of the Fraser River Canyon, British Columbia. Directed by Dr. Charles E. Borden, Department of Anthropology, University of British Columbia, this continuing research program has been supported through grants to Dr. Borden from the Leon and Thea Koerner Foundation, the Vancouver Foundation (H.R. MacMillsn), the University of British Columbia Committee on Research, and the Mtion&l Museum of Canada. From 1959 to 1961, attention was centred on site DjRi 3, a deeply stratified site which has produced a record of human occupation dating from the recent past back to the eighth millennium B.C. (Borden, I960, 1961, 1962). In 1961, investigations were begun at Esilao (site DjRi 5), a pit house village situated some 150 yards down-river from DjRi 3. Field work at Esilao was continued in the summer of 1962. One of the four house pits in this recent Coast Salish village was extensively excavated. Part of the data collected at Esilao in the course of these two field seasons contributes the basis for the present study. In 1961, the Esilao project was directed by Dr. Borden personally. Donald G. MacLeod, field foreman in 1962, was in charge of excavations during the second season, while Dr. Borden directed the program of work through repeated visits to the site. The writer, who was a member of the field crew in the summer of 1962, is well aware that without the full and meticulous records which were assembled during the two field seasons, the present study would have been impossible.  xiii  Many thanks go to Chief Peter Emery and to the Band Council of the Yale Indian Reserve for their permission to excavate on Yale Indian Reserve No. 21, on which the settlement of Esiiao is located; to Superintendent J.S. Dunn of the New Westminster Indian Agency for his cooperation: and to the members of the 1961 and 1962 field crew. The writer's appreciation is extended to Dr. Frederica de Laguna for her kindness in examining beads from Esiiao; to Mr. James Garner of the Vancouver City Museum for his helpful comments on glass trade beads j and to Miss Charmiantfestphalfor her preparation, in 1961, of the map of the area® Special thanks are owed to my wife, Marjorie, for her encouragement .and very great help throughout all stages in the preparation of this thesis; to my brother, Howard G. Mitchell, for help in the drawing of Figures 6, 7 and.  and to the other members of my family for their many  forms of assistance and encouragement. To my thesis supervisor, Dr. C.E. Borden, I owe a special debt for his patient direction during the organisation and writing of this study.  I. INTRODUCTION In many ways, the enquiry which follows may seem a departure from, conventional archaeological presentation. In part, this is because the study reflects a current interest in the convergence of archaeology and cultural anthropology and, in part, because the site calls for an original approach. The recent examination by Willey and Phillips (1958) of the merging of archaeological and anthropological theory has focussed attention on the relative positions of three branches of cultural anthropology: ethnography, ethnology, and archaeology. The comparative model developed by Uilley and Phillips presents a useful illustration of converging disciplines as the scale of analytical levels is ascended. Willey and Phillips share Taylor's (194-8) view that at the explanatory level of analysis, the archaeologist is in fact a cultural anthropologist. As a statement of archaeological purpose, the Hilley and Phillips model provides, In a broad sense, theoretical orientation for the following study. A more specific theoretical basis is found in de Laguna's The Story of a Tlingit Community: A Problem in the Relationship between Archaeological. Ethnological, and Historical Methods (i960). The following passages are especially suited for incorporation in these introductory remarks. Ideally . . . the archaeology of a people should enable the anthropologist to trace the record of the culture back into stages temporally prior to those which can be explored through ethnological  techniques or historical records. Admittedly the archaeological data, even under conditions of maximum preservation and most skillful excavation, will never give the couplete outline of a culture . . . . Even if he should discover . . . another Pompei, his ability to understand what he had found would depend upon the degree to which he has ethnological insights into the total culture of which the BSterial remains are the concrete expressions. The more remote the archaeological horizon from the related living culture or cultures, the more limited these insights will be . . . . Because exchanges between neighboring peoples tend to make the cultures within an area similar to each other and In a sense derivatives of the past of any one of them, the archaeologist may discover in other archaeological or modern cultures of the region where he works clues that shed light on his own particular finds . . . . The ethnologist faces difficulties comparable but opposite to those of the archaeologist. For the ethnologist who elicits by e-rery patient and skillful method at his command only a verbal account of how 'our people lived in grandfather's time,' fails to grasp clearly just those aspects of the culture which may best be understood in their material embodiments . . . . The archaeologist digging in a site known from historic records to have been occupied a century ago, and the ethnologist who listens to descriptions of 'how our people lived at that place in grandfather's time' are dealing with the same culture, and their different approaches should not simply result in pictures that complement one another by supplying what the other lacks, but should rather overlap perfectly at some points . . . (de laguna, 196011-2). The singular nature of the site considered in this investigation lies in the ready availability of cultural insights. Esiiao village Is located in the Fraser River Canyon near the limits of what is known ethnographically as Tait territory. Although within the boundaries of the Coast Salish, the site is only a few miles down-river from an Interior Salish division - the Lower Thompson. For both of these Canyon groups, the Tait and the Lower Thompson, detailed ethnographic information is available. Esiiao, which probably bad a long history prior to the time of Simon Eraser's journey down the river, was not abandoned until late in the  nineteenth century. It thus shares with other early historic sites in the Pacific Northwest an unusual ad-vantage accruing from the recency of contact. Considering, for instance, that James Teit, collector of the Thompson ethnographic material, obtained texts from an Indian who had seen Simon Eraser, the first white man to penetrate the Canyon, the unusual opportunity for cultural insight is evident. There is promise of many opportunities "where ethnographical and archaeological data will exhibit the "overlapping" sought by de Laguna. ¥e can summon other justifications for analysis of Esilao within an ethnographical framework. The village, situated some 100 yards downriver from Borden's 9000-year-old site (Borden, I960), might well be treated as the contact historic horizon of a very long sequence of Canyon occupations. Although we may view Esilao in this context, examination of the data through comparison with the earlier site would be pressture. The older material is in the process of analysis and is therefore, not yet available for comparison. Just as a temporal archaeological frame of reference is impractical, so also is an or^nisation solely dependent on spatial comparisons. Few contact or historic sites have been excavated systematically In Interior Salish territory and only one comparable Coast Salish horizon has been investigated. Any pertinent archaeological evidence will, of course, be used in analysis G f the Esilao village assemblage, but the greatest dependence is placed on ethnographic information. Preceding paragraphs have been concerned primarily with describing and justifying a particular analytic framework, we may now set forth the  main objective of this enquiry. Our proposal is to examine the two contiguous yet linguistically distinct Canyon groups, the Tait and. the Lower Thompson, and to undertake a comparative study of their cultures. The examination will involve both ethnographic and archaeological data. The archaeological study should reveal whether or not an overlapping of ethnographic and archaeological evidence — such as de Laguna mentions — is discernible. In addition the study may complement information known, from the ethnographies. Finally, the archaeological and ethnographic data may indicate whether the culture of the Esilao villagers had greater coastal or Interior affinities. In terms of practical organisation, this enquiry will assume the following form. In Part II, relevant ethnographic material is presented for the Canyon Indian groups from Wilson Duff's The Upper Stalo Indians of the Fraser River Valley, British Columbia (1952) and from James Teit's The Thompson Indians of British Columbia (1900). An attempt will be made to single out those culture traits which permit reasonable inferences regarding interior or coastal affiliation. Part III is a report of the Esilao village excavation and an analysis of the findings.. This examination is undertaken mainly with reference to ethnographic data set forth in Part II, but whenever possible, other archaeological evidence is considered. In Part IV, several conclusions concerning the archaeological findings are advanced.  LOCATION! AND MVIRONMENT As the accompanying map (Figure 1) illustrates, Esiiao is located within the Fraser River Canyon, close to its southern end. The Fraser Canyon is a transitional zone which links Coast and Dry Forest biotic areas (Munro and Gowan, 1947:12). Although no distinct boundary can be drawn between the biotic areas, the transition is fairly rapid. There is marked decline in precipitation, noticeable thinning of forest cover, and increasing replacement of coast trees by interior, species as the river is ascended. TABLE I  PREC IPITATIONi, SNOWFALL AND TEMPERATURE FOR'THREE FRASER CANXON STATIONS  Station  Annual Precipitation (Inches)  ¥inter Snovrfall (Inches)  Temperature °F.Monthly Average Min. Max.  Altitude (Feet)  Lytton.  18.33  50.2  27  71  574  Hell's Gate  46.33  69.4  27  69  374  Hope  62.12  54.0  31  65  152  NOTES:  (1) Averaging periods are as follows: Lytton - 30 yearsj Hell's Gate - Precipitation and snowfall, 10 years, Temperature, 11 yearsj Hope - 23 years. (2) Source: Province of B.C., 1961: 6 and 18.  Table I illustrates that between Hope and Hell's Gate, a distance of about 30 miles, annual precipitation decreases approximately from 62 to 46  6  inches, while in the next 30 miles (to Lytton) precipitation drops to 18 inches. Up-river from Hell's Gate, the noticeably thinning forest includes fewer lush coastal cedar and hemlock and an increasing percentage of dry belt yellow pine trees. Duff (1952:17) remarks on the significant Increase in snowfall between Agassiz, on the Lower Eraser River flood plain, and Hope, in the Canyon. Within 20 miles, the snowfall increases by two-thirds, rising from 33 inches to 54- (Province of B.C., 1961:18). A further rise is noted as the Canyon is ascended, but in the 30 miles above Hell's Gate, the snowfall lessens, falling from 69 inches at the latter station to 50 inches at Lytton. Throughout the Canyon, January temperatures average below freezing, and for Hell's Gate and Lytton stations, December's average also falls below 32°F. In contrast, the down-river station at Agassiz records no monthly averages below freezing (Province of 3.C.196l:6). In the sixty miles of Canyon between Lytton and Hope, the Fraser River drops some 425 feet in a tumultuous series of rapids and bendso Hutchison (1950:9-11) presents a dramatic description of the Eraser after its confluence with the Thompson at Lytton: Its direction Is still south but it has collided for the first time with the Coast Range, which runs from the southeast to the northwest sideways along the edge of the continent. Seeking a pass to the sea, the current moves in a quickened pace but yet parallel to the coast. It is now squeezed tight within the mountains and turns furious at its imprisonment. Its channel here is cut out of the living rock, its trench dug ever deeper to accommodate its distended body, its water convulsed in whirlpool, back eddy and hidden cavern . . . .  7  To observe the dimensions and power of this larger life you must crawl down the rock slides to the riverbank. There the smooth line of water as seen from the mountains turns into a paroxysm of dirty foam, rising and falling in steady pulse. The perpetual mist has coated the canyon walls with slime and the water has worn them smooth, squared them off like old masonry so that in places they might have been built by human hands. A few islands still stand in the channel, whittled down to narrow splinters and already doomed. The final product of this erosion, the white sand pulverized out of the mountainsides, is laid in glistening bars by every back eddy. . . . All other sounds, the human'voice, the vrhlstle of the locomotive a few yards off, are obliterated by the din of this caldron. The motion of the river seems to set the entire canyon in motion.. Before the spectacle of flux, the beholder turns dizzy and, looking ip, finds the cliffs closing over his head. The river is larger than it appears. At the gut of Hell's Gate, where it finally breaches the central spine of the mountains, it is only 120 feet wide, but its constricted body has bored a channel for itself 85 feet deep at low water and as deep as 175 feet in summer freshet. It moves here sometimes at the rate of 20 feet a second, too fast even for the passage of salmon. For another fifteen miles below Hell's Gate, the river twists through the Canyon until, just above Yale, it leaves the last dangerous series of rapids and flows, swiftly but less confined, southward to Hope. Here the river's course changes abruptly and it heads westward towards the sea. From Hope to salt water, the Fraser flows at a more leisurely pace.  ECOLOGY OF THE FRASER GANYON, In the preceding discussion on geography, the turbulent nature of the Fraser River in its passage through the Coast Mountains was noted. From its junction with the Thompson River to its exit from the mountains, the Fraser, narrowed by confining walls and blocks of mid-river rock, forms a succession of rapids too swift for canoe travel at even the lowest water. Travellers in the Canyon had frequently to disembark and make their way on foot past the roughest portions.  8  Just as it presents obstacles to the movement of men, so also does the Canyon hinder the passage of more ancient river travellers — the salmon®  Five species ascend the river at various times of the year on their  way to spawn in lakes and tributaries. All five species were taken in the Canyon, but the most important were undoubtedly spring and sockeye salmon (Oncorhvnchus tshawytscha and 0. nerka) whose major runs span the four months from June to September. Spawning runs of less important species extend well into the winter months. The salmon make their way easily 15) the lower 105 miles of the Fraser, until, just above Yale, they are slowed by swiftly flowing water. For the next four miles their progress is difficult, and they seek the help of every back eddy, the rest and protection of every swirling pool, in their battle with the river. Above this stretch of rapids are fifteen miles of calmer water disturbed by an occasional constriction of the enclosing banks.  Then, in a  slick pool at the base of steep canyon walls, the salmon rest before plunging into their next ordeal - the passage through Hell's Gate.  This is  just one of many powerful rapids to be fought before the fish reach their destination, for the river winds, narrows and plunges in a seemingly unending series of obstacles. Yet just those parts of the river that proved a barrier to water travel were of the greatest benefit to man. "Where the fish were forced to hug the banks, or crowd into back eddies and pools, they were easily taken. Vie know that man, 9000 years ago, was camped beside the rapids above Yale  during the salmon spawning season, and there is good reason to believe he was attracted there by the fish (Borden 1960:116)® therefore, a long history as a fishery.  The Canyon may have,  By the time Europeans first  settled on the lower Fraser, annual movement of Indians to up-river fishing locations had become a sizeable migration. Duff (1952:25) quotes from an entry in the Fort Langley Journal referring to the return downriver of 505 canoeloads of "Cowitchens." As if the bounty of fish were not enough, the Canyon has yet another advantage. Channeled by the narrow canyon walls, a constant up-river breeze accompanies then®in salmon runs. ¥arm and dry, the breeze provides ideal conditions for drying the huge catch.  Thus, the Fraser Canyon,  although not conducive to easy water travel, grants man an abundance of food and unrivalled opportunities for preserving it. Mountain-goat, deer, and black bear live in the mountains and valleys on each side of the Fraser, and such small animals as raccoon, beaver, groundhog, squirrel, wildcat, and marten are common throughout the Canyon. All of these animals were taken by Canyon groups, although the groundhog was considered less desirable by the Lower Thompson than by the Tait. Deer and mountain goat were sought by hunting expeditions which left for the mountains in early autumn and remained away from several weeks to months at a time. Migrating birds travel the Canyon in spring and fall and there is, in addition, a considerable resident population of grouse and smaller birds, particularly in tributary valleys.  The nature and abundance of  plant cover differs somewhat between the dry,, northern end of the Canyon and the moist southern portion. Regional disparities led to an extensive trade in plant foods. Berrying expeditions visited mountain slopes, usually in late summer, to gather red and blue huckleberries. In the Canyon, salmon-berries, thimble-berries, black-berries, Saskatoonberries, salal-berries, Oregon-grape berries and crab-apples are regionally abundant and were gathered wherever they occur. Bracken and many other plants were important sources of edible roots and bulbs. We can see that In addition to abundant and readily preserved fish, the Canyon provided for its inhabitants a varied diet of plant and animal foods.  HISTORY OF THE ERASER CANYON, The Fraser was one of the last great North American rivers to be explored. Sea voyages to the Northwest Coast during the late eighteenth century had repeatedly missed Its outlet, and although Alexander Mackenzie had travelled its upper waters as early as 1793, it was not until 1808 that the explorations of Simon Fraser revealed the river's course to the rest of the world. In the autumn of 1805, Simon Fraser crossed the Rocky Mountains and entered what is now the northern interior of British Columbia. Fraser had been instructed by the North West Company to establish fur-trading posts in the area to trace to its mouth a river discovered by Mackenzie and mistakenly believed to be the Columbia. Fraser was engaged in the first objective for scsme three years before he -was able to pursue the second.  11  The discovery of a supply route from the Pacific Coast to the interior of the continent was of prime importance to the North Mest Company. The Hudson's Bay Company had the exclusive use of a salt-water route to the heart of the northern fur area. Hope of offsetting their rival's advantage lay behind the far western exploration of the North West Company. Thus, in 1808, Fraser set out to follow the "Columbia" to its mouth. Embarking at Fort George, one of his recently-established posts, in late May, Fraser travelled without serious hindrance to the rapids north of Lillooet. From here, where his party was forced to cache its canoes, much of their journey was accomplished on foot until they reached the area below the Canyon. Fraser's Journal entries record increasing difficulty as the Canyon was slowly descended. Of the area near Hell's Gate, Fraser wrote: "I have been for a long period among the Rocky Mountains, but have never seen any thing equal to this country, for I cannot find words to describe our situation at times. We had to pass where no human being should venture" (Lamb,1960-96). After making his. way on foot through the narrow part of the Canyon between Spuzzum and Yale, Fraser embarked at the latter place in borrowed canoes and continued his trip to the sea, emerging, much to his disappointment, several hundred miles north of the Columbia's mouth. Greatly disheartened, Fraser returned to the up-river trading posts  12  to report the Canyon route unsuitable as a supply trail, is, link between the mouth of the Columbia and the northern interior forts was eventually made in 1312 when the Pacific Pur Company and the North West Company established rival posts at Kamloops, served, respectively, by supply routes from Astoria and the northern interior forts.  The two supply routes  were combined as one with the sale in 1813 of Pacific Fur Company posts to the other organisation. For many years, the Fraser Canyon lay outside the main paths of communication in the Pacific Northwest.  In 1827, the Hudson's Bay  Company (which had earlier merged with the North West Company) founded Fort Langley on the Lower Fraser delta, and the next year the Company's governor, George Simpson, re-examined the Fraser Canyon as a possible supply route. Although his party, travelling in the low water of autumn,, was able to navigate more of the river than Fraser had, Simpson too, concluded that the Canyon was impractical as a water way. However, the Oregon Boundary Treaty of 134.6 and the consequent loss of the inland supply trail, forced the Hudson's Bay Company to find a new route to the interior. Attention turned again to the Fraser Canyon and in 1847, A.C. Anderson established a tortuous brigade path over the mountains between Kamloops and Spuzzum.  The next year, a portage trail was cut between Spuzzum and  the head of navigation where Fort lale was then erected. The Canyon route, however, was judged too dangerous and only two brigades ever made use of it. By the next year, 1849, Fort Hope had been  13  built and the Canyon brigade trail and Fort Yale were abandoned in favor of a trail running eastward from the new post to the Similkameen valley (Creech,1941:257-262). For another nine years, the Canyon again lay outside the mainstream of Pacific Northwest development, until the I858 discovery of gold in Fraser River gravel bars brought the region very much to the attention of the world. With the descent upon the Canyon of several thousand miners, Fort Yale was reopened and for the nest two years it served as a supply ba se for the largest population north of San Francisco. However, by late 1859, the main body of miners had already moved upstream from the Canyon and the need for a supply route to the new workings became apparent. Once again the Canyon was by-passed when the choice fell to a more westerly route through Harrison Lake and a chain of rivers, lakes and portages to Lillooet at the upper end of the Canyon. Frequent transportation breaks made this new route impractical and the discovery of the Cariboo gold fields in i860, and 1861 required the development of an uninterrupted road to the interior. Attention turned to the Canyon and in the winter of 1861-62 the famed Cariboo Road was begun. By 1864, at its completion, the road stretched 385 miles from Fort lale to Barkerville in the Cariboo, permanently establishing the Fraser Canyon as the main transportation route to the interior.  u  II. ETHNOGRAPHY OF THE FRASER CANYON INDIANS The Indian groins of the Fraser Canyon, situated as they are on the boundary between two major culture areas, exhibit considerable mingling of the main areal characteristics . Wissler noted that because mixture of culture traits and accompanying indistinctness of boundary lines was inevitable, it nattered little to which of two adjoining areas the most marginal culture was assigned (Nissler,1923). For our purposes, however, sketching the main characteristics of the culture areas influencing the Fraser Canyon tribes will provide a basis for more detailed discussion of the two main groups inhabiting the Canyon. Northwest Coast Culture Areas Drucker (1955) distinguishes several characteristics of the Northwest Coast which set it apart from other aboriginal American cultures. These he summarizes ass . . . emphasis on woodworking? rectangular plank houses; specialized varieties of dugout canoes and emphasis on water transportation; untailored (wrap-around or slip-over) garments principally of plant, fiberj barefootedness; an economy built around fishing, with an elaborate series of types of fish traps, angling devices, and harpoons sea-mammal hunting, important both as food source and for prestige; relatively slight use of vegetal foods; lineage-local groip basic sociological unit; rank-wealth correlation defining status, and emphasis on individual status in social affairs; slavery; elaboration of ceremonialism (potlatch, dancing societies, wealth displays);and First Salmon and related types of ceremonies deriving from belief in immortality of game (Drucker,1955s186). Drucker noted the distinctively interior nature of much of Coast Salish culture, specifying, for his Salish-Chinook province, the following traits, most of which he considers of obvious interior origins  15  Mat-lodge temporary dwellings Coiled basketry® . . Woman's basketry cap (truncated conical form) Dog-wool blankets and.the double bar loom (Gulf of Georgia and Straits of Juan de Fuca only) Closely twined wool blankets ('nobility blankets'). . . Loosely-defined system of social rank Spirit-canoe ceremony Guardian spirit singing. . . Small steam sweat lodge (Drucker,1955:19l). Nevertheless, even when these traits are taken into account, there Is still a distinctive coastal stamp to Salish culture west of the Cascade Mountains. Plateau Culture Area;  Say (1939) examined several of what he considered  fundamental complexes of the Plateau Culture Area. He found, as had Drucker, on the Coast, considerable cultural diversity but his conclusion was that the Plateau possessed a distinctive character.  In the list below,  emphasis Is placed on those complexes which distinguish Plateau from. Northwest Coast culture. From Ray's publications (I939jl942), certain Plateau characteristics may be extracted: Birchbark and basketry vessels Semi-subterranean (earth lodge) winter dwellings Summer mat-lodges • ' . ' • ' • Birchbark canoes and use of snowshoe for winter travel (on northern part of Plateau) Buckskin clothing with tailored shirts and leggings Deerskin moccasins Leadership only loosely hereditary with achievement or ability emphasized Equality of men Wealth not coupled with rank or status Slavery of minimi significance Strong emphasis on pacifism Isolation of women during menstruation  16  Inhumation In talus slope or soil Guardian spirit quest Winter guardian spirit dance  THE GMIOEi INDI&HS Although in the following pages the Tait and Lower Thompson are presented as bearers of Northwest Coast and Plateau culture respectively, these two contiguous groups are not In one sense true representatives of their culture areas. They are members of a single linguistic stock, Salishan, and have a long history of association. An increasing body of evidence has led many writers to suggest past affiliations between the Northern Plateau tribes and those of the Strait of Georgia-Puget Sound region (Borden,1951|1954; Drucker, 1955:191; Duff,1956; Jenness,1960:228 Sanger,1962; Smith,1903:190ff; 1907:439; Strong, Schenck, and Steward,1930:145). Details of basketry, house construction and simplicity of ceremony are just some of the ethnographic resemblances, while the archaeological parallels include an early, shared tradition of chipped implements and an emphasis in both areas on massive stone carving. Consequently, quite apart from a mingling of relatively recent Northwest Coast and Plateau elements, the Canyon Indians may share many traits and complexes derived from a common ancient cultural heritage. In the ethnographic investigation which forms the substance of this chapter, items of obviously recent coastal or interior origin will be stressed for they are of greatest importance in assessing cultural Influences on Qanyon inhabitants. Emphasis will also be placed upon, those  FIGURE 1: M o p  of Lowar  Canyon  Arna  17  items for which there is possibility of overlap between archaeological and ethnographical materials. The following description of the Canyon Indians relies on Duff (The Upper Stalo Indians of the Fraser Valley, British Columbia) and Teit (The Thompson Indians of British Columbia) as the principal authorities for the Tait and Lower Thompson, respectively. Unless otherwise noted, ethnographic information is based on these sources and, unless the specific group is mentioned, information applies to Canyon inhabitants in general. The Tait; "Tait" means "up-river" or "cp-river people." The name was applied to a loosely-organised group whose villages were the easterncanost outposts of the Coast Salish. The Tait spoke an tp-river dialect of Halkomelem, a language shared by the Indians on the Lower Fraser River from the Tait to the Musqueam at the river's mouth, and by the Cowichan aid Nanaimo on Vancouver Island. Collectively, the Indians on the Fraser River were known as the Stalo, a name derived from the Halkomelem word for river. The Upper Stalo, adivision including the Tait, were distinguished from down-river grows in many respects. They spoke, for example, a different dialect and many lived in a distinctive type of dwelling. The Tait inhabited a large area on the Fraser River extending from a boundary shared with the Lower Thompson near Five Mile Creek above Yale, down-river some thirty-five miles to the territory of the Pilalt. Inland boundaries were quite indistinct. Hunting parties generally went west only as far as the watershed divide between Harrison lake and Fraser River waters,  18  and east to the lover slopes of the Cascade Mountains. The most important boundary was the up-river delineation between Tait and Lower Thompson. "Up to Five Mile Creek, at the mouth of which was the uppermost Tait village of (Tsawkwem), the fishing rocks were owned by the Stalo, above this by the Thompson!' (Duff,1952:19). This boundary corresponds closely to that observed by Fraser in 1808 (Lamb,I960:97). . Proceeding down-river from Tsawkwem, other village or camp sites along the rapids included Eayam, Esilao, and Kellelliktel (See Map 1). Esilao was recorded by Boas as Asilao and was cited as the village inhabited by the Qelatl, a division of the Tait (Boas,1894:454). The Lower Thompson: Just as the Tait are distinguished as the "up-river people," so are the Utamqtamux or "people below" separated from the other Thompson Indians. Hill-Tout (1900:501) explains that this name means, merely, "below-river" people or "down-river" people, and is "applied by these very people themselves to the Yale tribe below them . . . ." Following Teit we:shall refer to the Interior Salish-speaking people whose territory is outlined below as the Lower Thomson. The Lower Thompson Indian territory stretched from just below the village of Siska down-river to a few miles below Spuzzum. In the west, their hunting-grounds extended to Harrison Lake and the mountains east of the Lillooet River. They hunted southward to the headwaters of Nooksac and Skagit Rivers, and eastward to the headwaters of Tulflmeen and Coldwater Rivers. From the accompanying rap, this territory can be seen to surround, in part, the Tait.  19  As with the Tait, the most significant boundary was that in the Canyon below Spuzzum. A tradition recorded by Teit (1917:50) describes Spuzzum as the down-river door of the Thompson Indian "house" whose center was at Lytton. Spuzzum is acknowledged the southernmost Thompson village on the Fraser River (lamb,1960:97; Hill-Tout,1900:500; Teit,1900:169). Other settlements were spotted up-river from this point but none are recorded for the large hunting territory extending south of Spuzzum in the mountains.  Material Culture Structures The Pit House: The usual Lower Thompson winter dwelling was the semisubterranean earth lodge. It was known throughout Upper Stalo territory but was most common among the Tait. Depending w o n the severity of winter, these lodges were inhabited from December until February or March. Teit provides details of their construction (Figure 2), and as a large part of the archaeological Investigation at Esiiao was concerned with an examination of pit house structural features, we shall quote at length from his description: These^winter houses were generally built in the valleys of the principal rivers, within easy distance of water, and vere inhabited by groups of families related to each other, who, although scattered during the hunting and fishing seasons, dwelt together auring the winter. These dwellings rarely numbered more than three or lour at one place, and often there was but a single house. The size conformed to the number of people (from fifteen to thirty) to be accommodated.  20  A spot with loose soil was selected for the site of the underground house. The person who desired to build the house asked all hi s neighbours to assist. Frequently twenty or thirty people came so that the building was sometimes completed in a single day. They were given food by the owner of the house, whose relatives contributed from their store of provisions. The site was laid out in the following ways A bark rope was knotted at a distance of from, twenty to forty feet from one end, according to the proposed diameter of the house. A second rope was marked off the same length as the first. Then the two ropes were crossed on the ground at right angles, the middle being determined by eye. Sometimes the centre was determined by folding two ropes over and tying them together in the middle. Then they were laid down so as to be at right angles. The centre and each end were marked with a snail stake. With the four stakes on the circumference as a guide, a man marked a circle on the ground with a stick. Then the women began to dig the soil with their digging sticks . . . . They also used wooden scrapers with sharp, flat blades. The loose earth was put into large baskets with the hands ana by means of snail baskets. The contents of the large baskets were then dumped near the hole, to be used later on for covering the roof. Green timber was generally used for the heavy posts of the house. . . . This was measured with bark ropes, the length being determined by eye, in accordance with the diameter of the hole. Then trees were cut, barked, and. hauled to the building-site with stout bark rope. Generally these timbers were not squared. They were worked with wedges, hammers, and stone adzes. The thin poles used for the roof of the house were also barked, except when dry wood was employed for this purpose. They were cut, tied into bundles, and carried to the building-site with ordinary packing-lines by men or women. After the wood was obtained and cut, "toe upright braces (a) were erected. These were placed about fifteen inches deep in the ground, which was firmly pressed down by standing it with the feet and beating it with sticks. The tops of the braces were notched to support the rafters (b). The butt-ends of these were placed about two feet deep in the ground, one at each of the four points marked when the circle was laid out. The braces and rafters were securely connected with willow withes. The rafters did not meet in the centre. The side-rafters (c) rested on the ground and on the outside of the main rafters, at the place where these were supported by the uprights. The rafters were either notched for the reception of the braces, or they were simply tied on, while their butt-ends were embedded in the ground. Horizontal poles (d ) from one to two feet apart were tied to these rafters and side-rafters. They formed the support for the roof-covering. Above the place where the side-rafters"and main rafters join, the poles were placed much nearer together, often so that on the ends of the poles of two opposite sides rested the next pair of the  21  other two sides. The ends of the rafters were connected by four heavy timbers (e), which formed the entrance. This structure was covered with poles or pieces of split wood (f)> which ran from the ground to the entrance, . . their ends resting on the rafters and side-rafters. They were not tied to the framework. They were covered with pine-needles or dry grass j aisi then the entire structure was covered with earth, which was beaten and stamped down firmly. The Lower Thompson Indians, owing to the heavy rainfall prevailing in their country, lined these houses with large pieces of cedar-bark, the inner side out. A large.notched log (g), with its butt-end resting on the ground near the centre of the apartment, and the other end in the square hole or entrance, gave access to the house. This log, or ladder, was placed almost upright. It leaned against the west side of the entrance-hole, to which it was firmly lashed. The fire was at its foot, and separated from it by a slab of stone, which protected it from the heat. A groove was cut along the back of the log, from near the bottom to the top, to serve as a hold for the hand. The small end of the ladder, above the hole, was often rudely carved in the form of the head of a bird, animal, etc., or was painted in red or other color. m patterns. Sometimes these ornamentations represented the guardian spirit of the builder or principal man of the house, but usually they were for adornment only. The head man of the house sometimes painted new designs,- according to his dreams. Hie ladder was generally placed with its small end slightly leaning toward the east. Persons coming in or going out descended or ascended with their face toward the northeast, and the right hand in the groove. Some Indians claim that all the southern interior tribes made these ladders lean slightly toward the east, and that they all, with the exception of the northern Shuswap, ascended and descended in the manner above described. The northern Shuswap invariably took hold of the groove with the left hand, turning their face toward the southeast, and back to the fire, which was always built on the north side of the ladder. When entering the house, they gave warning by shouting 'A'la!' This was done that the women who were cooking might have time to protect the food from dust or dirt. The spaces between the four main beams were called rooms or houses, and took their names from the points of the compass, the mainraftersbeing placed N.E.-S.W. and N.W.-S.E. (Teit,1900:192-94-) . The range in diameter is listed by Bay (1942:177) as 12 to 22 feet. Considerably larger houses were noted by Hill-Tout (1900:512) for the  To follow page 21  FIGURE 2: Plan and Elevation of Pit House (after Teit, 1900:193)  22  Thompson at and below Lytton. The smallest he observed was 34 feet and the largest 39 feet, while his informants related that houses up to 70 feet in diameter were not uncommon. The Flank House: Fraser describes the first plank house he saw on his way down-river as "an excellent house 46 x 23 feet, and constructed like American frame houses. The planks are 3 or 4 inches thick, each passing the adjoining one a couple of inches. The posts, which are very strong, and rudely carved, received the beam across. The walls are 11 feet high, and covered with a slanting roof" (Lamb,I960:99)• This house was within Tait territory and probably very close to Esilao. It is easily identified with the small shed-roofed houses used by the Tait. One such dwelling was reportedly built at Spuzzum by people related to down-river Indians. The Grave House: Special elevated grave structures housed the Tait dead. Some of these grave houses, including one at Esilao, were apparently large enough to hold bodies from several families. Large cedar boxes holding the dead of single families were also used. Fraser described similar "tombs" at Spuzzum and later at Yale. The Sweat House: Both Tait and Lower Thompson built small dome-shaped sweat houses. & frame of arched vine-maple saplings was covered with fir or balsam boughs and the inside occasionally was lined with maple leaves© Larger sweat houses accommodating several people were roofed with cedarbark and sometimes covered with earth.  23  The Puberty Hut; Small brush huts similar to sweat houses were used throughout the Tait and Lower Thompson areas. According to Duff, the Tait differed in this respect from the down-river tribes where the practice of secluding a girl in a curtained room of the larger dwelling house prevailed. But as on the coast, Tait women were not segregated during later menstruations. In contrast, Thompson women were subsequently isolated. The Storehouse; An above-ground plank box cache five to eight feet square was secured to the branches of large trees or was raised on poles. Both Teit and Iiill-Tout (1900:513) mention bark-lined subterranean caches used by the Thompson* Hill-Tout says of the elevated sheds; "As a rule these structures are found only where the ground is rocky, or of such a nature as makes excavations difficult or impossible, as along the Fraser Canyon above Yale." The Mat Lodge: When the Lower Thompson were not in their winter pit houses, they dwelt in above-ground cedar-bark, mat-covered lodges. These had a rectangular floor plan and were shaped like an A-tent with end entrance. No information is available on the mat lodges which were presumably used by the Tait for temporary summer dwellings. Carved Grave Figures Life-sized or larger anthropomorphic grave figures were set upright near the burial box or grave house throughout the Canyon. Duff's informants had seen such figures at Esilao, Yale and Ruby Creek. It seems evident from Teit 's informants that the Upper Thompson only recently acquired  21,  from the Lower division the custom of using grave figures. Transportation. Canoes: The dugout appears to have been the prevailing form of canoe in the Canyon. Three distinct types used were the Nootka, a large canoe up to forty feet long, with high projecting bow and vertical stern-post; a smaller type, fifteen to thirty feet long with a characteristic mouthshaped notch in the low projecting bow; and the dley or shovel-nosed canoe. The first of these was used along the main river below Yale, the second throughout the Canyon while use of the third, a typically Coast Salish craft, extended beyond the Canyon to parts of the Northern Plateau. Snowshoes: The snowshoe in use in the Canyon was of a form considered best adapted for travel on steep mountains and moist snow. Meshes were wide and the frames small and roughly oval. By contrast, Upper Thompson bands used elongated snowshoes (up to six feet long) more suited to open country. Clothing Some buckskin clothing was occasionally worn by the Canyon Indians but its use was not common. Capes, robes, and aprons of woven cedarbark, woven mountain goat and dog wool, or sewn hides were the usual apparel. Both Tait and Lower Thompson went barefoot in summer or rainy weather and in wintertime wore buckskin moccasins.  25  Items-of clothing were frequently decorated with fringe, and the fringe in turn might have glass or shell beads or copper tubes strung on strips of skin. The copper tubes used by the Lower Thompson were about six inches long and one-half inch in diameter. Personal Adornment Ear ornaments were worn by both men and women. These ornaments consisted of bark or skin threads passed through holes in the ear from, which were hung dentalium shells, colored beads and brass, or sheet copper. Necklaces of shells, claws, seeds , bone disks, and. teeth were replaced in more recent times by large and small colored beads obtained from the Hudson Bay Company or from the Okansgan Indians. Occasionally, pendants of sheet copper up to three inches square were attached to necklaces. Erom Eraser's Journal (Lamb,1960s99-100) we learns  "Their  (Indians at Xale) ornaments are the same as the Hacamaugh [Thompson] nation make use ofj that is to say, shells of different kinds, shell beads, brass made into pipes hanging from the neck, or across the shoulders, bracelets of large brass id.re, and some bracelets of horn." Face and body painting was common and -varied in extent from a single dot on each cheek to covering the entire face and much of the body. Bed ochre, the best of which was apparently obtained from the Okanagan Indians, was most often used. Tattooing was confined mostly to women. Usually a few straight lines radiating from the mouth were made, but occasionally the back of the wrist  26  •was also decorated. Lower Thompson men who had married women from the coast sometimes permitted deformation of their children's heads but the practice, even among the Coast Salish Tait was not common. Household Utensils Birchbark Containers? Cradles and baskets, as well as other containers were occasionally made of birchbark. With neither Canyon group does this type of vessel seem to have been so important as coiled basket containers. Basketry Containers; Round basket kettles were used by the Lower Thompson for stone-boiling of food, but conflicting testimony surrounds Tait use of baskets for cooking vessels. One informant who cited their use for cooking purposes also mentioned that boiling-stones were handled with two shovelended sticks rather than with split-stick tongs. Duff observes that both stone-boiling in baskets and use of two carrying sticks are interior traits definitely absent from the down-river groups but shared by Tait and Lower Thompson. Coiled baskets of two main types were produced. One type, with a cedar slat foundation, was made also by the Lillooetj the other, with a round cedar root-bundle as foundation, was the typical Canyon form. Designs were imbricated in black, red, and white. large open-work baskets of cedar twigs were used for carrying fish. These were not known up-river above Spuzzum. Mats? The Lower Thompson Indians wove mats of cedar-bark strips. It  27  is assumed that the Tait used these also® ¥ooden Vessels;  Dugout wooden vessels (some over four feet In length)  have been recorded for the Upper Stalo, and shallow wooden trays or dishes of varying size are noted by Teit. Wooden boxes were used in the Canyon, but the typical Northwest Coast skarffed and bent corner was not, according to Teit informants, employed In their construction. However, Teit (1900:202), writes that "square boxes and buckets bent of wood, in the same style that prevails on the coast, are in use." Stone Vessels: Stone containers are known ethnographically only in the Thompson portion of the Canyon. Here they were used for paint or ochre, for catching drippings of oil, and for grinding tobacco and berries. Tools and Weapons Bows: Bows were made of yew or vine maple and the best were reportedly those with deer-sinew backing. Bows were sometimes covered with snakeskin. Arrowss A shaft of rosewood or serviceberry was often singly charred and sharpened, but more often a point of chipped stone or, more recently, iron, was affixed. The Lower Thompson obtained glassy basalt for their arrowheads from the headwaters of the Skagit River, a region they visited during hunting expeditions* The Tait tipped some arrows with ground slate points.  28  Spears; Spears ranged from three to six feet in length and were tipped with spear-heads of similar shape and material to the arrow-heads. Clubs; War-clubs of wood or stone and sturgeon-clubs of wood were employed. Stone clubs were about two feet long and flattened into a blade beyond the handle. The Lower Thompson used, as well, composite clubs of stone, hide, and wood. These consisted of a wooden handle surmounted by a hidewrapped stone. Wood-working Tools; Duff illustrates a D-type adze from Yale and states that an almost identical one from Yale is in the Provincial Museum. This adze type was in common use among the Lower Thompson, also. In contrast, the typical up-river form was the elbow-adze. The D-type adze was apparently employed for fine work on canoes, paddles, and house-planks. Wapiti antler wedges and hand mauls are recorded for the Lower Thompson, but comparable Information is lacking for the Tait. For the Lower Thompson, there are also recorded stone chisels fastened into handles with sockets, chipped stone knives, beaver-tooth knives, and stone-point drills. Fire-making Apparatus; Fires were started with drills and fire was carried from place to place by means of a slow-match made of cedar-bark or glowing coals in large clam-shells. The latter appears to have been the Tait practice, the former, Lower Thonpson. Smoking; Tait informants knew only of elbow pipes. They were not aware of the straight or tubular pipes found in Canyon archaeological sites. The eldest Thompson informants could remember use of tubular pipes but they  29  stated that such pipes were not as common as the ordinary elbow type® Both Tait and Lower Thompson elbow pipe bowls were made of local steatite, and the stems, of at least Lower Thompson pipes,., consisted of sixto eighteen-inch maplewood tubes® The tobacco used was "a genuine wild tobacco (Nicotians, attenuata) which grew in the warmest valleys. The leaves were gathered,, dried, and greased, and when used were broken up and mixed with bearberry-leaves, which had first been dried or roasted over a fire" (Teit,1900s300). The Lower Thompson carried tobacco and pipes, in a special coiled basket, resembling, in shape, the modern angler's creel, k hole in the centre of the lid allowed the pipe stem to protrude.  Subsistence Fishing Fish taken in the Canyon included all five species of salmon (spring, sockeye, humpback, chum, and coho), sturgeon, steelhead, cutthroat trout and Dolly Varden char. Trout were usually eaten fresh. Most other fish were taken only for immediate consumption in the wintertime, but the bulk of the catch during other seasons was preserved. The most important species were spring aid sockeye salmon, their major runs occurring, as stated earlier, during warm and windy preserving weather. Dip-nettings Spring and sockeye salmon were taken in dip-nets, operated  directly from rocky points or from platforms built out over the water. Fishing stations, nearly all of which were located in the Canyon above Yale, were named and family owned. The dip-net used by both Tait and Lower Thompson was formed from a large elliptical hoop of vine maple bound to the end of a long handle. "A conical net of Indian hemp was attached to bone rings which slid around the hoop, and was held open by means of a line held in the fisherman's hand. 'When the fisherman felt a fish swim into the net, he released the line, the rings slid down around the hoop, and the net formed a bag around the fish" (Duff,1952:63). Bag-nettings  On his return trip, Fraser saw the Indians fishing down-  river from Yales  "Their nets, that resembled purses, were fixed to the ends  of long poles, and dragged between two canoes" (Lamb,1960s114)« A second type of bag-net, common, apparently, only in the Tait area, was also used between two canoes but attached by ropes rather than poles. The top edge of the mouth was buoyed up by floats and the bottom held down by stone weights. Nets and Ropes Most nets were made of Indian hemp fibre (Apocynum canrabium) which was gathered by the Lower Thompson and the Tait or, more commonly in the latter instance, obtained by trade with the Thompson. In the process of manufacture, cartridge-shaped shuttles and net gauges, both of maple, were utilized. Although the best ropes were those made of Indian hemp, twisted hazel or cedar withes were frequently used for securing drying racks or house  31  planks or mooring canoes® Harpooning; As a means of taking salmon, harpooning -was important only in the early spring when the river was low and the water clear. The harpoon used at this time had two fixed foreshafts of unequal length In the Tait area but of equal length in the Lower Thompson territory. Tait salmon harpoon points were of the composite toggling type. Valves of deer or wapiti antler or mountain goat horn held a round cutting point of antler, bone, or charred wood. More recently, salmon harpoons have been tipped by iron nails. Teit illustrates, without description, a Thompson harpoon point which appears to be of a composite toggling type (Teit,1900;251,fig.231). Sturgeon harpoons were also of the composite toggling type, but were armed with ground slate cutting blades about one inch wide and up to three inches long. Sturgeon were sought out by feel, as the canoe drifted downstream over them. When the harpooner felt the cutting blade scrape over the horny plates of the fish's head, he raised the harpoon shaft and drove in the points. Hook and Line; Since spawning salmon do not feed after they have entered fresh water, they cannot be taken by hook and line. Sturgeon, however, were caught by this method throughout the summer. The hook was made by binding one or two heavy bone barbs to the end of a wooden shank. Weirs; Weirs were used by the Tait for salmon and sturgeon below the Canyon area. Fish were either taken in traps, speared with a three-pronged  32  leister or noosed* Neither weirs nor traps were commonly used by the Lower Thompson. The up-river people used weirs to stop ascending salmon so they might be speared easily® Preservations Upper Stalo women prepared the fish for drying by cutting down from the back, and laying out flat each flank. The two sides were joined by the belly skin, and the backbone remained attached at the tail. The women then scored the flank pieces every half-inch, cutting the flesh but not the skin underneath. The backbone portion was similarly scored and the fish were then hung up to dry. The Lower Thompson procedure was identical except that the first slit was made from the belly. Teit observed in 1900 that the wooden-backed steel knife used by older women closely resembled the old ground-slate knives. Drying racks, where the prepared fish were then hung, were usually built on the highest available rocky points. Although open to the warm winds on all sides, the rack had a rain and. sunlight-proof covering. Fish were hung in these racks for about three weeks. Dried salmon were then placed in an elevated storehouse for winter use. Salmon oil was often extracted, particularly from the sockeye, a species considered less desirable for drying than the spring. Salmon heads were dried for later use in soup, as trap bait, or as torches. The roe was sometimes buried and allowed to reach a near-rotten state. When removed, it had the consistency of cheese and could then be eaten raw or boiled. Sturgeon were usually smoked, while eulachon, taken only by the Tait, below the Canyon, were apparently preserved by a combination of drying and  33  smoking. Eulachon oil -was not extracted. Hunting Although not so important an economic activity as fishing, hunting was well-develcped. Black bear, mountain goat, deer, elk, ground-hog, marmot, and beaver were taken, as were many species of birds. Tait and Lower Thompson used hunting dogs, bows and arrows, spears, clubs, and many traps, including pitfalls, deadfalls, snares and nets. The use of nets is recorded in Fraser's Journal in an entry made shortly after the explorer penetrated the Upper Stalo portion of the Canyons "I examined a net of a different construction from any I had hitherto seen. It was made of thread of the size of cod lines, the meshes were 16 inches wide, and the net 8 fathoms long. With this the natives catch Deer, and other large animals" (Lamb,1960:98). Vegetable Foods A wide variety of roots, berries and seeds was consumed. Fruits were commonly mashed in a basket or wooden bowl, spread in the sun to dry, and stored away as flat cakes. Lower Thompson Indians looked on the upper division as a source of certain roots made available through trade.  Social Structure Social Units The Extended Family: The most important economic and social unit and the largest exogamous group was the extended family. Formed about a nucleus of  34-  males (brothers, sons, grandsons, etc.) the extended family included wives, children, and other dependents. The Village; One or more extended families might occupy the small Canyon villages. Populations seem to have been quite fluid, and Duff observes that In marked contrast to down-river groups, Upper Stalo villages or locallineage groups lacked any tradition of descent from mythical ancestors. Lower Thompson villages, on the other hand, were more stationary than those of up-river groups, and many families wintered for generations, at the same spot. At the same time, the village community was not a permanent social unit. The Tribe; The tribal concept was not developed in Thompson or Upper Stalo groups. The Tait, with several unnamed clusters of villages, were so loosely bound that Duff is reluctant to call them a tribe, and the Lower Thompson had no named group larger than the local village. There were undoubtedly kinship ties knitting the Canyon villages but there seems to have been no recognized political or economic unit functioning above the village level. Kinship Part of the Tait child's training was thorough lecturing on his position in an extensive web of relationship. One informant spoke disparagingly of the neighboring Thompsons because of their lack of regard for kinship bonds. During the late fall, Stalo Indians would visit their relatives. Tait  35  people would travel down-river to visit, pick cranberries, and occasionally to stay all winter. Social Stratification HasM  Personal qualities were held to be of paramount importance in de-  termining an individual's rank. However, both Tait and to a lesser extent, Upper Thompson, regarded high birth and wealth as important ways of attaining respected personal characteristics. A wealthy man, particularly one who gave liberally of his riches, was highly thought of. High birth was expected to result in a superior personality, and to the extent that people of high rank were groomed for their status, were made distinctive through such physical marks as headflattening and nose or ear pendants, aid tended to intermarry, the Upper Stalo, at least, lived in an hereditary class society. Slavery: From the  SOCK what  contradictory statements of his informants,  Duff (1952:84) concludes: Stalo slavery did not differ in kind from the comolex as it existed universally on the North-west Coast. Slaves were held, obtained by capture or purchase, not permitted to marry freemen, and passed on their status to their issue. . . . On the Plateau, Ray found slavery quite widespread . . . but developed to the above degree only among the Thompson . . . . Among both Stalo and Thompson, then, the complex was developed to an approximately equal degree, but among both it was atypical of the basic orientations of the culture.  The Society in Operation Life Cycle of the Individual Birth and Infancy: The newborn Tait or Lower Thompson infant was tightly  36  swaddled and bound into a basketry cradle, at the bottom of which was a hollow tube for carrying off urine. He was nursed for a year or two and finally left the cradle only when he could walk away from it. Puberty? Tait and Lower Thompson girls marked special puberty obserTOnces. At the onset of the first menstrual period, the girl was secluded in a bough-branch hut. She was encouraged to great activity, fasted and bathed frequently in cold water. For high-ranking Tait families, a daughter's puberty called for an elaborate potlatch ceremony. Lower Thompson but not Tait boys took part in special ceremonies on attaining puberty. The nature of these ceremonies depended upon the boy's aspirations. Those who wished to become great hunters practiced hunting and shooting in a ceremonial way. Fasting and long trips into the mountains were part of a guardian spirit quest. Marriages Shortly after their puberty observances, girls would usually marry. Boys were generally a little older. For the Tait, there was little freedom of choice among high-ranking families, where a spouse of equal rank was required. Among the Lower Thompson, the wealthier, more successful or more industrious, and so more distinguished families tended to intermarry. Although a married couple might move around fairly frequently, residence was usually patrilocal. Among Tait families of high rank, particularly among those which were polygynous, residence was more rigidly patrilocal. One of Duff's informants suggested that some polygynous marriages were attempts to establish good relations with neighboring Indian  37  groups. Throughout the Canyon area, the levirate and sororate existed. Deaths A shaman supervised preparation and wrapping of the body for burial in the family grave-box or grave house. The Tait funeral ceremony was marked by two feasts, spaced four days apart. At the second feast, food and valuables were burnt. A common practice with both Tait and Lower Thompson was to take up the bones of a deceased relative and re-cover them with new wrapping material. A Lower Thompson tradition states that long ago the dead were buried, but for many generations prior to Christian influence, they followed the down-river custom of placing the bodies in elevated grave-boxes. Political Activities Leaderships The office of chief did not exist, positions of leadership in important affairs falling to men who were highly respected. Within each extended family was one man who made the everyday decisions on matters involving the family. "In multi-family villages, these heads were no doubt loosely ranked by prestige, with one man standing above the others and holding the most sway over the village as a whole" (Duff,1952:81). The power of this individual over unrelated families must have depended greatly on personal qualities. Lower Thompson leaders, in particular, were noted for their wealth, wisdom, oratory, or warring ability. As one of the main characteristics for which the village leader was respected was pacifism, fighting leadership was usually accorded to another individual, one of the warriors. Leadership was not, strictly speaking,  38  hereditary, but the fact that a man was the son of a "chief" gained hi a certain standing which, if it was backed by the necessary personal qualities, led to a descent of "chieftainship" from father to son® Concept of Property; With the exception of fishing stations, resource areas were not owned by individuals, families, or villages. All Canyon inhabitants seem to have had equal access to the surrounding mountain hunting areas. Salmon dip-netting stations, however, were owned by families. Of the Tait fishing stations, Duff (1952:77-78) writes: With few exceptions, these were rocky points in the canyon above Yale. Nominally the station was owned by the head of the family; however, all of his descendants could claim the right to use it, and he was considered extremely selfish if he forbade anybody, related or not, reasonable use of the station. Usually there was no problem, at least in recent times, as practically everybody who wanted" to fish could claim the right through kinship to use at least one station. . . Formerly, most of the stations in the canyon were owned by families in the villages close by, although through the web of kinship, most people along the river could and did claim the right to use at least one . . . . The status of the Cowichan, Nanaimo, and other nonStalo tribes who. . . came to this fishery every summer is not clear. My informants did not know of this former summer invasion. Perhaps this pattern was upset in the gold-rush days, and my informants knowledge did not extend that far back. Or perhaps more likely, the outsiders came as relatives, privileged guests, or claimants on less desirable stations. Religious and Ceremonial Activities The Gterdian-spirit Complex: Although all Lower Thompsons sought guardianspirits, evidently not all Upper Stalo did so. In place of specific spirit powers, hunters, fishermen,.; warriors, or gamblers often used spells to help them. The shaman, however, almost without exception, was required to  39  seek his power through quest for a guardian-spirit. Sleeping out overnight or for longer periods under rigorous weather conditions and fasting were normally a part of any spirit quest, the objective of which was a dream, vision, or supernatural experience in some fixed form. Other people might receive a guardian-spirit unsolicited. This often took the form of a spirit song, commonly acquired during the winter dancing season,. Winter dancing began in the Canyon early in the fall and lasted, according to one Tait informant, until the first spring salmon was caught. The First Salmon Ceremony; Neither Teit nor Ray (1939) describe a first salmon ceremony for the Thompson, but Duff refers to a Thompson woman at Yale who said such a ceremony wss held at Spuzsum. This was undoubtedly related to the Coast Salish practice, for the ceremony was held in Tait villages from Tsawkwem to down-river settlements. The Potlatchj. Competitive potlatching with required return of wealth was known among the Tait, although it does not seem to have been an important means of establishing rank. Many important Tait events were marked by a feast and gift-giving to the assembled guests and it was expected that the most numerous and impressive of these feasts would be given by people of high-rank. Wealthy Lower Thompson people also gave feasts and distributed gifts but it is not known if these "potlatches" commemorated specific events. Apparently no return gifts were expected.  40  Inter-group Relations Trades According to Teit (1900:178), "villages of the Lower Thomson Indians had little intercourse with one another, owing to the difficulty of travel in the Fraser Canyon. Communication between Spuzzum and the villages of the Coast Salish was fairly easy. . . e" Teit (1909:536) mentions that articles from the Coast reached the Shuswap via Chilcotin or Lillooet intermediaries. Haeberlin, Teit and Roberts (1928$156) in discussing the basketry trade point out that very few Lower Thompson baskets were traded to up-river groups. The authors give two reasons; the Lower Thompson made only enough baskets to fill their requirements, and "the trade route for basketry from their region was interrupted by a cross route from the direction of the Lower Lillooet, which reached the Fraser River at Lillooet and Bridge River." The Fraser Canyon does not, then, appear to have been an important trade route between the coast and interior. In light of the obstacles to travel presented by the river, this is not surprising. Warfare: Duff regards the Upper Stalo as basically a pacific people who fought little among themselves but were frequently attacked for slaves and loot by surrounding tribes, and were thus occasionally provoked into revenge raids. The Lower Thompson had but few trained warriors and considered themselves inferior in warfare to the Lytton band. They claim to have been on good terms with all the surrounding tribes, and never to have sent out any war-parties. Their relations with the Coast tribes and Lillooet were on the whole very amicable; and  a  these tribes never attacked them, and were seldom attacked by them. The Upper bands of the Lower Thompsons were different, however, for they occasionally sent- war-expeditions against the Lower Lillooet, and frequently against the Coast tribes. . . . Their enemies seldom returned to retaliate. . . . Once a large party set out from the coast to have revenge for a bloody raid inflicted on them by the Thou©sons. They passed by Spuzzum without attacking the people there and were hospitably entertained. . . . The Lower Thompsons were on very friendly terms with the upper bands and the Okanagan, aid when their hunting-parties met members of the latter tribe in the mountains, they invariably interchanged presents (Teit,19005270-271) Although there was some friction between Tait and Lower Thompson over fishing-stations, on the whole, Canyon relations seem to have been singularly peaceful. Marriage; While most Upper Stalo married outside their own village, only people living near the boundaries, or high-ranking families married outside the Upper Stalo areas. Twenty-five marriages in which one or both partners were Stalo were spread over four generations . . . . In eighteen of the twenty-five cases, Stalo men married. Stalo women. Only three found wives in their home villages, but seven more found wives within about 10 miles. In the other seven cases, Stalo men found wives outside the area two Chehalis and one Hope men at Douglas, two lale men among the Thompson, one Pilalt among the Squamish, and one Langlev man at Lummi (Duff,1952:95-96). In his 1894 report on the Indian tribes of the Lower Fraser Pdver, Boas (1894.:454-4.56) includes two linked genealogies. A three-generation sample from both genealogies dating back to about the tine Esilao was occupied shows as wide a geographical range as that mentioned by Duff. Boas lists the village origin of only one partner of each marriage, but nevertheless, the pre-contact geographical dispersion is significant. Of twelve marriages shown for three generations of one family.  uz  four partners come from a village north of Harrison Lake; two from Esiiaoj one each from Port Douglas, Lillooet, Nicomen, Cowichan, and Saanich and one partner is listed as a Musqueam (B0asjl894.s4.56).. Of seven marriages for the equivalent three generations of the other family, four partners came from Chelhalis; two from Esiiao j and one from Kwantlen. An additional note shows that the mother of one Esiiao partner was a Thompson (Boas, 1894sfacing 4-54)® In a passage cited earlier, Duff questions the status of non-Stalo tribal groups who came to the Canyon fishery. The basis for this reported general use of choice fishing locations may have lain in part, as Duff suggests, in widely dispersed kinship bonds, such as those described in the above genealogies. This conclusion is strengthened by Suttles' recent enquiries (i960) into the connection between Coast Salish affinal ties and subsistence activities. Fewer data are available concerning intermarriage in the Lower Thompson, portion of the Canyon. Teit refers to frequent intermarriages between the Lytton tend and nearby bands of the Upper Thompson. The ease of communication between Spuzzum and the Upper Stalo villages led to repeated contacts, and intermarriages were, therefore, not Infrequent.  43  SUMMARY AND ANALYSIS OF ETHNOGRAPHIC MATERIAL Table II is a list of selected traits drawn from the preceding ethnographic material. Two criteria for inclusion were: first, that Duff mention the trait in his Upper Stalo ethnography? and second, that the trait show a definite coastal or interior relationship. The Upper Stalo were assigned a pivotal position in selection of data because Esilao Village is within their territory. Coastal or interior trait affinities were established with the aid of the three volumes from the University of California Culture Element Distribution Series compiled by Barnett (1939), Drucker (1950),. and Ray (1942). Drucker's and Barnett's studies were used because they are complementary. However, with but one exception, (item 11), no Gulf of Georgia Salish trait was considered coastal unless it appeared also in the Northwest Coast list. As the two main culture areas exhibit considerable regional variation, Table II includes only those sub-areas most likely to have influenced the Canyon groups. These sub-areas are the Northern Plateau and the Wakashan or Central Coast region. Of the twenty-seven Tait items, eleven are of coastal and sixteen of interior orientation. The Tait and Lower Thompson share twenty-three traits, eleven to the exclusion of any coastal groi$>, and six to the exclusion of the Northern Plateau groups. Six of the Tait and Lower Thompson traits are shared with the Gulf of Georgia Salish but not with the Wakashan area. Of these last six, only one is not shared also with the Northern Plateau.  44  TABLE II DISTRIBUTION, OF SELECTED CULTURE TRAITS Key; NWCs. Northwest Coast Culture Area j GS: Gulf of Georgia Salishy Ts Taitj LT: Lower Thompson} NP: Northern Plateau Sub-culture Area j X: Trait present? (X); Trait present but apparently unimportant j ?»; Trait listed but doubtful© Trait  NWC  GS  T  LT  NP  1« Salmon split dorsally  X  X  X  2 e Blanket robes  X  X  X  3® Plank houses  X  X  X  X  X - (X)  5® First-salmon, ceremony  X  X  X  (X)  6. Semi-lunar slate knives  X  X  X.  X  7® Wooden dishes  X  X  X  X  8. Head deformation as status Indicator  X  V A  X  (X)  X  X  X  X  X  X  11. Use of dog wool  X  X  X  12. Coiled basketry  X  X  X  X  '•ftT .F  X  X  X  14. Isolation for all menstrual periods  X  X  X  X  15. Projectile points of chipped stone  (X)  X  X  16. Leadership based on respect, not heredity  (X)  X  X  X  X;  X  >.'.  18® Dip-nets, scoop-type with slip rings  X  X  X  19. Stone-boiling in. baskets  X  X  X  X  X  Potlatch with equivalent or better return  9® Corpse placed in elevated box 10. Dug-out canoes  13. Dome-shaped sweathouses  17. Semi-subterranean winter houses  .  X  .• X  ?  20. Two-piece wooden tongs  (X)  21. Sinew-backed bows.  X  X  X  22. Snakeskinrcovered bows  X  X  X  X  X  (X)  X  X  25. Smoking with elbow or tubular pipe  X  X  X  26. Cradle urine conduit  X  X  X  27. Seclusion, of pubescent girls in separate hut  X  X  X  23. Bark baskets 24. Buckskin shirts  4-5  A list of traits, such as the one presented in Table II, does not lend itself to numerical analysis, since such comparisons do not consider, for example, the relative importance of the individual traite Recognizing these qualifications, we may, nevertheless, make some general observations concerning the results of our comparative ethnographic study® Of significance to our enquiry is the high proportion of items shared by both Canyon groups. Approximately 85 percent of listed Tait characteristics are found also in the Lower Thompson - a relatively strong indication of the uniformity of Canyon culture. This uniformity of culture justifies use of both Tait and Lower Thompson ethnographic sources for interpretation of archaeological data at Esiiao. Conversely, this same uniformity may permit generalization of the archaeological findings to an even larger area than that held by the Esiiao villagers or the Tait. Thus, the results of our excavations m y add not only to knowledge of Tait culture but also, with reasonable reliability, to our knowledge of Lower Thompson culture. Finally, the above analysis of Canyon Indian traits bears on a subsidiary objective of this study. Since about 60 percent of the Tait and 70 percent of the shared Tait and Lower Thompson elements are of Interior orientation, we would infer that Canyon culture aligned more with the interior than with the coast.  III. ARCHAEOLOGICAL INVESTIGATION AND ANALYSIS Site Descriptions Esilao, or DjRi 5 in the Borden (1952) site code scheme, is a pit house village situated in the northeast corner of Yale Indian Reserve Number 21 (National Topographic Series, Sheet 92 H/ll Nest Half). Designated by the Canadian Department of Indian Affairs as a fishing station, the reserve is located on the east bank of the Fraser River about two-and-one-half miles up-river from the Tom of Yale® Esilao is built on a bench of compact, gravelly soil which extends from the steep eastern face of the Canyon to a high rocky point which projects into the main stream of the Fraser. Immediately to the north of Esilao, an unnamed, intermittent creek has at one time carved a deep ravine. The creek's present channel cuts through the bench in a southwesterly direction between the village and the Canyon's east wall. The village is approximately 100 feet above mean river level and 130 yards from the river's edge. Its elevation is close to 300 feet above sea-level. Threehundred yards to the south of the village, a large all-weather stream, Siwash Creek, enters the Fraser. Deciduous trees and shrubs including broad-leaf and vine maple (Acer macrophyllum and A. circinatum), white birch (Betula papyrifera), and hazel (Corylus sp.) grow on the floors or rims of all house-pits except number i, (see below). The surrounding evergreen forest is re-establishing itself over much of the site, and all pits, again with the exception of the fourth, support a growth of grand fir (Abies grsndis) saplings. On house-pit 1 is the only large evergreen, a grand fir of 18-inch diameter  FIGURE  3 : M a p of E s i i a o V i l l a g e  47  at breast heighto The dominant ground-cover throughout the village is twisted stalk (Strentopus amplexifolius) although Oregon grape (Mahonia sp.) , salal (Gaultheria shallon), and sword fern (Polystictium munitum) are also common. Oregon grape is particularly abundant in house-pit 4® The relative positions of the four house-pits at Esilao are shown in the accompanying nap (Figure 3)« Major dimensions are as follows! TABLE III INTERIOR DIMENSIONS OFTOEXGA.VATEDHOUSE-PITS  PIT  :  "LENGTH*  WIDTH*  (feet)  (feet)  1  40  35  4  sub-rectangular  2  60  60  9  circular  3  52  48  6  sub-rectangular  4  42  42  2 and 7  circular  NO.  •  .  —  —  •  —  RM-FLOOR  HEIGHT (feet)  ^  ^  PLAN:  Measurements taken across pit between rim crests. House-pit 1, set a little apart from the others, has rounded and slumped walls. Several old trees growing on it suggest that it is the oldest pit. The relative ages of numbers 2 and 3 are difficult to determine from, external appearances. They share a common rim but neither pit seems to  •  St> >  5  AS  overlap the other. Inside walls are steep and each pit has a thick covering of humus® Of the pits, the fourth would appear to be most recent because its excavation has cut into the rim of number 3. A small, fenced plot with four graves is in the floor of this pit and there is a tall, painted wooden cross on the rim facing the river. The absence of any trees from this pit m y reflect recent care of the graveyard. Rock-lined Pitss Four shallow, rock-lined pits are present in the eastern part of the village. They range from four to seven feet in diameter ami from eight to twenty inches in depth. One lies to the east of the Figure 3 map area. None could be associated with the occupation of any particular pit house * House-pit Number Is Four of the largest trees in the village grew on the rim or floor of house-pit 1. Their presence indicated this was probably the oldest pit, and it was chosen for systematic excavation. Surmounting the northwest rim was the largest tree, a 30-inch d.b.h.*white birch; on the west rim, the 18-inch d.b.h. grand fir mentioned above; on the southeast rim, another white birch (24-inch d.b.h.); and in the centre of the depression, a dead maple of 20-inch d.b.h. The maple trunk had broken about six feet from the ground and fallen across the northeast rim. The tree had been dead for several years. It was covered with a thick layer of moss and all but the largest branches bad rotted away, although the main trunk wood was still sound® *Diameter at breast height.  49  The northwest and. north portions of house-pit 1 had been altered by two recent activities: 1. A Canadian Northern Railway construction camp had been built about 1910 - 1912 on the narrow bench north and northwest of house-pits 1 and 2. The camp structures do not appear to have affected house-pit 2, but levelling operations for three of the four buildings altered some portions of house-pit 1. Earth from outside the north rim was thrown into the pit while part of the west rim was removed and apparently used to fill a nearby depression. There was also evidence that a path from the camp had once traversed the pit through the west and east rims. 2. Recent probings by an amateur archaeologist had disturbed a segment of the north rim's outer slope® Excavation Procedure: The first systematic investigation at Esilao began in 1961 when the site was surveyed, a benchmark established, and an excavation grid laid out. A five-foot by thirty-foot trench was cut through the east rim and the five-foot square at the eastern extremity of this trench was carried down fourteen feet through continuous cultural strata. Probing the house-pit floor with a sharpened stake disclosed four post holes. All of these were cleared out and, for each, the surrounding fivefoot square was subsequently excavated to the contacted floor level. Excavations in 1962 extended the east-west trench begun the previous year and carried it through the west rim. A north-south trench was also dug, the two cross-trenches meeting at the house-pit centre.  50  The long profiles of these trenches clearly showed the occupation layer and compacted rim core of this dwelling. Squares in the quarters between cross-trenches were dug, therefore, only to the bottom of the house floor deposit and to the compacted surface of the rim. The badly-disturbed northwest quarter, which also contained the large sprawling birch tree, was, except for one five-foot square, not excavated. All of the northeast quarter occupation layer was removed, and this quarter, in spite of a partial overlay of disturbed material, produced the best structural details of the rim and floor. Comparable features were unearthed in the southwest quarter, although time precluded couplete excavation of this last area. The extensively excavated southeast quarter yielded no structural details. The exact locations of all artifacts from the house floor squares and from the east-west trench were recorded with reference to a fixed point, Datum A — the benchmark established in 1961. This reference point is located outside the pit's north-east rim. It is an old iron pin set as a corner post when Indian Reserve 21 was surveyed. All horizontal grid measurements are taken from this pin and its top is arbitrarily established as the datum plane height from which all vertical measurements are made. For artifacts in the north-south trench, other than those in the house floor squares (S 15' - S 45'), exact positions are not recorded but the provenience is given in levels, each two inches deep and two-and-one-half feet square. These levels are horizontal and measured from the datum plane. In the areas where deposits were removed only to the compacted rim surface,  51  no vertical measurements were recorded and horizontal measurements were restricted to location, of the two-and-one-half foot square from which the artifact was taken. All material was screened, normally through six strands to the inch wire mesh. During or after a heavy rainfall, however, four strand mesh was used. Either of these meshes would stop all the large size trade beads found at the site and the six strand mesh would hold most of the smaller beads. For a short while, all buckets were double-screened ~ through six strand mesh and fly-screen — but as this process yielded no additional beads, the procedure was abandoned.  Structural Data Stratigraphy Although brief mention will be made of strata underlying the house-pit, our concern here is with those strata clearly identified with the construction and occupation of pit house 1. The accompanying Figures 5 arid 6 show profiles at west 10 feet and south 25 feet respectively. Where these profiles cut the north, east and south rims, there is a marked similarity to the strata. Each rim shows the loose grey loam that lay within the house-pit depression on top of the compact floor. Each also reveals a compacted red-brown loam forming the rim' core, and beneath this another grey layer. In the eastern and southern rim there is an easily discerned grey layer over-lying the compacted rim core. Because of the disturbance at the north  52  end of the north-south trench, this last layer is missing from the outside slope of the north rim section®  It is also absent from the disturbed west  rim section, where the lip seems to have been cut down well into the redbrown rim core. A schematic reconstruction, of one lip section is given in Figure 7. Here, the top grey layers, both inside and outside of the rim, are identified with the pit house occupation. They probably contain a mixture of midden dumped by the occupants down the slope of the pit house roof, decayed remains of the wood and bark house covering, and part of the earth layer which once covered the entire structure. The red-brown loam is shown as the artificially built-up rim. Material for this was probably scooped up during excavation of the house floor. The lower grey layer may belong to an earlier and smaller pit house on the site. In composition and content, this layer is similar to the upper grey layer. Both contain a marked concentration of fire-cracked cooking stones and of artifacts. Ground-stone tubular pipe remains, for example, are confined almost exclusively to these two grey layers. Beneath the lower grey layer in the rim, and forming part of the pit house floor is a compact red-brown layer identified as a pre-pit house stratum. Below this again, and also forming part of the pit house floor, is sterile, yellow-brown gravel. It is suggested that when the builders of pit house number 1 began work5 the site was already occupied by a smaller house-pit whose centre was somewhat to the west of the bigger one. Evidence for this Is found in the  53  west rim section, where the lower grey layer shows a pronounced slope, conforming, perhaps, to the outside slope of the smaller house-pit. In the west rim profile of pit house 1 is a stratum (labelled "Light Red-brown Loam" in Figure 6) which nay be part of the smaller house-pit rim. In contrast, the lower grey layer in all other sectioned rims is virtually horizontal and overlying compact red-brown loam, indicating that it was midden, laid down beyond the house structure. When the second house was constructed then, its builders incorporated part of the older pit's west wall in the new structure, pushed out the other walls and deepened the floor depression, A post mold buried deep in the east rim may have anchored the butt of one hip rafter from the smaller house although another possible interpretation of this same post mold is given later. Structural Features Plan; The shape of house-pit 1 (Figure 4) is best described as subrectangular. Measuring from one exposed inner wall to another, its greatest dimension after excavation is 42 feet and its least, 36 feet. The long dimension, oriented northwest - southeast roughly parallels the Canyon wall and river course. Although the encircling rim was broken at its western edge, there is no indication that this was a side or tunnel entrance. As outlined earlier, this seems instead to have been a recent disturbance resulting from levelling and filling for the adjacent construction camp.  FIGURE 7  Scharnatic Reprasentatiorv of Rim Profile  5k  Earth Walls; Three rim profiles show the steep inside face of the compacted rim core. For the south, east and north profiles these measure seventy, sixty-five, and seventy-five degrees respectively for an average slope of approximately seventy degrees. In both the northeast and southwest quarters, where large sectors of the rim face were exposed, the slope generally conforms to the above average, although near the western face it becomes somewhat less steep. Floor: The saucer-shape floor Is at its maximum depth about three feet below the surrounding surface level. Its minimum depth, at the edges of the floor, varies from eight to ten Inches. Compact red-brown loam forms the surface over most of the floor area, although in the centre of the depression and towards the west rim the floor is formed of yellow-brown gravel. It is suggested that this sterile layer underlies the red-brown loam in at least the western half of the house ana that scooping out the floor has exposed the gravel® In the northeastern and southwestern quarters, the floor surface is broken by a low bench, three to four feet wide, which extends inward from the wall paralleling the latter. This bench is less distinct in the southeastern than In the northeastern quarter where there is a pronounced sixinch rise from the house floor to the bench. Overlying the floor to the south and east of the central hearth (see below) were several thin patches of white sand. Snail pieces of charcoal and flecks of ochre were mixed with the sand.  55  Hearths : Stains of ash and charcoal were noticeable to the west, south and east, extending from beneath the large maple stump at the house-pit centre. This stump was not removed, but enough of the hearth was exposed to obtain the following details. The main bed of ash, which at its thickest was four inches deep, was contained in a shallow, rock-lined depression. The estimated diameter of this hearth depression is two feet. Rocks lining the saucer were all of tabular shape and uniformly two to three inches thick® Eight feet to the southeast of this main hearth is a smaller concentration of ash and charcoal. The proximity of a main house post (less than two feet away) would make this a hazardous location for a fire, unless, of course, the wood was protected by a rock slab in much the same manner as was the central notched ladder of most pit houses© Notched-log ladder; No feature was found that could be considered definitely the anchor location for a notched-log ladder. However, a large flat rock north of the central hearth may have leaned against the ladder and protected it from heat. Posts and Rafters: Four clearly defined rock-ringed post holes were present in the floor area. These holes formed a rectangle which, as can be seen in Figures U and 8, is aligned with the rectangular pit outline. Dimensions of these holes and their distances from one another are given in the accompanying table. The holes were filled with grey, sandy loam, and as no wood remains were found it is assumed the uprights were removed and  56  not permitted to rot in place® TABLE IV  UPRIGHT BRACE POST BOIE DIMENSIONS Post hole (See Fig.5)  Diameter (inches)  Depth (inches)  '.A'  12  15  A  • B'  12  2G  10'  3  C  8  15  16'  12'3"  D  9  16  12'3» 15«3"  Distance from other post holes  G 9'  D  Ho clear indication was found of the butt end location for each corresponding hip rafter. At two of the four pit corners were rock concentrations which may have been part of the rafter anchoring. But of the two rock piles, only that opposite upright brace D was in reasonably good condition and even it had been disturbed by the amateur archaeologist's digging. This rock concentration lay inside the compacted earth rim, approximately fifteen feet from post hole B. No trace of wood was found in the rock feature. The only other possible hip rafter butt feature noted was a rock pile in the corner opposite post hole G. However, this lay within the house, . only seven and one-half feet from its post hole. As already noted, this portion of the rim had been badly disturbed during levelling for the adjacent camp construction. The foundation of one camp building cut into the housepit -wall and if a hip rafter butt had been anchored fifteen feet from post  57  hole C, it would have lain in the corner of this building. The rocks inside the rim at this point were probably piled there by the construction, camp builders. In the corner opposite post hole B there was no concentration of rocks at a distance of fifteen feet but three feet beyond this was a large rock partly buried in the outer slope of the compacted rim. As this rock lay in line with the expected rafter location, it may have been part of the anchoring for the hip rafter butt® Although there were many rocks on the long downhill slope opposite post hole A, no concentration could be positively identified with the hip. rafter butt for this corner. Similarly, of the many rocks scattered about the outside slope, none f  could be definitely considered 'side rafter anchoring features. Two email rock concentrations shown on the accompanying plan (in the northeast and southeast quarters, Figure 8) were placed too far beyond the rim crest to be the locations of side rafter butts. The post mold buried in the east rim and. tentatively identified with an earlier pit house may equally well have been the anchor point for a side rafter of the later house. With four hip rafters and a corresponding number of upright braces, the appearance of this house must have conformed to the general Northwestern Plateau type described by Hay (1939:134). There are also a number of specific parallels between the archaeological data and Teit's description of the Thompson house (see Figure 2).  3 FIGURE 8 DjRi 5: PIT H O U S E FEATURES  A N  58  Teit illustrates a 70 degree slope for the inside earth vail, a figure in agreement with the excavation data. Similarly, Teit's reference to a below-floor depth of 15 inches for the butt end of each upright post approximates our findings. Both the central hearth and the large rock slab protecting the ladder base are other points of consensus. However, in several important respects, house-pit 1 differs from the Thompson form described by Teit. The contacted, above-ground rim and saucer-shaped rather than level floor are marked departures from the house excavation described and Illustrated by Teit and, indeed, the above ground rim is a feature not mentioned in any Plateau ethnography. Other features revealed by the excavation but not described in. ethnographic sources are rock concentrations at hip rafter ends and about the base of main iprights, and a low earth bench rising from the floor in parts of the house. This latter feature may have served as a sleeping platform — much like wooden benches in the typical west coast plank house. The sub-rectangular floor plan Is a significant difference, allying the house form more with a Lillooet variant of Northwest Plateau earth lodge rather than with the Thompson type. The Lillooet house had four rafters aid uprights but no side rafters, structural details consistent with the excavation data from house-pit 1 where the rectangular plan and failure to locate their bases would suggest side rafters were not used.  Artifacts Because the pit house was built on ground which had been occupied for  59  many centuries, the process of construction, involving the scooping out of a saucer-shaped floor, and the re-depositing of this removed material in a built-up rim, resulted in considerable mixing of artifacts. More recently, disturbance of the northern rim of the house has further mixed artifacts from different strata.. Therefore, only certain undisturbed layers within a portion of the site are considered reliable and artifacts from these layers only are catalogued ana analysed below. Artifacts included are those recovered from: the grey loam overlying the pit house floor and the grey loam on the outer slope of the rim. No artifacts from the artificially built-up rim (compact grey-brown loam) or from the pre-pit house strata (contact redbrown loam) are considered. Similarly, artifacts from the disturbed grey loam on the north lip and at the west end of the east-west trench are rejected. The lower grey loam layer, associated with a possible earlier pit house, is revealed only where the two cross-trenches cut through the rim. Although the artifact count is therefore.not comparable to that recovered from the upper grey layer, the sample obtained allows for comparison of at least some classes of artifacts. For the sake of simplicity, we shall distinguish between the two grey loam layers as the upper and lower grey deposits, recognizing that this superior-inferior relationship is true only of the rim deposits. Each artifact from Esilau is labelled with the site code (DjRi 5) aid is assigned an individual artifact number. However, since all artifacts  60  are from this one site, when items are referred to by number in the following pages the site code is omitted.  Stone Industry Chipped Stone Projectile Points: The typology used by Strong (1935:88-89) and Strong, Schenck, and Steward (1930:78-79) provides an adequate basis for descripti of all but four chipped stone projectile points associated with the occupation of Esilao house-pit No.l The following classification includes all but the four aberrant forms found in House-pit 1: Mi. Mot Stemmed A, Leaf-shaped b. Pointed at one end 1. Convex base 2. Straight base B. Triangular a. Straight base lo Two side notches b® Concave base 1. Two side notches c. Convex base S» Stemmed A. Contracting stem a. Shouldered only b. Shouldered and barbed  61  B. Parallel-sided stem a. Shouldered only b. Shouldered and barbed c. Neither shouldered nor barbed C. Expanding stem a. Shouldered only 1« Convex: ba se 2. Straight base 3. Concave base b. Shouldered and barbed 2. Straight base 3- Concave base c. Neither shouldered nor barbed Of 123 complete or fragmentary points from the upper grey deposit, 80 show the diagnostic features necessary for classification. Similarly, of 55 specimens from the lower grey deposit, 33 exhibit diagnostic features. Tables V and VI summarize quantity, dimensions and relative importance of the -various point types from the two grey deposits. A comparison of the percentage distribution of point types in the upper  and lower grey deposits is given in Table VII. The strong correlation, exhibited in this table, in spite of the disparity in sample sizes, indicates virtually no distinction between the two deposits on the basis of point type. The most important types in both assemblages are therefore considered together in the following discussion.  TABLE V .DIMENSIONS OP PROJECTILE POINTS FROM. UPPER GREI DEPOSIT  TIPS:  N  ffibl NAb2 NBa : NBal NBb HBbl .NBo  2 5 2 5 2 10  UNSTEMMED  M  Sfta SAb SBa SBb SBc SCal SCa2 SCa3 SCb2 SCb3 SCc STEMMED  TOTAL  2 1 2 5 49 15 6 4 1 1  LENGTH (cm) R . X N 3 »3-4»4 2.7-4.2 2 «f) 1.2-3.0 1.8-1.9 1.6-3.4 1.6-2.8  3.9 3.5 2.3 2.1 1.9 2.3 2.3  2 5 2 5 2 4 4  WIDTH (cm) R ; ,. X N 1.3-2.2 1.0-2.0 1.3-1.5 .8-1.7 1.4-1.6 X «X^X .9-1.6  1.8 1.4 1.4 1.2 1.5 1.3 1.3  2 5 2 5 2 9 4  THICKNESS"* (cm) W/L INDEX R ,X N .. R /. X ft .5-«9 0.7 «5~®-7 ,. 0.6 0.4 .2-.5 0.3 •3-.5 0.4 .2-.6 0.4 »1— .3 0.2  2 5 2 5 2 9 4  30-67 37-48 52-71 52-67 74-89 35-69 44" 70  43 41 62 58 81 57 56  2 5 2 5 2 4 4.  $ OF TOTAL 2.50 6.25 2.50 6.25 2.50 12.50 37.50  2.7-3.0 -  -  2.0-2.8 2.8-4.7 1 »4-3.4 1.5-3.9 2.3-3.6 2.0-3.1 =  —  2.9 CO 1 a5 2.5 3.5 2.7 2.7 2.7 2.5  2 -  1 4 4 8 10 4 3  «E»  3.0  1  1.8 2 1.6 1 1.1-1.3 X«2 2 1.8-2.0 1.9 2 .9-2.3 1.7 4 1.1-1.7 1.4 9 .8-2.5 1.5 15 .9-2.6 1.7 6 X • 2 © V 1.9 3 1.9 1 1.2 1 1.6-1.9 .  -  .3-.5 „  -  •1- .3 .3-.7 .3- .6 .2-.6 .3-.7 .3-o6 —  0.4 2 0.4 1 0.2 2 0.2 5 0.5 4 0.4 9 0.4 15 0.5 6 0 40.5 1 0.5 I  59-63  61 «  2 «  73 69 50 55 46 55 77  «  «,  1 2 4 8 10 A 3  09  40  1  66-71 32-79 39-78 37-60 39-72 70-87  ™  2.50 1.25 2.50 6.25 5.00 11.25 18.75 7.50 5.00 X« —1,25  _50  62.50  80  100.00  Width Explanation;: KI: Number j Rs Rangej X: Mean; W/L Index: Length x 100  63 TABLE VI DIMENSIONS:OP PROJECTILE POINTS FROM LOWER GREY DEPOSIT  TIPE  Ns  NAb 2  2 3 1 4 2  NBa 1  NBb NBbl NBc  TfflSTEMMED SAa  SAb SBa SBb  SBc SCa 1  SCa 2 SCa 3 SGb 2  :  LENGTH (cm) Y " A, R M  2.o»2;7 2.4 3.3-3 .4 3.4 1.6 1.9-3.5 2.7 2.4  2 2 1 2 1  WIDTH  R 1.1-1.6 1.2-1.9 -  1.3-1.5  (cm) Ni  1.4 1.7 1.2 1.4 1.4  2 3 1 A  2  THICKNESS (cm)  R • X  :  0.5 .2-.7 0.5  0.2  • 2—. A 0.3 0.3  1: 2 3 1 A 2  W/L INDEX  R  X  55-59 56-58  57 57 75 57 63  56-58 .  .  B 2 2 1 2 1  12 1 1 2 3 2 2 8 1 1  $ OF TOTAL  6.06 9.09 3.03 12.12 6.06 36.36  -  2.2—2.4  2.3-3.5  3.2-3.7 1.8-2.8 2.3-4.0 -  3.0 -  2.3 2.9 3.5 2.3 2.9 2.6  2.8  1  «  na  2 3 2 2 6 1 1  -  1.7-2.3 1,3-1 1.0-1.6 1.3-1.9 -  •L X 1.1 2.1 1 «4 1.3 1.5 Xo I)? Xa  1 1 2 3 2 2 8  1 1  -  .3-.5 .3" .6 ,2-.5 .2-.5 -  -  0.3 0.4 0.2 0.4 0.5 0.4 0.3 0.5 0.4  1 1 2 3 2 2  8 1 1  .  46-48 63-82 -  51-57 33-69 -  -  47 «=  47 73 41 54 56 54 50  1  a,  2 3 2 2 6  T. a  1  3.03 3.03 6.06 9.09 6.06 6.06 2A.25 3.03 3.03  STEMMED  21  63.64  TOTAL  33  100.00  Explanations N: Number: Rs Range: X: Mean:. W/L Index: x 100 ' ' Length  64  1. Unstemmed, triangular, side-notched points (NBalj HBblj Plate 1:5, 6,8,9; Plate 11:3-6): The only two chipped stone points illustrated by Teit (1900:242) would fit this description. Their importance at Esilao is shown in Table VII where they total 18.75 percent of the upper grey deposit point assemblage and 21.21 percent of the lower deposit point assemblage. TABLE VII RELATIVE IMPORMGE OF POINT TXPES (Presented as percentage of total points for each deposit)  Point Type  Upper Grey Deposit  HAbl KAb2 MBa IBal NBb NBbl NBc Total Unstemmed  2.50 6.25 2.50 6.25 2.50 12.50 5.00 37.50  SAa SAb SBa SBb SBc SCal SCa2 SCa3 SCb2 SCb3 SCc Total Stemmed  '  2.50 1.25 2.50 6.25 5.00: 11.25 18.75 7.50 5.00 1.25 1.25 62.50  Lower Grey . Deposit —«'  6.06 9.09 3.03 12.12 6.06 36.36 3.03 3.03 6.06 9.09 6.06 6.06 24.25 3.03 3.03 ....  63.64  6:5  At Sanger's (1962) Chase site on the Northern Plateau, the NBal form (Sanger's NBa2) totals 19 percent of the points. These are also common among recent Carrier (Borden, 1952s PI.1,10-15). Two NBbl specimens from the upper grey deposit at Esilao exhibit the asymmetrical basal spur (Sanger's NBd form) found on eight points at Chase and on three reported by Borden (1952s42) from Carrier territory. NBal and NBbl forms were also important in the ColumbiaWindermere Lakes area where they totalled 50 percent of the chipped points found (Borden, 1956:89). In the McNary Reservoir, the two forms (Osborne's Ila and lib) constituted 14 percent of the points found (Osborne, 1957s80). Collier et al's NBb2 type (NBbl in the classification adopted here) formed 11 percent of the chipped points found in the Upper Columbia region. (Collier et al, 1942s62). Unstemmed, triangular, side-notched points are rare on the coast. Four specimens are known from Stselax (Borden, pers. comm.) and three from Cattle Point (King, 1950:16). Stemmed, with parallel-sided stem and barbs (SBbj Plate Isl5,l6; Plate II$14-16: Of the points falling into this classification, one from the upper grey deposit (Plate Isl6) and two from the lower grey deposit (Plate IIsl4,15) have concave edges. These forms (Cressman's Type II) are fairly common at the Dalles where, for example, at the WS-4 Roadcut they comprised 12 - 13 percent of the levels one and two chipped point assemblages (Cressman, 1960:61). SBb forms, including specimens with concave, straight and convex edges, totalled 24 percent of the Dalles-Deschutes collections studied  66  by Strong, Schenck, and Steward (1930s Table 4) but only two percent of the total points from the Upper Columbia (Collier et al, 1942s63). Similarly, Borden (1956:87) found only one SBb form during his East Kootenay survey. An SBb form with concave edges was found at Stselax (Borden, pers. comm.). 3. Stemmed, with expanding stem and shoulders (SCal, SCa2, SCa3; Plate Is 1S-24J Plate 11:17-24) s  The three variants of the SCa form total 37.5  percent of the upper grey deposit point assemblage and 33.33 percent of the lower grey deposit points. In the Upper Columbia region, SCa forms accounted for five percent of chipped points found (Collier et al, 1942:62) while the Kootenay River Valley (Borden, 1956:89) yielded a much higher proportion of SCa found - about 44 percent. At Chase, (Sanger, 1962:22) they form less than four percent of the point assemblage. According to Smith's (1950:18) compilation, the SCa forms total ten percent of the Dalies-Deschutes collections. Isolated specimens occur at Stselax (Borden, pers, comm.) and Cattle Point (King, 1950:17). Table VIII compares the proportions of stemmed to unstemmed chipped points from six sites or survey areas. The percentages have been calculated from sources cited on the previous pages. Pour points which did not fit the classification used in this study are described below: 1. No.2622 (Plate 1:11): Length, 2.4 cm.; width, 1.0 cm.; thickness, .3 cm.  67  TABLE VIII: COMPARISON OF SIS®©) AND UNSTEMME3> POINTS FOR SEVEN SITES OR SURVEY AREAS (Figures are percentages of total chipped point assemblages)  Site or Survey Area  Stemmed  Unstemmed  Kootenay River Valley  82  18  Dalies-Deschutes  81  19  Esilao  63  37  Upper Columbia  34  66  Columbia-Windermere lakes  30  70  McNary Reservoir  22  78  Chase  17  83  This point is basically triangular, but has roughly parallel sides for the basal half of its length. Found in the upper grey deposit. 2. No. 2617 (Plate 1:12): Length, 2.2 cm.; Width, 1.1 cm.j thickness^ .4 cm.. This point has projecting shoulders and an expanding, eoncavebased stem. It was found in the upper grey deposit. 3. No. 2172 (Plate 11:10): Length, 1.8 cm., width, 1.9 cm.j thickness, .2 cm.. Long lateral spurs project from the base of this small, triangular point. It was recovered from the lower grey deposit. 4. No. 1993 (Plate 11:9): Length, 1.9 cm.j width, 1.4 cm.j thickness, „4 cm A single lateral barb extends from near the base of this markedly asymmetrical point. It was found in the lower grey deposit.  68  Drills: Two complete (one is illustrated in Plate 1:28) and two fragmentary drills were found at Esilao, from the ipper and lower grey deposits respectively. Table IX summarizes their dimensions,, TABLE IX DIMENSIONS' OF DRILLS  Deposit  Artifact No.  Length (cm)  Width Pile (cm)  Width Base (cm)  Thickness Pile (cm)  Upper  9399  4.6  .6  1.5  .5  n  4651  3.5  .7  1.9  .5  Lower  2081  1.6 +  .7  X  .3  u  3651  1.9 +  .7  1.4Jr  .5  Teit (1900:183)' refers to drilling of wood with stone points although no particular form of drill is described. Many of the unfinished and fragmentary stone pipes found at Esilao showed coarse drilling. The similarity between the average drill hole dimension (about .7 cm.) and the width of drill piles (see Table IX) nay indicate the use of stone drills for at least a part of the pipe manufacturing process. Chipped stone drills are reported from the Dalles (Cressman, I960: 53), McNary Reservoir (Osborne,1957:81), the Upper Columbia (Collier et al. 1942:64.), Central British Columbia (Borden, 1952: PI.1,25-27), the East Kootenays (Borden, 1956:95-96) and Chase (Sanger, 1962:24). Specific  69  parallels in shape of base are found in the McNary Reservoir and at Chase, Scraper-Knives;•A complete analysis of the chipped knives at Esilao was not made. Table X summarizes the dimensions of three chipped scraperknives (Plate IIIs2) from the upper grey deposit. They resemble an implement illustrated by Teit (1900:184) and identified by him as a stone TABLE X DIMENSIONS OF CHIPPED SGRAPER-SNIVES FROM. UPPER GREY DEPOSIT  Artifact 3298  Length . 2.1  Width 1.6  Thickness  8866  2.1  1.4-  .5  tt  8781  4.2  1.8  .5  11  Remarks Steep, unifacial retouch  knife. Since the three specimens from Esilao seem better adapted for scraping operations (the blades are thick and cutting edges steep), they are called here chipped scraper-knives. Chipped End-Scrapers: The many hundreds of flake scrapers from this site were not analysed. They must comprise 30 to 40 percent of the total assemblage if all unretouched but used flakes and crudely retouched flakes are included. One well-made end-scraper from the upper grey deposit Is included in this study because it has a specific ethnographic parallel. Based on a basalt flake, it has steeply retouched sides and end (Plate III:l). The business end is rounded, and slightly expanded. In side view, the  70  specimen curves down at the scraping end. The length is 3.2 cm.; width of stem, 1.9 cm; width of scraping head, 2.2 cm.; thickness, .6 cm.. Teit illustrates a scraper of this form (1900sl85) and refers to its use in dressing hides. Borden observed a similar scraper being used by a Carrier woman to dress a hide. The tool was made of iron, but was hooknosed, rounded, and of approximately the same size as the Esilao specimen. Choppers? Crude choppers were found in both upper and lower grey deposits. These tools are not considered diagnostic artificats and were not examined in this study. However, we may note that although fairly common at Esilao, they are rare in recent Coast Salish assemblages (Borden, pers.comm.). Boulder Chip Scraper-Knives (Plate VI:2,4.): These crude cutting or scraping implements are split from the surface of large boulders or cobbles. Although one or more edges may show wear, the tools display no evidence of retouching but appear rather to have been kept only until use dulled their edges® Table XI shows no significant difference between the two grey deposits although there does seem to be a tendency for scraper-knives in the lower deposit to be on the average smaller than those in the upper. With the discrepancy in sample sizes, little importance can be attached to this. Our ethnographic Investigation disclosed no specific references to the use of boulder chips by Canyon inhabitants. However, when we consider that among the Shuswap such tools were the usual form of skin-scraper (Teit, 1909sA73), their presence in quantity at Esilao raises the question of  71  whether skin preparation and skin clothing may not have been of greater importance in the Canyon than the ethnographies would indicate., Their use as knives, particularly for preparation of fish, is also a possibility. mBEE XI QUANTITIES AND DIMENSIONS OF BOULDER CHIP SCRAPER-KNIVES Long Dimension (cm)  Upper Grey Deposit  Lower Grey Deposit  Number  % of Total  Number  under 6  25  17.5  7  19.0  6-8  34  2/+.0  11  28.0  8 - 10  43  30.0  11  28.0  10 - 12  27  19.0  3  7.5  12 - 14  7  5.0  3  7.5  14 - 16  4  3.0  3  7.5  over 16  2  1  2.5  Total  142  100$  Total  39  % of Total  100$  Boulder chip tools have been reported archaeologically on the Plateau at Lytton by Smith (1899:146-14-7) j on the Upper Columbia by Collier et al (1942:67) J in the Kootenays by Borden (1956:92-93) J in McNary Reservoir by Osborne (1957:48), and at the Dalles by Cressman (1960:53). The McNary median size, 8 - 1 0 cm., is identical to that for Esiiao. An Upper Columbia  72  figure of 77 percent of scrapers measuring between two and six inches in diameter is roughly equivalent to Esilao's 78 percent between six: and fourteen centimeters® Ground Stone Ground. Slate Points; Only two ground slate points were found, both in the upper grey deposit, and both of SAa form, although there was considerable difference in the widths.. TABLE XII DIMENSIONS OF GROUND SLATE POINTS Artifact No.  Length (cm)  'Width (cm)  5275  8.3  3.1  .3  2.6  4740  7.1  1.4  .3  £  Thickness (cm)  Stem Length (cm)  Figure 9: Slate point No. 5275 No. 5275 (Fig.9) is roughly ground over the entire surface of one face while on the opposite face, only the blade is ground. The blade edges are  73  slightly convex and bifacially bevelled with a rounded bevel. In crosssection, the roughly blocked out stem is lenticulate. No. A740 is similar to the other point in all measurements except width. In contrast, all surfaces are ground and the stem cross-section is rectangular. The blade has an unusual edge - the top 1.9 cm. of length is rectangular in cross-section, and only the middle 2.6 cm. of the point is bevelled to a cutting edge. No explanation is offered here for this edge treatment. The blade is well-finished and does not seem to be in the process of manufacture® The first of these points, because of its size and wide, rough stem, may be interpreted as an end blade for a knife. No mention is made of such knives in the Canyon ethnographies. The second point may have armed a projectile® Slate arrow points, mentioned by Duff's Tait informants, are known neither ethnographically nor arenaeologically from the Plateau. Ground Slate Knives: Only four complete or nearly complete ground slate knives were found (Fig.10). Table XIII records the many fragments of ground slate. Distinctions are made on the basis of size or cross-section. Described as "Thick" are fragments of a uniform thickness measuring greater than .3 cm. "Tapering" fragments exhibit a marked taper in cross-section. At their thickest edge, they are all greater than .3 cm. "Thin" fragments measure .3 cm. or less in thickness and do not display any noticeable taper apart from the bevelled cutting edge. A further distinction is made  a.  "  —  —  •  FIGURE 10: Slate knlvas  —  (xi)  3161  between those fragments bevelled to a knife edge and those without» The four complete or nearly complete knives, all from the upper grey deposit, are described below® 1. No. 2506 (Plate 111:4; Fig.10a): The ends of this rectanguloid knife slope in from a long dimension near the cutting edge of 13 cm. to 11.8 cm. at the back. The depth of the blade is 6.4 cm.. The blade averages .3 cm. thick at the back and tapers evenly to the knife edge. TABLE XXII GROUND SLATE KNIFE FRAGMENTS  Description  Upper Grey Deposit Knife No Knife - Edge Edge  Lower Grey Deposit Knife No Knife Edge Edge  Thick  3  15  -  Tapering  3  6  1  176  197  52  Thin  4  64  Notches have been chipped into the back at intervals of about 1.5 cm.. 2. No. 6525 (Fig. 10b): The ends of this knife are rounded. Length, 9.1 cm.j depth of blade, 3.4 cm., tapering to 2.8 cm. at one endj thickness, .3 cm.. 3. No. 4641 (Fig. 10c): The ends of the knife expand from a long dimension of 6.5 cm. at the cutting edge to 8.2 cm. at the back. The blade depth is 4.7 cm. and the thickness .3 cm.. One end is straight, the other slightly convex. 4. No. 4649 (Fig. lOd): The blade of this knife is 9.2 cm. long, the greatest  Upp«.r Grey Deposit  Lowci* Gray Deposit  FIGURE  11 :  Slat<z knife ands  depth 5.2cm. and the thickness .3 cm.. One end curves from the cutting edge into a high rounded back. The other end slopes out and up for 2.3 cm. before curving sharply inward to the high rounded back. The 2.3 cm. long sloping end has been bifacially ground to a cutting edge « End fragments of six upper grey and two lower grey deposit specimens Indicate there were many variations in outline. Without complete knives, no attempt is made to classify them. They are shown in Figure 11. The use of ground slate knives for butchering fish is reported in both Canyon ethnographies although the specific form of knife is not described beyond mentioning a curved blade and short handle (Teit, 1900s234) <> The notches in No. 2506 (above) are probably to facilitate halting of a wooden back. The smaller No. 6525 shows wear on the back and may have been used without a haft. Ground slate is abundant at many coastal sites (Drucker, 1943:51-52). At Marpole, for example, slate fragments total about 3000 out of 8000 artifacts in the University of British Columbia collection (Borden, 1962:13). In the Canyon, thin ground slate knives, similar to those which predominate at Esilao, have considerable time depth. At DjRi 3 they have been radiocarbon dated at 2360 ± 60 BP (Borden, 1961:1). Smith (1899:140) reports ground slate from Lytton and a small amount from Kamloops (1900:414). Sanger (1962:35-36) found only two rubbed slate artifacts at Chase. Ground slate is rare at other northern Plateau sites  76  and is apparently absent from the southern portion of the Plateau® Crude Slate Knife (Plate 111:5): This rough, slate slab, 12.7 cm. long, has been bifacially ground on one long edge to form a knife. Judging by the wear along the back, no handle was attached. The knife averages .6 cm. thick and in outline, tapers from a maximum width of 8.5 cm. to about 5.5 cm.. The back is fairly straight, but the ends and cutting edge are all excurvate. Knives of this form are not mentioned in the Canyon ethnographies, nor are archaeological parallels known. Celts: Complete or fragmentary celts came only from the upper grey deposit. Of five specimens, two (2608, 9036) were intact. 1. Ho. 2608 (Fig. 12): A well-shaped adze or chisel blade, 5.1 cm. long, 3.4 cm. wide at the bit and tapering to 2.8 cm. at 1 cm. from the rounded poll. Thickness of the body is 1.1 cm. The bit is square and ground asymmetrically bifacial with broadly-rounded bevels. The celt is Bade of serpentine. ..  Figure 12: Celt No. 2608.  77  2. Ho. 4-650! A. 10.8 cm. long fragment of a roughly shaped serpentine adze blade. The fragment includes part of a unifacially bevelled bit, a vertically ground side and part of one roughly finished face® 3. No. 8248; A 2.1 cm. long fragment of a well-polished, unifacially bevelled nephrite celt bit. 4. No. 2228:. A 3.7 cm. long fragment of a ground and polished, object that appears to be part of a celt edge® 5. No. 9036: This boulder chip is considered a celt because it has been ground at one edge to a sharp chisel or adze bit. The rest of the chip is unfinished• The length is 6.7 cm., width 5.7 cm., and the average thickness 1.4 cm.. Grinding extends back about 1.4 cm. from the bit on one face and 2.2 cm. on the other. Celts are not uncommon at coastal and interior sites. Parallels to the one complete form from Esiiao (2608) are to be found at Chase (Sanger, 1962:31) where five of twenty-six blades tapered from bit to poll; at Cattle Point (King, 1950:35)5 and at Stselax (Borden, pers. comm.) where blades of similar size and shape have been unearthed. The use of stone adzes in the construction of pit houses was mentioned in the lengthy quotation from Teit presented earlier, and their use was discussed also in the section on wood-working tools. Carving Chisels: Two small chisels or carving tools were found in the lower grey deposit. 1* No. 5456: This small and beautifully finished tool is made of transparent jade. It measures 3.0 cm. long, is .6 cm. wide at the bit,  78  and tapers evenly to .2 cm. at the poll. Behind the bifa dailybeveiled bit, the thickness is .3 cm. and the body tapers to .1 cm. at the poll. The cross-section is sharply rectangular except for 1.7 cm. measured from the poll end. For this length, the cross-section is also rectangular but the edges are rounded. 2. No. 9541s Of approximately the same dimensions as the above, this: chisel is less carefully made. It is 2.7 cm. long, and the parallel sides are .5 cm. apart for 1.9 cm. of the length. The tool has a .4 cm. wide stem, flattened and twisted so that the planes of bit and stem are at a 4.5 degree angle to one another. The thickness is greatest at the point where stem and body join (.4 cm.) and from here the body tapers to a convex, bifacial bit and the stem quickly flattens to a uniform .2 cm. thickness. The body is roughly rectangular in crosssection for approximately „9 cm. of the length behind the bit and is circular from there to the stem, which is roughly rectangular. This tool is made of serpentine. No ethnographic or archaeological precedent is known from Plateau or Coast for these minute chisels or carving tools. Their possible use in the manufacture of stone pipes will be discussed below. Pipess A total of 62 pipes or pipe fragments was unearthed at Esilao. Among the 34 pipes or pipe fragments from the upper grey deposit and 28 fragments from the lower grey deposit was only one complete .pipe (Plate Vs9)„ This study cannot, therefore, attempt a pipe typology but can only describe  79  the fragments found and indicate any consistencies in bowl, stem, or stem end forms. None of the 62 pipe or pipe fragments appeared to be remains of elbow type pipes. In the following discussion all pipes are assumed to have been of the straight or tubular form. The following terminology is employed. "Bowl" refers to the enlarged end of the pipe in which the tobacco was placed. It includes all of this end from the opening to any discernible juncture with the stem. "Neck" . refers to junction of bowl and stem. The "stem" is all that portion of the pipe from the neck to the stem end. If the stem end is expanded or in any way set off from the rest of the stem, it is referred to as a "stem end piece." Some of the expanding or trumpet-shaped bowls may be described as flaring and others as enclosed or tulip-shaped. In Tables XIV aid. XVII, outside bowl diameters are estimated by matching rim fragments with circles : of known diameter. This assumes a symmetry of bowl opening which may not be true of all specimens. The one complete pipe (8317; Plate Vs9) is a short, "tapering cylinder with no noticeable break between bowl and stem portions. It measures 3.8 cm. in length, and Is 1.1 cm. in diameter at the bowl opening and .9 cm. in diameter at the stem end. The hole Is „5 cm. in diameter at the stem end and the inside bowl diameter is .8 cm.. Several unconnected grooves encircle the pipe 2.3 cm. from the bowl opening. The interior of the bowl is darkened, presumably by burned material. No. 5487 (Plate  VslO)  is similar  80  TABLE XXXII BOWL FRAGMENTS FROM UPPER GREI DEPOSIT  Artifact and Plate Nos. 2734(V:1)  Fragment Size (cm)  Est. Dia. (cm)  7.9x5.6  6.0  -  -  Thickness: (cm) .5  .2-.4  ^540 2 -  fv^) ^  3.7x2.7  2.5  .2-.3  9029  2.0x1.7  1.7  .3  Remarks  Longitudinal section of a roughly abraded, wide flaring bowl. Inside of bowl shows some carving and abrasion scratches running from bowl opening to neck. Bowl length is 5.2 cm. Five fragments, all located in S.E. part of house, joined to form most of a flat sided, pentagonal enclosed bowl. Exterior is smoothly ground; interior roughly carved to a circular and not pentagonal shape. Rim edge is ground flat and square to the long axis. Bowl length, 4- cm.;, greatest diameter, 3.4 cm.; bowl opening diameter, 1.7 cm.; greatest interior diameter, 2.3 cm. Longitudinal section of a smoothly finished, enclosed bowl. On exterior are fine abrasion narks, but interior is ground smooth. Bowl length, 3.1 cm., Interior bowl length, 2.7 cm.; bowl opening diameter, 2 cm.; greatest interior diameter, 2.4 cm.. The stem hole is smoothly drilled and .5 cm. in diameter. Complete bowl of a small, flaring pipe with no marked distinction between stem and bowl. Bowl interior is shallow, .6 cm., and flares from a .6 cm. diameter stem hole to 1.2 cm. at opening. Exterior is well finished; interior shows concentric  81  TABLEXVII(continued)  Artifact and Plate -Bos.  Fragment a• oize (cm)  •< Est. Dia. (cm)  Thickness (cm)  Remarks  abrasion or drilling marks from stem hole to bowl opening. Interior of bowl shows traces of burned material. 8497 (V;6)  2.8x2.6  3.0  1273  2.3x1.9  3.0  .2-.3  8738  2.5x2.2  4.0  .1-.15 Rim fragment of thin, well-finished flaring bowl. Inside shows traces of burned listeria 1.  9745  3.2x1.3  5.0  •3-.5  Rim fragment of roughly abraded and carved enclosed bowl.  2341  4.6x1.1  7.0  .3-.5  Rim fragment of a large flaring bowl. Exterior is roughly abraded interior well-finished.  1953  3.2x1.7  5.0  .2-.3  Rim fragment of an enclosed bowl, roughly abraded on exterior but well-finished inside.  2549  3 «4xl.5  .2-.3  Bowl fragment with well-finished exterior and interior.  8570  1.6x1.5  .3  Well-finished bowl fragment with traces of burned material inside bowl.  8168 8169  1.7x1.5  9192  2.2x.8  .3  2.0  Rim fragment of a roughly abraded enclosed bowl. Interior shows carving. Rim fragment of a well-finished enclosed bowl.  Two fragments, well-finished inside and out, which join to form part of a rim.  ©3  Rim fragment, well-finished inside and out but too small for estimation of diameter.  82  TABLEXVII(continued)  Artifact and Plate Nos .  Fragment Size (cm)  Est. (cm)  6500  8715  Thickness (cm)  Dia.  Remarks  «4*" ®5 Bowl fragment with roughly abraded exterior and. well finished interior.  -  2.4x1.8  .4  Neck fragment with smoothly finished bowl interior and .f diameter stem hole. Roughly abraded exterior.  TABLE XV  STEM FRAGMENTS FROM UPPER GREY DEPOSIT Length (cm)  Dia'. (cm)  Hole Dia. (cm)  Remarks  8071  7.6  1.4  .6  (Plate Vsll) Roughly circular cross-section. Smooth, uniform hole diameter®  6884  3.4  1.2  .5  (Plate Vs8) Circular crosssection. Smooth, uniform hole diameter.  7243  7.7  1.8  .8  Hole roughly drilled.  5158 .  2.8  Artifact No.  8994  .8  aa  .9  .4  Longitudinal fragment of roughly abraded exterior, roughly drilled hole. Short section of pipe neck, broken at two encircling grooves« Possibly re-used as a bead ®  83  TABLE XXXII STEM END FRAGMENTS FROM UPPER GREI DEPOSIT Artifact and Plate Nos.  Length . (cm)  Stem Die. (cm)  5083 5084. (V:l6)  10,6  1.2  .6  Two fragments joined to make a complete stem from neck to stem end. The stem is well-finished, circular in cross-section, and at the neck curves smoothly into the bowl. Hole diameters are .8 cm. at stem end, .6 cm. at stem centre, and .5 cm. at neck. The .8 cm. diameter tapers gradually into the .6 cm. The .6 and #5 cm. diameter holes meet 2.5 cm. from the neck. The stem, end piece is a bird head, carved in the round, with well-defined nostrils, beak and oval eyes. The stem enters the top of the head and the stem end is in the base of the head. The downward curved beak projects 1.3 cm. beyond the stem end.  4872 (VsX3)  3.7  1.4  .5  Stem end is plain. Hole diameter at stem end is .7 cm..  9465 (Vs15)  2.1  1.2  .6  Plain stem end. Exterior is roughly abraded.  5487 (VslO)  2.7  1.1  .5  Stem end expands slightly to diameter of 1.2 cm.. Hole diameter at stem end is .8 cm., tapering gradually to smaller diameter of the stem. Stem has broken near a partially-encircling groove, 2.3 cm.from the stem end. Although there is no sign of burned material in the stem end hole, this may be a small pipe bowl, similar to 8317 below.  9390  2.6  1.0  .5  Highly polished, plain stem end.  (V:14)  Hole Dia. (cm)  . • .  Remarks  84.  TABLE XVI (continued)  Artifact . and Plate Bos®  8182  (Vsl2)  Length (cm)  4.5  Stem Dia. (cm)  Hole Dia. (cm)  -  .5  Remarks  This fragment has apparently been carved to shape* It measures 1 cm. in diameter at the stem end, tapers to .8 cm. at the middle of the fragment and then expands to 1 cm. at the "broken end.  to this pipe and may be a bowl rather than a stem end® Five pipe-blank fragments were recovered from the upper grey deposit* Of these, three broke during the drilling operation and the other two during early stages of manufacture. 1. No. 9247 (Plate V:3): Length of blank fragment, 7.1 cm.j length of bowl, 4.3 cm.| diameter of bowl, 4 cm.. The well-finished exterior of this blank indicates an enclosed form for the finished bowl. At the centre of the bowl end is a pecked conical depression .6 cm. in diameter, possibly meant as the starting point for a drill hole. The stem has broken at the end of a drill hole„ 2. No. 4713 (Plate V:2): This appears to be the blank for an enclosed bowl® It is roughly abraded to shape and was drilled from the stem end to .7 cm. beyond the point where the blank later broke. Length, of fragment, 6.7 cm.j diameter of bowl end, 3 cm.j greatest diameter of bowl, 3.6 cm.; diameter of stem at neck, 2.4 cm., diameter of drill hole, .5 cm.. 3. No. 9028s Roughly abraded bowl blank fragment containing portion of drill hole in neck. Length of fragment, 4.2 cm.j diameter of flaring  85  bowl, 3 cm.j length of bowl, 3.2 cm.. 4. No. 5197s Hough blank of a large pipe. Length of blank, ll.A cm.j diameter of bowl, 4.5 cm.j length of bowl, 3.6 cm.. Blank has been sawn and abraded on much of the surface, but bowl appears to have split before any drilling was begun. 5. No. 8912: Longitudinal fragment of a blank. Length of blank, 5 cm.j diameter of bowl about 2.5 cm.. T&.B1M OTI BOML FMQ®iT8 PROlVt LOWER GKEI DEPOSIT Artifact and Plate NoSo  Size (cm)  5466 (IV:2)  4.0x2.0  2403  3.0x1.8  (r i  „ __ „ 3.2x3.7  ,v UV.1J  Est. Bowl Dia. (cm) 4.3  3.7  Thickness (cm)  Remarks  .6-.8  Thick, unfinished rim fragment, flaring bowl, roughly abraded. Rim is ground flat; inner bowl surface shows coarse carved lines spiralling into bowl.  .4- »5  Thick rim fragment, roughly abraded on exterior; smoothly polished interior with carved lines spiralling into centre. Rim ground flat. Enclosed bowl.  .5-.7  Bowl and stem fragment. Interior bowl length, 2.1 cm.j bowl opening diameter, 2.9 cm.j stem diameter at • neck, 2.5 cm.j diameter of stem hole, .7 cm.. Flaring bowlj exterior shows rough abrading, but is smoothed by wear. Interior has concentric circles surrounding stem hole — drilling or abrading marksand-carving cuts.  86  TABLE XVII (continued)  Artifact and Plate Nos.  Size (cm)  Est. Bowl Dia® (cm)  Thickness. (cm)  Remarks  1610 (TVs4-)  3.7x2.2  1930  1.9x1.2  -  .2-.<4  Bowl fragment;; smoothly ground inside and out®  1620  1.9x1.7  -  .2-.3  Slightly curved section, possibly from large bowl. Smoothly ground inside and out.  6973  2.2x1.3  -  .2-.3  Bowl fragment smoothed inside and out. Signs of heat or burning on inside surface.  2247  1.5x1.2  -  .1-.3  Rim fragment of large, incurving bowl. Smooth inside and out; rim ground flat.  2082  1.9x1.8  4..0  .1-.3  Rim fragment with rounded edge; smooth inside and out.  2300  1.9x1.2  -  .15-.25  6226 (IVS3)  1.7x2.0  2.2  .2-.5  .6  Fragment of unfinished bowl base ana stem. Drill diameter, .6 cm.. Exterior roughly abraded; interior shows carving in bowl; smooth, parallel-sided drilling in stem®  Thin, well-finished bowl base and stem fragment. Rounded rim fragment of small bowl. Little flare; interior shows signs of burning. Stem diameter about 1.6 cm®.  87  TABLE m i l STEM FRAGMENTS * FROM. LOWEfc GREY DEPOSIT Artifact No.  Length (cm)  Dia. (cm)  Hole Dia. (cm)  7411  3.5  1.4-  .6  9539  1.4  -  2333  2.8  -  Remarks Roughly abraded exterior with encircling groove at one end where stem has broken. Stem shows wear. Drilling smooth, polished. Roughly abraded exteriorj smoothed interior of about 9 cm. in diameter.  -  Longitudinal sliver from.stem — exterior and interior smoothly finished.  TABLE SIX STEM. END FRAGMENTS FROM LOSER GREY DEPOSIT Artifact and Plate Nos.  Length (cm)  Stem Dia. (cm)  5244(IV:15)  5.7  1.7  .7  Stem end plain, tapering slightly. Outside roughly abraded. Drilling smooth but stepped (as in Plate IVslO). Drilling from stem end and from bowl end meet 4»2 cm. from., stem end.  1974 (IV:13)  2.1  1.1  .7  Stem end bevelled. Smoothly finished throughout. Drill hole shows longitudinal striations.  7477  3.1  .5  Beginning 1.8 cm. from stem end, stem flares to a 1 cm. long stem end piece with rounded diamond-shaped cross-section. Diagonals of diamond are 1.7 and 1.5 cm.. Smoothly  (IV:8)  Hole Dia. (cm)  Remarks  88  TABLEXVII(continued) Artifact and Plate Hos.  Length (cm)  Stem Dia. (cm)  Hole Dia. (cm)  Renerks  drilled hole is countersunk.2 cm. at the stem end. Exterior is well finished. 2080 (IV:16)  6.1  1.5  .6  Stem circumference has been fairly smoothly abraded to nine longitudinal facets. Drill hole is smooth and of even diameter throughout fragment® Stem end piece is 2 cm. longj of circular crosssection; 2.1 cm. at greatest diameter, tapering to 1.2 cm. at stem end. Large diameter curves abruptly but smoothly to stem.  24-02 (IV:17)  7.1  -  -  Stem is thick, lenticular in crosssection, measuring 2.3 cm. x 1.8 cm.. Stem end piece is 2.4 cm. long, of sharp diamond-shaped cross-section with 2.8 cm. and 2.4 cm. diagonals. Widest stem diameter and narrowest stem end piece diagonal are on same plane® End piece sharply set off stem. The .9 cm. diameter stem end hole and 1.1 cm. hole from bowl end meet 4<>6 cm. from stem end. A marked ridge separates the two different diameters. The holes are roughly drilled.  9506 (IV:12)  7.4  1.5 (Square)  .5  Stem end piece is a sharp diamondshape in cross-section, with 2.4 cm. and 2.2 cm. diagonals. .7 cm. from squared end, end piece tapers to a 1.4 cm. constriction which is 1.3 cm. from the stem end and of circular cross-section. This expands to a squared stems at 2.2 cm. from stem end. 4*5 cm. from stem end, the squared edges taper, apparently toward a circular cross-section. A .9 cm. diameter hole at stem end is  89  TABLEXVII(continued)  Artifact and Plate NoSo  Length (cm)  Stem Dia. (cm)  Hole Dia. (cm)  Remarks  drilled 1.8 cm. deep to meet the .5 cm. stem hole. Flat surfaces are coarsely abraded. 5437 (IVs9)  2.1  5368 (IVS14)  1705 (IV:11)  •7  Longitudinal fragment of a roughly squared stem end piece.  2.3  .4  Stem end piece of roughly elliptical cross-section. End piece is 1.6 cm. long. Greatest height (2.2 cm.) and width (2.7 cm.) are at end where stem aid end piece meet. From here, tapers to height of 1.6 cm. and width of 2 cm.. Hole is drilled evenly and is of uniform diameter.  6.1  .7  Longitudinal fragment of what appears to be a blank for a stem and carved stem end piece similar to that shown in Plate Vsl6 (5083-5084, above). It is roughly abraded and has split lengthwise along drill hole.  -  In addition to 1705, above, five fragments of pipe blanks were recovered from the lower grey deposit. 1. No. 9490 (Plate IV:6): A longitudinal fragment of what was apparently to be a cigar-shaped pipe. The length is 6.1 cm.. From 2 cm. diameter at bowl end, the pipe expands to 2.2 cm. at 1.5 cm. from the bowl end and then tapers gradually to 1.7 cm. diameter at 1.2 cm. from the stem: end. From here it flares abruptly to 2 cm. diameter. The fragment  90  has been coarsely abraded and exhibits a semi-circular cross-section® It has been drilled from each end with a .6 cm® diameter drill. The length of hole drilled from the bowl end cannot be determined as it terminated in the other half of the blank®  The hole from the stem end  is 3®5 cm. long and so angled that if continued, it would have emerged from one side of the blank. The drill hole is smooth-walled, of uniform diameter throughout, and is flat at the terminal end® No. 5318 (Plate IV:5): The longitudinal fragment of a conical, tubular pipe. It is 7.5 cm. long and tapers from 2 cm. diameter at the bowl end to about 1.2 cm. diameter at the stem end. The exterior is coarsely abraded to a semi-circular cross-section. One side shows the edge of a saw groove. The blank has been drilled to 1.9 cm® depth from the bowl end. The drill hole is about .6 cm. in diameter, tapering, at the tip, to .3 cm. diameter. Throughout Its length, the hole shows coarse encircling grooves. The entrance of the hole flares to 1.3 cm. diameter. The stem end is not drilled, but there is a .1 cm. diameter, .1 cm. deep hole at the centre of the stem end. This may have been intended as a starting dimple® No. 5201 (Plate IYslO): The longitudinal fragment of a small, unfinished pipe bowl and stem. The length is 2.8 cm.; stem diameter, about 1.5 cm.; bowl diameter, about 1.9 em.. A .7 cm. diameter hole has been drilled from the bowl end in three "steps" (see Plate IV:10) with a square-tipped instrument. The exterior is roughly abraded. Both ends are squared and ground flat.  91  4* No. 2282s A 7 cm. x 5 cm. fragment of a roughly blocked out bowl and part of the stem. Bowl diameter is about A.5 cm. and the bowl length about 4 cm.. 5® No. 1700s A 5 cm. x 4.5 cm. roughly abraded bowl blank siMlar to the above. The bowl diameter is about 5 cm.. Between the pipe assemblages from the two grey deposits, there is a noticeable difference in the distribution of stem end pieces. With the exception of the carved bird head, all six stem ends from the upper grey deposit are plain (Table XVI). In contrast, of nine stem ends in the lower grey deposit, only two are plain (Table IX). The others all have expanded end pieces® In bowl forms, too, there are differences between the two deposits. The upper layer has eight enclosed and four flaring bowls. The lower layer has three flaring and two enclosed® As explained in the introduction to this section, we cannot, on the basis of only one entire specimen, develop a pipe typology. However, in Table XX, a preliminary organisation of the many fragmentary pipes found at Esilao is attempted. Stem cross-sections are, with only three exceptions, circular or nearly circular in cross-section. The three exceptions, all from the lower grey deposit, are faceted (2080), square (9506), or lenticular (2402) in crosssection0 The Canyon ethnographies refer to more recent elbow pipes in discussing smoking. Teit's informants considered the tubular pipes very  92  ancient, while Duff's informants were not aware of their use in the Tait area » TABLE XX CLASSIFICATION OF PIPE FRAGMENTS FROM UPPER AND LOWER GREI DEPOSITS  T  yP®  I.  Conical pipe  II.  Cigar-shane pipe  III.  Trumpet-shape pipe A. Bowls 1. Flaring ' 2. Enclosed  a) Pentagonal B. Stem Ends 1. Plain .  a) Bevelled 2. Stem End Piece a) Zoomorphic carving  Plate and Fig. Ref.  Upper Grey Artifacts  Lower Grey Artifacts  I¥s5 Pig.13a IV:6 Fig.13b  9029  5318  -  9490;  I¥:lj V:1  2734, 8738 2341, 9028  5466, 3198 6226  V:5  2772, 8497 1273, 9245, 1953, 9247 4713  2403 , 2082  V:4  2351 et al  IV:15 Fig.13c  4872, 9465 54^7, 9390,  IV:13 Fig.13d V:l6  5244  8182  -  1974  5083,- 5084  b) With diamond cross-section  IVsl2, 17 Fig.l3e,f,g  -  7477, 2402, 9506  c) With circular or elliptic cross-section  IV:14, 16 Fig.l3h,i  -  2080, 5368  Cto follow Page <?z)  a.  I  c.  O l  "N  h.  FIGURE 13*. PI pa Forms  93  From the relatively high proportion of partially finished blanks, we may conclude that the inhabitants of Esilao manufactured their own pipes at the village. The abundance of worked steatite detritus (see below) would also indicate manufacturing on the spot. From the worked steatite detritus and partially completed specimens, we can suggest the following sequence of operations in the production of a trumpet-shape pipes: 1. A suitable cobble or slab of soeatite was cut into a wide, flat slab by sawing, 2. This slab was then coarsely abraded to the desired thickness somewha t larger than the greatest diameter of the finished pipe bowl, 3. The slab was then sectioned by sawing. The width of the section would again be somewhat larger than the greatest diameter of the finished pipe bowl, U» The section was sawn to a length slightly longer than that of the finished pipe® 5® Through abrading and possibly carving, the exterior shape was roughly blocked out® The bowl end and stem end were abraded flat to produce a level drilling platform® 6. The stem end appears to have been drilled first® A small starting depression was pecked or carved at the centre of the stem end. In drilling the stem, a stone drill was used at first. After the limit of the stone drill was reached, the hole walls were reamed smooth and the hole continued by a drill of uniform diameter and squared  94  cutting tip. The nature of this second drill is not known. From the uniform, flat-ended appearance of some drill holes, it may have been a copper tube or bar of iron. The depth of the stem end drill holes varies, but in no case does it seem to have been carried completely through the pipe. 7. With a stone drill, the bowl end was then drilled, probably no further at first than the neck® 8. It is not known how the bowl was hollcvred out, but the final stages of shaping the interior were apparently done with small, sharp carving tools — possibly of beaver incisors, bone, quartz, crystal or jade — and abraders. The two small ground stone carving chisels described above may have been specialized implements used at this stage of pipe manufacture® 9. The stem hole was then extended from the bowl base to join that drilled from the. other end® 10. The exterior was then shaped with successively finer abraders, the final polish being accomplished with scouring rush (Teit, 1900:182). In use, the pipes probably had a long wooden stem such as that described by Teit (1900:300) for elbow pipes. Several of the Esilao specimens have larger holes at the .stem end than for the rest of the stem. These may have been for the insertion of a wooden stem. Tubular pipes have been found at many Plateau sites. From the McMary Reservoir, Osborne (1957:57-58) describes three cigar-shaped pipes with disc end pieces and two trumpet pipes with plain stem ends. In the Upper  95  Columbia, Collier et al (1924:72-74) found five entire tubular pipes. Of the two cigar-shapes, one has a disc end piece and the other an end piece much like that on Esiiao specimen No. 2080 (Plate IV:l6). The other Upper Columbia forms include a conical pipe and two trumpet pipes, ore enclosed and the other flaring. Both have plain stem ends. Sanger's Chase specimens (1962:32-35) are all trumpet shape and. have plain stem ends. A.t Lytton, Smith (1899:154-155) describes trumpet and cigar shapes, both with end pieces, and a conical pipe. On the coast, Smith recorded trumpet pipes at Port Hammond (1903s 180-181) and on Vancouver Island (1903:181). A single bowl fragment of a large flaring pipe was found in early 1963 at Beach Grove on the Fraser delta (Lee, pers. comm.). In size and shape it is similar to the unfinished Esiiao specimen No. 2734 (Plate V:l), and in finish it also resembles the finer pipe fragments from the Canyon site. On the whole, pipes are rare in coastal sites until early contact times when elbow forms of European manufacture or pattern make their appearance® No specific archaeological parallels can be found for several of the Esiiao specimens, including the pentagonal, enclosed, trumpet-shaped bowl; the zoomorphic stem end piece carving; the stem end pieces with diamondshaped cross—section; and the stems of square and lenticular cross—section® Saws: (Plate VTI:2,3) Nine schist or sandstone saws were recovered from, the upper grey deposit and a single schist one from the lower grey. In addition, a considerable number of ground slate knife fragments from, both  96  layers have smoothly blunted cutting edges. These, too, may have been used as saws. Table XXI summarizes the dimensions of upper grey deposit savs. TABLE XXI  DIMENSIONS'. OF SMS FROM UPPER GREY DEPOSIT Artifact No.  Length (cm)  Width (cm)  2790  13.7 +  7.7  .9  Schist  Cutting edge parallels rock grain  8206  14.2  5.7 +  .7  11  tt  9628  12.2 +  7.9  .9  ri  it  2550  6.7 +  6.7 +  1.1  it  tt  6168.  8.3 +  8.5 +  1.3  it  it  2724  10.1 +  5-9  1.1  tt  tt tt  Thickness (cm)  Material  6822  9.8 .+  4.8 +  1.0  tt  7848  7.8 +  3.8 +  .7  ti  7488  7.6 +  6.9 +  .9  Sandstone  Remarks  Cutting edge at 20° angL to grain Cutting edge parallels rock grain.  The one saw fragment from the lower grey deposit is 8.7 cm. long (incomplete), 5.8 cm. wide and 1 cm. thick. It is made of relatively fine schist and the cutting edge is at a 65° angle to the rock grain. Thompson use of "gritstones" for cutting jade and serpentine is noted by Teit (1900:182) and presumably this refers to stone saws. That sawing was an important stone-working technique in the manufacture of tubular  97  pipes has already been discussed. Further evidence of sawing is found in the worked steatite detritus (see below) where 32 out of 39 pieces of steatite show sawing. The schist of which the saws are made contains varying quantities, of garnets — hard crystals that must have greatly contributed to the cutting power of these implements. Moreever, those saws whose grain slopes into the cutting edge seem to have a particularly coarse bite. Schist or sandstone saws have not been recorded for any Plateau sites although Smith (1900) mentions their use in the Thompson River region and Sanger (1962) found, at Chase, rectangular schist slabs which, although exhibiting no signs of use, may have been saw blanks. The early use of saws in the Fraser Canyon and in the delta region of the river has been reported by Borden (1961:1; 1962:13, Plates 4k,6i; cf. also Smith 1903:167). King (1950:39) suggests that four objects from. Cattle Point were prepared for use as saws. File (Plate VII:l) t One fragment of a thick, ground slate object from, the upper grey deposit has smoothly worn edges. In contrast, the flat faces, although ground, exhibit little wear. The length of the fragment is 6.9 cm.; width, 2.5 cm. tapering slightly to 2.3 em.; and. thickness, a uniform .8 cm.®  The cross—section is rectangular.  Wo ethnographic references to the use of slate files can be found, but such tools are known archaeologically from Stselax (Borden, pers. comm.). The triangular basalt files which Osborne (1957:55-56) records for the McNary Reservoir differ both in shape and material from the Esilao specimen.  98  Abrasive Stones and Whetstones (Plate VII:4,5); A. distinction is made between these two forms of abraders on the basis of coarseness. Whetstones are made of fine grain rock, abrasive stones of coarser grained materials The upper grey deposit contained four whetstones and five abrasive stones; the lower grey, two whetstones. Details of these are given in Tables XXII and XOII. : TABLE m i DIMENSIONS OF ABRADERS FROM. UPPER GREY DEPOSIT Artifact Type No * 9398  Shape  Whetstone Tabular  Thickness Grain Number of (cm) Faces Used  4.7  2.7  7.6  4.5  .6  II  12.6  9.2  1.6  II  Triangular 15.2 (Plate VII:5)  13.7  3.9  it  1  Tabular 20.4 (Plate VII:4)  5.1  1.6  Med.,  2  13.1  10.6  2.7  II  2  Irregular  8.1  6.1  1.1  II  1  II  Irregular  9301  II  II  484-6  N  Abrasive  Width (cm)  22.1 +  8389  6801  Length (cm)  5149  II  9308  it  7397  II  Tabular  9.9-+-  6.8  1.7  8682  II  Irregular  8.0 +  3.1 +  X  Tabular  Fine  Coarse II  1  2 1  1 1  99  TABLE XXXII  DIMENSIONS'. OF WHETSTONES- FROM. LOWER GREY DEPOSIT Artifact No.  Shape  Length (cm)  Width (cm)  Thickness Number of (cm) Faces Used  2401  Irregular  8.6  5.8  1*4-  2  2061  ti  8.5  6.0  .7  1  Abraders are a class of artifacts implied but not specificallymentioned in Canyon ethnographies. The abundance of ground slate knives in both deposits at Esiiao indicates the importance of abraders and abrasive techniques. Many of the bone and antler implements described ethnographically for the Canyon but not preserved in Esiiao deposits must also have been ground and sharpened by abrasive stones and whetstones. Abraders, common in some parts of the Northern Plateau (Sanger, 1962:29j Smith, 1899:144; 1900:14-7), are scarce in far northern (Borden, 1952:32), eastern, and southern Plateau assemblages (Borden, 1956:84} Collier et a l . 1942; Osborne, 1957). They are found as well on the Coast (Borden, 1955:17; Drucker, 1943:57; Smith, 1903:167-168) and at the Dalles (Cressman, 1960:57). Spindle Whorl (Plate VI: 1): One half of a ground siltstone object recovered from the upper grey deposit has been tentatively identified as a fragment of a small spindle whorl. The complete object, if it was circular, measured 8„4 cm. in diameter and had. a 1 cm. diameter perforation drilled near its centre. The hole has been biconically drilled and then routed out  100  to almost perpendicular walls. The object averages .5 cm. in thickness and the rim edges, in some places squared, are in others rounded or pointed. Both the hole and outside rim show wear polish. Teit (1900:191) noted that the spindle of the Lower Thompson Indians was the same as that described by Boas for the Songhish of Vancouver Island except that both the disc and shaft of the Thompson were of wood. Spinning apparatus was an important part of the dog and goat wool weaving complex shared by the Coast Salish and Lower Thompson groups. Fraser, while in Tait territory near Esilao, remarked that the Indians had blankets made from the wool of wild goat and from dog's hair and noted that the dogs had been lately shorn (Lamb, 1960:99)* The whorl from Esilao is probably too small and slight to have been used for dog or goat wool. It may have been used on such light plant fibres as nettle• Spindle whorls are not recorded archae ologically from any Plateau site except Krieger's (1928:12) Wahluke excavations where there are suggestions of both a spindle whorl and the use of dog hair. On the coast, spindle whorls of stone, antler, bone and wood are known from a number of sites (Drucker, 1943:58; King, 1950:49). A decorated rim fragment from a bone spindle whorl and an undecorated antler whorl were found at Stselax (Borden, pers. comm.). In addition, a large, decorated steatite whorl was found in the upper portion of the topsoil zone at Site BjRi 3, about 150 yards north of Esilao village (Borden, pers. comm.). Personal'Ornament Stone Beads: Few stone beads were recovered from either grey deposit at  101  Esilao, but the disproportionate number (seven beads., including two blanks) from the lower layer, when collared with the number (three) from the more extensively excavated upper layer, probably reflects the advent of glass trade beads. Trade beads (see below) are restricted to the upper grey: deposit and the decline in importance of stone beads is very likely related to this® All of the beads listed in Tables Till, and "KM are disc-shaped — flat and circular with a central hole — although there is considerable variation In diameter and thickness0 TABLE H O V  DIMENSIONS OF STONE BEADS FROM UPPER GREl DEPOSIT Artifact Dia. Thickness Dia. of (cm) hole (cm) •No. (cm)  Remarks  9392  1.3  .3  .2  Biconically drilled; irregular shape  2659  .9  .2  .1  Drilled from one side or may have split.  2714  .9  .3  .2  Biconically drilled  TABLE XT? DIMENSIONS'.-OF STONE BEADS FROM..LOWER GREY DEPOSIT Artifact Dia. Thickness Dia. of No. • (cm) (cm) hole (cm)  Remarks  2162  1  «4  •2  Biconically drilled; rounded edges  9157  1.0  .6  .5  Straight drill; may have been part of pipe stem  84-07  .8  •4  .2  Biconically drilled  102  TABLEXVII(continued) Artifact Dia. Thickness, Dia, of hole (cm) (cm) (cm) No. 2266  .8  .3  .3  9224  .7  .2  .15  Remarks Drilled from one side ti  In addition to the beads listed in Table XXV, the lower grey deposit yielded a bead blank, composed of two joined and unfinished beads (Plate 3X:14) • The blank is 1.3 cm. long, .7 cm. wide and .4 cm. thick. An encircling, sawn groove divides the length into two beads, one of .7 cm. length and. the other .55 cm. Evidence of similar encircling grooves' is found at both ends of the blank where other beads have been broken away. Each of the joined beads has been biconically drilled with a .1 cm. hole® From the above information we can reconstruct at least part of the bead manufacturing process. Long flat bars of steatite were apparently first ground to a fairly uniform width and thickness. Encircling grooves were sawn into the bar to form the individual bead blanks and each bead was drilled. The bar of beads was then broken up into separate, square beads ready for rounding. Stone beads are not recorded ethnographically for the Canyon, although Teit (1900:222) mentions and illustrates horn, bone or shell beads of the same disc shape from Thompson territory. The last two beads listed in Table XXIV and the last three in Table XKV (one Is illustrated in Plate 3X:13) are similar in general size and shape to the stone disc beads so abundant at some Coast sites. At Marpole, for example, in addition to the 8000 artifacts there was a cache  103  of over 20,000 shale disc beads near the western extremity of the site (Borden, 1950:19; pers. comm.)*' They are also known from the top soil zone at DjRi 3 in the Fraser Canyon (Borden, 1961:1), from the McNary Reservoir (Osborne, 1957:33) and from the Dalles (Cressman, 1960:57)® Stone Pendants: No stone pendants were recovered from the lower grey deposit® Of the four from the upper grey deposit, three were suspended from perforations and one from a girdling groove near the top® 1® No. 2584 (Plate 3X:6); This is the broadly-rounded top fragment of what appears to have been a wide, triangular shaped pendant. The fragment, which measures  U cm. wide, 3.2 cm. long and is .5 cm® thick,  includes a biconically drilled perforation centred .5 cm, from the top. The perforation is .2 cm. in diameter where the drilled cones meet. The edges are rounded and both faces decorated. On one face are nine incised oblique lines, four to the left of the hole and five to the right, converging towards the hole. On the obverse face are one long and two short lines to the left of the hole and a single short line to the right, converging in the same manner as those on the other face. As the lightly incised design on this face seems to be partly worn away, this may have been the back of the pendant. 2. No. 7367 (Plate JLiA) s This pendant has a broadly rounded top similar to that of the preceding one, but the complete shape is that of a semicircle rather than of a triangle. The pendant is complete and measures 1.9 cm. long, 2.6 cm. wide and. is .5 cm. thick at the base, tapering to .2 cm. at the top. The perforation is ®3 cm. in diameter,  104  biconically drilled and centred .6 cm. from the top. Neither face is decorated. 3. No. 5418 (Plate 3X:3): Shaped like a narrow, elongated form of the above, this pendant, when found, had a rounded top coupleting the biconically-drllled perforation. The pendant is 2.3 cm. long, 1.3 cm. wide, and .3 cm. thick. The .2 cm. diameter perforation is centred about .4 cm. from the top* 4. No. 8533 (Plate IX:l): If this highly decorated steatite object was a pendant, it must have been suspended from an encircling groove near the top. The length is 4.3 cm.j width at the base 1.1 cm. and at the tcp .7 cm.j thickness .7 cm. at the base and .3 cm. at the top. It is roughly rectangular in cross-section. Six sawn grooves notch each side, but except for the top and bottom two, these are staggered rather than paired. Below the suspension knob, one face is divided into three approximately equal fields by four sawn grooves. The top field contains three incised lines running down obliquely from the left and one line angling down from the right to cross these. The middle field has three oblique grooves angling down from the left. Of these, the top two are continuations of lines started in the upper field® The bottom field has three oblique lines running from upper right to lower left® The opposite face Is not divided into fields but has four oblique grooves running from the left and four from the right. These lines.  105  are not symmetrically spaced but seem to run from the side notches on left and right to notches on the opposite side. The face with the three fields shows the greatest amount of wear polish. Ho specific references to the use of stone pendants are made in Canyon ethnographic sources although Teit (1900:223) refers to the use of large sheet copper pendants. Possibly these metal forms had replaced stone by the time the first Plateau ethnographies were compiled. No. 5418 has a parallel on the coast in a snail steatite pendant found at Stselax village (Borden, pers. comm.) and is similar in outline and size to a double-holed pendant from McNary Reservoir (Osborne, 1957:57).  Miscellaneous Objects Hammers tones (Plate ¥111:3,4-) J  These crude implements have such a wide  temporal and spatial distribution that they are seldom considered diagnostic features. The greater number recovered from the upper grey than from the lower grey deposit may reflect the more extensive excavation of the former. Ethnographic reference to the use of hammerstones is lacking for the Canyon. Any of the forms shown in Tables XX¥I and XXVII could have served a variety of purposes, but the two flattened and pointed jadeite cobbles from the upper grey layer (No. 6156, 2725J Plate ¥111:4) may have had a special use. Both show a concentration of battering about the pointed end and it is suggested that these hammerstones were used to blank out ground slate artifacts. Hammerstones are reported archaeologically from Chase (Sanger, 1962:26), McNary Reservoir (Osborne, 1957:46-47), the Upper Columbia (Collier et al,  106  1942:69-70), the Kootenays (Borden, 1956:83), the Dalles (Cressman, 1960:54), Cattle Point (King, 1950: 35-36) and the Northwest Coast (Drucker, 1943:50). Hand Maul (Plate VIIIslj Fig.14): A basal fragment of a well-shaped and polished hand maul was found in the lower grey deposit. The maul had split along its main axis and again across the shaft 7.3 cm. from the striking face. Although the shaft was circular in cross-section, measuring 5.3 cm. in  Figure 14: Hand maul fragment diameter at the top of the fragment, the flared striking head appears to have been rectanguloid, with the one measurable diagonal 7.7 cm. The striking face is convex and shows little use® In Its possession of a flared striking head, this maul is similar to  107  -3MEB.OT  DIMENSIONS OF H&MMIRSTONES FROM UPPER GREY DEPOSIT  Artifact No. 6156  Shape  Flattened cobble  Length . (cm)  Width Thick(cm) ness (cm)  Remarks  9.4  9*3  4.2 Jadeite, shows use around entire circumference, particularly at point.  2725  "  12.1  9.2  2.1 Jadeite, battered on point  7676  »  15.2  8.6  3.4 fettered ends, sides; also used as anvil-2 pits on one face, one on other®  8792  Small cobble  8.5  3.9  1.2 fettered ends  8881  »  6.7  4.4  2.1 Tapers to battered, pointed end.  and on outer rim at top.  2802  Oval cobble  14  8.7  7  6888  Elongated cobble  17.2  7.3  4.3 Battered one end, sides and pitted both faces  5U9  Elongated  14*0  6.4  3.6 Battered both ends, sidesj pitted both faces  Battered on both ends, and on centre each face  8015  "  15.0  5.6  4.6 fettered both ends, one face  9481  »  17.7  5.1  4.2 Battered ends, one face near middle j rectangular .crosssection  7676  »  8711  Piriform boulderchip  7.4-+- 4.9 13.9  10.3  3.4 Broken; end battered 3.5 Narrow end battered  108  the general forms used by the Thompson. However, it differs in detail from the specific mauls illustrated by Teit (1900:183, Figs. 120,121). ... TABLE XXVII DIMENSIONS OFffiMMERSTONESFROM. LOWER GREI DEPOSIT . Artifact No.  Shape  Length Width Thick(cm) (cm) ness (cm)  Reaarks  2175  Prismatic cobble  9»2  6.5  5-9 Battered both ends and at edges of prism  9320  Flattened cobble  13.3  6.2  2.2 Battered both ends, sides  In hand mauls and adzes, the. Esilao villagers had two of the three main implements of the coast wood-working industry. The third tool, the wedge, was usually made of wood or antler - materials not preserved in Esilao deposits. But antler wedges are recorded for the Thoiqpson (Teit 1900:183) and are known archaeologically from near—surface deposits at Site DjRi 3 (Borden, pers. comm.) . There is also abundant ethnographic evidence of the importance of wood-working in the Canyon, from the cutting of pit house timbers to the construction of split-plank houses and carving of canoes and grave figures® Archaeologically, hand, mauls have a wide distribution, on the Plateau and Northwest Coast (Borden, 1956:83; pers. comm; Collier et_al, 1942:69; Cressman, 1960:54; Drucker 1943:50; Osborne, 1957:54; Smith, 1903:156). Specific parallels to the Esilao striking head fragment are found on the Fraser delta at Marpole and Port Hammond (Smith, 1900:156) and Stselax (Borden, pers. comm.) and on the Plateau in the Upper Columbia (Collier  109  et a l . 1942:69) and at Lytton (Smith, 1899: figs.22-27). Stone Vessel (Plate VIII:2): Part of a decorated stone vessel was found in the upper grey deposit. The hollow is pecked and ground into the surface at one end of a 12.8 cm. long by 8.8 cm. wide rectanguloid block of  Figure 15: Reconstruction of stone vessel chlorite schist. The present diameter of the hollow is 6.3 cm. and the depth, 1.9 cm.. The edge of the depression is surrounded for half of its circumference by a 1.4 cm. wide lip, raised .3 cm. from the flat surface  110  of the rock slab. On this flat upper surface are incised five lines. Four of these, parallel and spaced 1.3 cm. apart, run from the bowl rim to the end of the rock slab® The fifth crosses the others at right angles approximately .9 cm. from the rim's outer edge. If the rim is held level, it becomes apparent that part of the bowl is missing, and that the original hollow depth must have been close to 2.4 cm.. The complete vessel may also have been symmetrical, In which case its total length would have been about 16 cm. and Its height, at the centre, A.8 cm.. The bowl would then have been flanked by two decorated wings (Fig.15). Black stains and wear polish found on the fractural surfaces as well as in the bowl indicate the vessel has been used since it was broken. In several places, the inside surface of the bowl is ground smooth. Within the base is a rough, possibly natural, angular depression .4 cm. by 3.0 cm® wide and 1 cm. deep. Several uses for stone vessels were noted in our ethnographic study. The nature of the dark stains has not been determined, but from the worn inner surface of the bowl, this vessel seems to have been used in a grinding operation — possibly in the production of pigment or tobacco. Mortars are never very common in Plateau sites. They are reported from Kamloops (Smith, 1900:413), Chase (Sanger, 1962:27), Lytton (Smith, 1899s 139), the Upper Columbia (Collier et al, 1942:75), the McNary Reservoir (Osborne 1957:63) and the Dalles (Cressman, 1960:56-57). On the coast, excavations at Marpole have produced several bowls, both decorated and plain (Borden, 1950:14,19? 1955:18; Smith, 1903:184-186), and Drucker (1943:54-55)  Ill  refers to their abundance on the Northwest Coast® At none of the above sites was there a vessel significantly similar to the Esiiao specimen® Worked Steatite (Plate Vis 3): Grouped under this heading are 38 pieces of steatite, all of which shov? some form of working® Hone were clearly recognizable as partially completed objects, and most are probably detritus® They are of interest both because they indicate that such steatite objects as pipes were actually made at Esiiao and because they disclose some of the stone working techniques® Table XXVIII summarizes various techniques exhibited by the 38 specimens. TABLE XXVIII TECHNIQUES OF WORKING STEATITE Technique  Sawn from L, faces, broken  Upper Grey Deposit  Lower Grey Deposit  1  Sawn from opposite faces, broken  17  3  Sawn from 1 face  10  3  13  2  1  1  incised Shaped by abrading Carved  112  The one specimen sawn from four faces has been cut and broken from the end of an elongated piece of steatite. This cutting technique nayhave been used for sawing pipe blanks to length® The most common technique is to saw from opposite faces (Plate VI:3) until the two cuts are less than 1 cm. apart. The part to be removed is then broken off. Fragments with cuts sawn from one face only may indicate a different technique or may merely result from breakage of the more common two-cut sawing procedure®  There is some evidence in favor of this latter con-  clusion. The four specimens with both a single saw cut and abrading show grinding only on the face lying at the top of the cut. The opposite face is invariably rough. Eleven of the sawn specimens were first abraded®  Large pieces of  steatite may have been sawn and coarsely abraded to one desired thickness and then sawn to width. The final blocking out operation would be to saw the piece to length. All of the saw cuts are wide and apparently produced by thick, coarsegrained tools similar to the schist and sandstone saws described earlier. Incised grooves have been cut into the surfaces of four pieces. Of these, two also show abrading, one, abrading and. sawing, and one shows incising alone. The grooves have been cut with a stoutly pointed implement but no purpose for the incisions could be ascertained. In some cases, they possibly represent starting grooves for a sawn cut. The one specimen showing carving was also abraded and incised. The  113  carving marks are essentially the same as those inside unfinished pipe bowls® Ochres Fragments of red and yellow ochre were recovered from both upper and lower grey deposits® In the house floor area of the upper deposit, ochre flecks were particularly abundant about the central hearth® The use of ochre for ceremonial or decorative purposes is a widely distributed trait® Teit refers to the decorative painting of face and body in various colors, Including red® Problematical Ob.jectss Eleven objects, all of steatite or chlorite schist, are reported below® Eight came from the upper grey deposit and three from the lower grey® From Upper Grey Deposits  1® No® 1243: Possibly a pendant, this carved and sawn object has an encircling groove 1.1 cm. from its widest end® The length is 7.3 cm., width 1.9 cm. tapering to 1.1 cm., and the thickness «5 cm. tapering to .3 cm.. The thickest end is also the widest. One side is straight; the other convex. 2. No® 8691 (Plate IX:5): This is a surface fragment of a decorated and drilled object. The fragment is irregular in shape and measures about 2.5 cm. x 2.1 cm. x .3 cm.. Four deeply-carved grooves cross the ground surface at various angles. A .1 cm. conical perforation is centred on the object and drilled from the decorated side.  114  3. No. 1199 (Plate 1X82): Possibly a pendant with the suspension knob broken off, the object is tear-drop in shape. Its crosssection is that of a rounded rectangle. The length is now 3.8 cm., width 1.5 cm. tapering to .7 cm., and the thickness 1.1 cm. tapering to .4 cm. An encircling groove is sawn about .4 cm. above the base and on one face two more incisions parallel this at .3 and .8 cm. above. A sawn line from the other face spirals up around the side. 4. No. 2602: Rectangular and coarsely ground to shape, the artifact measures 3.5 cm. x 2.1 cm. x .7 cm.. 5. No. 8477: This specimen is roughly rectangular in shape. The four long faces are slightly convex. One end is sawn and abraded smooth, the other sawn from all four sides to a depth of .4 cm. and then broken. The length is 3.8 cm., width 1.9 cm., and thickness 1.9 cm.. 6. No. 9309: A cylindrical object with top and bottom polished, the sides are smooth but angular. One diameter (2.4 cm.) is slightly larger than the other (2.1 cm.). The larger diameter has a raised, polished disc of 2.1 cm. diameter on its surface. 7. No. 1618: This is a rectangular object drilled with a .2 cm. diameter asymmetrically biconical perforation towards one end. The dimensions are 1.7 cm. x .9 cm. x .5 cm.. 8. No. 6375: A cylinder 6.8 cm. long and 2.1 cm. in diameter, the specimen is roughly abraded to shape. On one side is a 2.1 cm. long, .2 cm. deep saw mark paralleling the long axis. One end has a 1„3 cm. diameter, and a depression .3 cm. deep. This end shows much wear polish both in the hollow and on the rim®  115  From Lower Grey Deposits 1. No. 5354 (Piste IXsl5): This is a well-made, ground and perforated disc, 3.2 cm. in diameter and .4 cm. thick. The hole is biconically drilled.  2. Mo. 7135 (Plate Hsl6) s This carved object may be a pendant. It is 4.5 cm. long, 1.7 cm. wide, and .6 cm. thick tapering to .15 cm. at the possible suspension end. At this end is a .2 cm. diameter asymmetrically biconical perforation. The base expands into a circular portion, in one side of which there is an asymmetrically drilled biconical perforation of .1 cm. diameter. Approximately .6 cm® above the circular base are two asymmetrical lugs projecting from each side. The carving may be zoomorphic, in which case the lower perforation would represent an eye and the lugs either fins or feet. 3. No. 5468 (Plate IXs12): This is a fragment of an object carved in the round. It is shaped roughly like a bird's body with the portion where the head would be, broken off. Parallel incised lines on both sides may represent wing feathers. Similar lines on a projecting but partly broken tail, may also depict feathering. If the carving does represent a bird, then the legs are brought forward under the body in an unusual position. Wood Industry Bark Rolls; These cylinders of thin, papery bark are often charred at one end and are generally tightly rolled. The rolls are probably made of  116  birchbark. Birch trees are common at the site and their bark is a remarkably durable material. Tables XXLK. and XXL summarize the dimensions and appearance of these artifacts® Ethnographic sources do not mention bark torches. Teit (1900:252) refers to an Upper Thompson harpooning technique involving the use of torches but this does not explain the presence of bark torches in the Canyon, where harpooning was less important, or their relative abundance at the habitation site® TABLE 2X3X  DIMENSION OF M M ROLLS FROM UPPER GREY DEPOSIT Artifact No®  Length (cm)  1206  1.7  7727  —  Dia® (cm)  Charred  Not Charred  ®5  X  -  X  Remarks  Several folded and flattened. fragments  No explanation can be offered for the marked difference In distribution between the two deposits. No bark rolls clearly showing signs of burning were found in the upper grey deposit. Ar chaeologically, bark rolls are known from a number of sites on the northern part of the Northern Plateau (Sanger, 1962:67; Smith 1899: 160 ff) and from a recent Carrier site in the northern interior (Borden, 1952:33)®  They are also found in the Canyon in the topsoll zone  at site DjRi 3 (Borden, pers, comm.) and occur in strata underlying the pit house at Ssilao.  117  TABLEXXXII(continued) DIMENSIONS OF BARK ROUS FROM LOMER GREY DEPOSIT Artifact No.  Length (cm)  Dia . (cm)  Charred  Not Charred.  Remarks  2046  10.1  1.7  A.  TT  Tightly rolled  2105  7.3  1.9  X  Tightly rolled  2100  11.3  1.5  X  2159  5.1  1.1  X  Loosely rolled  2101  3.0  .7  X  Tightly rolled  2117  5.0  1.4  X  Loosely rolled  2436  2.7  1.1  X  9142  1.2  1.4  X  Tightly rolled  2435  3.0  .8  X  Loosely rolled  2253  5.7  .9  X  !1  tt  tt  Wooden Haft; Fragments of wood adhering to an iron point (Plate X:2) are described below in the discussion of iron trade goods. Trade Goods Classed as trade goods are objects of Asian, European, or European-American manufacture. Trade goods at Esilao include numerous glass beads, and objects of copper, brass and iron.  US  Glass Glass Beads; The upper grey deposit yielded 120 glass beads. Hone were found in the lower deposit. Several shapes and sizes were represented in the bead assemblage but the greatest number fall into either of two main classifications: large (.8-.9 cm. diameter), light blue spherical beads (Plate 3X:7); and small (.35-»5 cm. diameter), light blue spherical beads (Plate 1X59). Both the large and small types show considerable variation in shape, ranging from forms flattened at the holes, through spherical to almost oval form® In Table XXXI, further distinctions are made. Each bead, subjected to a small, intense source of light, was reclassified on the basis of light transmission. This test revealed that the small beads, externally TABLE XXXI .DIMENSIONS Al® QUANTITIES OF SPHERICAL BLUE GLASS TRADE BEADS .... Light Blue Opaque Does Not Transmit Light  Transmits Light  IS  2  Medium (.6-.7 cm.)  -  -  Small (.35-.5 cm.)  34  65  Large (.8-.9 cm.)  Colour  Dark Blue Translucent  8  •  119  Identical, were actually of two distinct types. Those which transmitted light had a blue-green translucent appearance and the interior of the bead contained many minute air bubbles. The significance of this distinction is not understood, but it probably reflects differences in manufacture and possibly in origin. Table XXXI also includes a small sample of eight dark blue translucent beads (Plate IX:10). These beads are noticeably translucent without the aid of an artificial light source. Descriptions and dimensions of the miscellaneous beads from Esiiao are presented in Table XXXII. The most striking characteristic shared by all beads from the Esiiao assemblage is that all are blue. Various shades ranging from light to dark are represented, but no beads of any other colour were recovered. Blue beads show a wide distribution in Western North America from the time of early contact. Cook (1785:II 357) observed that the natives of southeast Alaska had sky-blue beads, upon which they set a high value. Strange (1928:36), in the same area in 1786, also noted the rejection of all beads except sky-blue. In 1808, Fraser found great quantities of blue beads in the possession of the first Thompson Indians he met (Lamb, 1960:84). On the Lower Columbia, Townsend, in the mid 1830's, noted that blue glass trade beads were a standard of value (Quoted in Osborne, 1957:95). More recently, Teit (1900:261) mentions blue beads in his list of Thompson Indian commodities of trade. Archaeologically, blue beads are fairly common at some Plateau sites.  120  TABLE XXXII DIMENSIONS OF MISCELLANEOUS- GLASS TRADE BEADS  Artifact No.  Shape  Dia. (cm)  Length (cm)  .5  .6  Lt. Blue Opaque  Plate 11:11  .6  Lt. Blue »  Plate IXsll  Colour  Remarks  4686  Cylindrical  6515  it  8745  ti  .5  .6  Lt. Blue Translucent  Full of bubbles  9033  t!  .5  .5  Lt. Blue Translucent  11  .3  .3  Lt. Blue Translucent  ti II  5198  H  5177  ir  .3  .2  Lt. Blue Translucent  54-50  11  .2  .2  Lt. Blue Opaque  6859  Oval  .7  1.2  Dk. Blue Translucent  Full of bubbles  .7  1.0  Med. Blue Translucent  Plate IXS8 Fragment; hole dia ® .2 cm. Flattened at ends  2751  IT  4-671  Spherical  X aX  1.0  Lt. Blue Opaque  6852  ir  .8  .6  Bright Lt. Blue Translucent  6583  ti  .7  .6  Med® Blue Translucent  7246  it  .6  •5  Med. Blue Opaque  8772  1!  •4  .3  Bright Med. Blue Translucent  Less Translucent than No. 2751  121  TABLE XXXII (continued)  Artifact No.  Shape  25AS  Spherical  9820  »  Dia• (cm)  Length (cm) .2  •2  Colour  Remarks  Bright Med. Blue Translucent Lt. Blue Translucent  Of type commonly known as seed bead  In the McNary Reservoir, Osborne (1957:98) records that of 642 beads, about 75 percent were blue and the remainder largely white. Collier et al (1942:104) found in one burial on the Upper Columbia, 3350 blue and 2800 white beads. Of twelve spherical glass trade beads found at the mouth of the Fraser at Stselax (Foster, n.d.:15), nine are dark blue and one light blue. In coastal southeast Alaska, de Lagum (1956:211) has found pale blue glass beads of two sizes. Specific comparisons of the Esilao beads with specimens from other archaeological sites are difficult. Detailed descriptions of beads are lacking, and even such terms as "light-blue" or "dark-blue" are relative to the observer and the assemblage he is studying. However, we can say that the blue beads from Esilao differ from those found by de La guns in southeast Alaska (de Lagum, pers. comm.) and that they differ also from the Stselax beads. The Stselax , assemblage includes beads of other colours and shapes, including the large faceted variety traded by early Hudson's Bay Company posts in the area. Beads were used as clothing decoration and personal adornment by  122  both Canyon groups. They are illustrated by Teit (1900) in Figures 195, 196, 199? and 200, although no colours are given® Copper The uniform thickness of large copper objects discussed below indicates that most if not all copper at Esilao was of European or European-American manufacture® Tube Beads: Tables XXXIII and XXXIV summarize the dimensions of rolled copper tubes from the upper and lower grey deposits. The greater number of specimens from the upper deposit probably reflects a genuine increase in trade goods and not merely better preservation of the metal® Tubes found in the lower grey deposit were no more corroded than many from the upper layer. As we observed in our ethnographic study, Fraser (Lamb, 1960:100) saw brass or copper tubes "hanging from the neck, or across the shoulders" of the natives at Yale. Teit (1900:222) speaks of copper tubes attached to buckskin shirt or belt fringes. One tube from Esilao (8414; Plate X:3) enclosed a .3 cm. diameter hide thong, further evidence that buckskin clothing may have been more important in the lower Canyon than the ethnographies would suggest. Possibly those tubes of uniform diameter throughout were used as necklace beads, while those exhibiting marked taper or flattening at one end (1986, 6844, 2212, 8728, 2801, 6494) may have been fringe decoration® Copper tubes are known archaeologically from a number of Plateau  123  IABIE m i l l DIMENSIONS OF COPPER TUBES FROM: UPPER GREI DEPOSIT  Plate No.  Length (cm)  Outside Diameter (cm)  1085  X:4  9.0  .7  Joint butted, not overlapping  S/JL4  Xs3  8.7 +  .7  .3 cm. dia. thong runs full lengi  1986  Xs7  4.0 +  .7  Tube curves and. tapers to ®5 cm. dia „  4«4*  .5  Slightly flattened to ®4 cm.  3.8 +  .6  Roughly shapedj triangular crosssection at one end  1294  3.4  .6  Flattened to ®3 cm. at one end  8466  2.8  .45 .7  Tapers to «4 cm. dia. at one end  Artifact No®  2058 6844  Xsl3  Remarks  2212  Xsl4  3.4  8821  Xs6  3.8 +  8728  Xs9  3.0  .5  Tapers to ®4 cm. dia. at one end  8828  3.1  .7  Flattened for over half length  5172  2.14-  .8  Fragment  2.3  .5  Carefully shaped; lapped edge straight  8102  2.1 H-  .7  2801  2.3  .6  Tapers to .5 cm. dia» at one end, .4 cm. dia. at other end  6495  2.1 -h  .45  Tapers to .35 cm. at one end  7705  1.9 -h  .45  Fragment  9312  Xs8  124  TABLEXXXII(continued) •DIMENSIONS "OF COPPER TUBES FROM' LOWER GREY DEPOSIT Artifact No.  Length (cm)  Outside Diameter (cm)  2134  4.6  7  Bark fibre adhering; ends corroded  2170  1.9  •35  Ends corroded  Remarks  sites including the McNary Reservoir (Osborne, 1956:96-97), the Upper Columbia (Collier et al, 1942:102), Yakima (Smith, 1910:88-89), and Wahluke (Srieger, 1928:13). Sanger (1962:70) suggests that two creased pieces of copper foil at Chase may have been opened tubes. A single copper tube bead was found on the coast at Stselax (Foster, n.d.:17). Ring (Plate X:10): A 4.6 cm. long flat band of .7 cm. wide sheet copper, curved to form an oval ring of 1.5 and 1.0 cm. diameters, was found in the upper grey deposit. The rounded ends overlap .7 cm.. In its present shape, this object is too small to have been a finger ring. However, if it once had little or no overlap it may have been a finger ring, Teit (1900:223) mentions the use of metal finger rings by Thompson Indians. They are known archaeologically from the McNary Reservoir (Osborne, 1957:248). Bracelet (Plate Xsll) s. This badly corroded object is formed from a tend of folded copper varying from .8 cm. to 1.4 cm, in width. A sheet of copper has been folded into four layers, the edges hammered and the whole  125  bent into an incomplete oval of 4»1 and 3.7 cm. diameters. Teit (1900s223) refers to anklets and bracelets of copper, and Fraser (Lamb, 1960:100) spoke of "large brass wire" bracelets on the Indians at lale» Three wide copper bracelets were found in Upper Columbia burials (Collier et al, 194.2:103). Stains on some Chase burial remains indicated where copper bracelets and anklets had been® Jingle (Plate X:2l): The long sides of a sheet copper rectangle have been rolled together to form a partially closed cone. The jingle is 3.3 cm. longj .5 cm. in diameter at the enclosed end spreading to 1,6 cm, wide at the opened end. Within the enclosed portion is the knotted end of a suspension thong. The object was recovered from the upper grey deposit. No specific ethnographic reference is made to copper jingles. From the presence of the buckskin suspension thong, we may speculate that this was a variation of the copper tube fringe decorations. Two similar objects have been found at coast sites. Smith (1907: 308) reports one from Comox on Vancouver Island and one was recovered at Stselax (Foster, n.d.slS). Pendants: The four copper pendants described below were recovered from, the upper grey deposit® 1. No. 5277 (Plate X:24): A. roughly triangular pendant of thin sheet copper, this object is 7.7 cm. long, 5*9 cm. wide at the base tapering to about 2 cm. wide at the top. The base is straight, but the edges  126  Figure 16s Copper Pendants are excurvate and the top rounded. A .2 cm. x .3 cm. perforation is centered .8 cm. from the top. The hole has either been drilled, or, if punched, the excess metal has been removed through careful grinding. When found, the buckskin suspension thong was still attached. The hitch used was'that commonly known as a lark's head (Kephart, 1957s 11,280). The bight of the suspension thong is passed through the hole and the ends then drawn through the loop. For the .3 cm. diameter thong used, this:form of attachment required either a large or elongated hole. No. 8997 (Plate X:20)s A roughly squared circle in outline, this )  pendant is deeply dished at the centre. It measures 5.5 cm. between  127  the flattened sides and 5»4- cm. from flattened top to bottom. Each diagonal is 5-8 cm®. The sides rise .9 cm., and the top and bottom .6 cm., above the dished centre. The circular perforation, centred .4 cm. from, the top and measuring .25 cm. in diameter, has been formed by punching through from the concave side. The resulting ridge of metal on the opposite side has been smoothed but not entirely removed through grinding. Four roughly concentric circles of dots have been punched into the concave side. ' These occur at approximately .6, 1.2, 2.3 and 2.6 cm. from the centre, measuring' towards the right side. While the two inner circles are roughly circular, the outer pair conform more to the pendant outline. The inner circle consists of 11 dots; the next of 18; the next of 24; and the outer circle of 28. These dots are about .1 cm. in diameter, and somewhat angular in shape. They are about .05 cm. deep and raise a correspondingly high protrusion on the opposite, convex side of the pendant. The copper is of a uniform thickness and is made of material thicker than pendant No. 5277. No. 6911 (Plate X:23): Of a similar outline to the above No. 8997, this thin sheet copper pendant measures 2.1 cm. between the flattened sides and 1.9 cm. between the top and bottom. The circular .15 cm. diameter perforation is centred .2. cm. from the top. It has been formed in the same manner as the perforation in No. 8997, and exhibits a similar ground-down lip.  128  4® No. 5193 (Plate X:22): Now an asymmetrical, truncated triangle, this pendant nay once have been almost square in form. It is 2.2 cm. long with the base measuring 2.2 cm. and the top 1.5 cm. wide. The horizontally elongated perforation is .4 cm. from the top and .1 cm. from one side. This opening has been formed by punching through a. .5 x .15 cm. hole with a wide, flat perforator. A flap of copper from, the hole has been bent up and flattened against the back of the pendant. A .2 cm. deep strip at the top of the pendant has been bent down and pressed against the back. The elongated perforation suggests that the suspension thong may have been attached as described above for No. 5277. According to Teit (1900:222) the Thompson wore sheet copper ornaments suspended from the helix of the ear. They also used large (three-inches square) sheets of copper as necklace pendants• Copper pendants are recorded for the McNary Reservoir (Osborne, 1957:97-98), the Upper Columbia (Collier et_al, 1942:102,104), Yakima (Smith, 1910:92), Wahluke (Krieger, 1928:13), the East Kootenays (Borden, 1956:96), and Chase (Sanger, 1962:70-71). Sheets: Eight small, roughly rectangular pieces of copper were found in the upper grey deposit. None were intentionally perforated but some were well-finished. Table XXXV summarizes the dimensions of these objects.  129  :  TABLE S U  DIMENSIONS OF SHEET COPPER PIECES FROM UPPER GREY DEPOSIT Artifact No.  Length . (cm)  Width (cm)  2527  3.6  1.9  2315  2.3  1.2-1.5  6667  4.0  .9  2609  2.6  .3-1.2  2625  3.1  1.1  2775  3.7  .2-.5  9027  2.5  1.1  2654  3.2  1.0-1.2  Remarks Roughly rectangular; thickness -varies thicker at one side and end; edges rough (Plate Xsl9) Width tapers to one end; very thick copper (Plate Xsl5) Curves vp sharply at one end; edges rough (Plate X:17) Tapers to one end. All edges slightly excurvate and smooth; uniform thickness Smooth edges Edges rough; tapers to rounded point at one end; twisted about 30° around long axis; rough edges Thin (Plate X:l6) (Plate X:18)  Brass Pendant (Plate X:12): Shaped roughly like a stemmed triangular point of SBb xorm, this possible pendant Is 9.5 cm. long and at its greatest width is 3.0 cm.. The concave sides taper to 1.0 cm. width at the top. The length of the stem is 1.1 cm. and its width, .7 cm.. The thickness of the object is a uniform .35 cm. throughout. There are smooth, rounded edges  130  on the blade, stem, base of stem and on one shoulder, but the other shoulder and point are roughly broken. The point or top has also been notched on either side about .4 cm. from the broken end, possibly for a suspension cord. Running from each corner notch and converging at the top notches are two roughly V-shaped grooves of varying depth and width. These grooves may be decoration or an unfinished attempt to section the metal for some other use. The other face is smooth and und.ecorated. Mo ethnographic or archaeological parallels for this object are known. It was found in the upper grey deposit. Iron Points: Two iron points were recovered from the upper grey deposito 1. Ho. 2760 (Plate X:l): This is a triangular point of NBc form with excurvate edges. The base is rough, perhaps indicating the former presence of a stem. The length is 3.5 cm.; width 1.6 cm.; thickness about .15 cm.. Covered with a heavy layer of corrosion, the point was found in post hole B. 2. No. 1030 (Plate X:2): A leaf-shape point of M.h2 form with the tip. broken, the length is approximately 5.6 cm.; width 1.4 cm.; basal width 1.3 cm.; blade thickness .1 cm.; basal thickness .15 cm.. Two fragments of wood, each 2.1 cm. long, are adhering to the basal half of each face of the point. A fragment on one face measures 1 cm. between two parallel sides. This may Indicate roughly the diameter of the arrow shaft. Beginning about .6 cm. above the base and extending a further 1.3 cm. are. the remains of what may have been binding material.  131  Teit (1900:241) refers to the replacement of stone arrow points byiron and illustrates (Fig. 222, d,e) two points apparently made of iron or copper® Archaeologically, iron points have been found in the McKary Reservoir (Osborne, 1957:99-104), on the Upper Columbia (Collier et al. 1942:106) and at Chinlac, a Carrier site (Borden, 1952:42, PI.I,36).  SIMM&RY AND. ANALYSIS OF MCHAE0L0GICAL INVESTIGATION  In the following discussion of the archaeological investigation, no attempt will be made to summarize the ethnographic and archaeological "overlapping" which was sought in this study. That problem is more appropriate to the concluding chapter where the results of the two studies are consolidated. Here we will be concerned first with two problems which will not be raised in the following chapter — namely, a comparison of the upper and lower grey deposit assemblages, and dating of the pit house occupation — and secondly, with a resume of the inter-areal comparisons made on the preceding pages® Upper and Lower Grey Deposit Assemblages: Too little of the lower grey deposit was excavated to allow for valid comparisons of all classes of artifacts. However, in spite of the differences in sample size, it is possible to observe considerable cultural continuity between the two assemblages. Where the number of specimens from each deposit is fairly large, as with the chipped points, or ground slate knives, the uniformity  132  is striking® Thisconsistency: makes the few disparaties all the more apparent® Reasons for some differences are probably related®  Thus, the presence  in the upper grey layer of all but two of the copper objects, and all of the brass, iron and glass items is part of a single phenomenon — the influx of trade goods® Possibly related to this, moreover, is the disproportionate number of disc beads in the lower grey layer®  The advent of glass trade  beads could well have led to decline in the use of aboriginal beads® Attention is drawn to two other differences between the assemblages but no explanations can be offered®  First, charred bark rolls were  confined exclusively to the lower grey deposit®  Secondly, a high pro-  portion of the stone pipes in the lower grey layer had stem end pieces® In contrast, pipe stem ends from the upper grey deposit were in most cases plain® Dating of the Occupation layers; Only a tentative dating can be attempted at present. The abundance of trade beads and trade copper in the upper grey deposit would place the pit house occupation at no earlier than the late eighteenth century. On the other hand, the presence of only blue beads would argue that the house was likely abandoned before the Hudson's Bay Company was actively trading on the Lower Fraser or in the southern interior®  The common Hudson's Bay beads were of two types®  Ore was bit©,  but of a cylindrical and facetted form, and the other spherical or  133  cylindrical and red. Neither of these types was found at Esilao. Since the Hudson's Bay Company had taken over the Northwest Company's Kamloops post in 1822 and had established Fort Langley on the Lower Fraser delta in 1827, it seems probable that Esilao pit house 1 was deserted before either of these dates. Although Fraser (Lamb, 1960:8^) comments upon the abundance of blue beads on the Fraser River above the Canyon, he does not refer to such beads in his description of the Indians at Yale (lamb, 1960:100). As his cataloguing of native ornaments if fairly detailed in this particular journal entry, the omission of blue beads is perhaps significant. We may, then, tentatively place the abandonment of pit house 1 at some time between Fraser's journey (1808) and the arrrval of Hudson's Bay Company trade goods (1822 - 1827). Inter-areal Comparisons: Table XXXVI summarizes site comparisons made during presentation of the archaeological data. In the table, several sites have been grouped under single headings. Northwest Coast (NWC) included Stselax, Msrpole, Port Hammond, Cattle Point, and coastal sites on Vancouver Island and north. Esilao (E) includes both upper and lower grey deposit assemblages. Northern Plateau (HP) includes Lytton, the Thompson River sites, Chase and Chinlac in the central interior. DallesDeschutes (D-D) includes those sites examined by Strong, Schenck, and Steward and by Cressman. The Upper Columbia and McNary Reservoir headings include the studies of Collier et al and Osborne, respectively. Under the East Xootenays heading are Borden's Columbia-Windermere Lakes and  134  Kootenay River .Valley'surveys, •A. different mode of presentation is adopted from that used in Table II — the summary of the ethnographical study. The archaeological data allows, more accurate estimates of the in?)ortanee of the traits in each area. Table XXXVI, therefore, uses a key Indicating whether a trait Is absent, rare, present, common or abundant,, In the compilation of Table II, no trait was included unless it showed a demonstrable coastal or interior affiliation. This rule has also guided the selection of traits in Table XXXVI. Of 18 traits listed for Esilao, 14 are shared with the Northwest Coast sites, but of these, 8 are rare on the coast and 2, common or abundant on the coast, are rare at Esilao. Thus of the 18 Esilao elements, only 4, or 22 percent, show significant affiliations with the Northwest ' Coast. With the Northern Plateau, Esilao also shares 18 traits. Of these, one is rare in the interior and another rare at both Esilao and on the Northern Plateau. The 16 elements important in both Esilao and Northern Plateau assemblages represent 78 percent of the Esilao traits. Finally, the results of our archaeological comparisons indicate that of the 18 Esilao traits with discernible coastal or interior affiliation, 67 percent have greater interior affinities.  135  TABLE XXXII DISTRIBUTION OP TRAITS REVEALED  BT THE ARCH&EOLOGIGftI  INVESTIGATION AT ESIIAO  Ke^s  miGt Northwest Coast; E: Esiiao; NPs Northern Plateau;: D-Di Dalles-Deschutes; McN: McNary Reservoir; UCs Upper Columbia; EKs East Kootenays. Xs Trait present; At Abundant; Ct Common;: Rt Rare; Absent. ?%  Unknown.  Trait  NWC  E  NP  1. Ground slate knives  A  A  R  -  2, Ground slate points  C  R  R  -  -  3. Spindle whorls  X  X  »  -  -  Disc beads  A  R  -  X  X  5. Stone Vessels  C  ?  X  X  X  ' X  »  6. Abraders  c  c  X  X  R  R  R  7. Unstemmed, triangular sidenotch points  R  G  c  «s»  c  G  G  3. SCa points  R  C  X  c  X  X  G  9. Flake scrapers  R  k  &  c  G  c  G  10, Boulder-chip scraper-knives  R  c  X  X  X  X  X  11, Stone pipes  R  G  c  JL  X  X  12, Copper tube beads  R  c  X  c  G  G  13, Blue glass trade beads  R  C  ?  C  G  „  H* SBb points with concave edges  c  R  X  «  c  X  »  «  15, Chipped stone drills  -  X  X  X  X  X  X  X  „  16, Stemmed points predominate 17, Bark roll torches 18, NBbl points with unilateral spur  -  X -  c  G  X  X  D-D  «.  uc  EK  McN.  •  •  „  X «  136  IV. CONCLUSIONS  '  In the introduction we stated that this study reflected a current interest in the convergence of archaeology and cultural anthropology and quoted several relevant passages from de Lagum (i960)* We may now review the findings of our investigation and assess the usefulness; of the adopted theoretical orientation.. First, a restatement of the objectives of our enquiry is necessary. We proposed, primarily, to undertake a comparative study of two adjacent, yet linguistically distinct, Canyon groups, the Tait and. the Lower Thompson*. The study, it was felt, might reveal an overlapping of ethnographic and archaeological data and that, moreover, ethnographic and. archaeological evidence would indicate whether the culture of the Esilao villagers aligns more closely with either the coast or the interior. The Comparative Study: In Part II, we examined ethnographic material bearing on the two Canyon groups. The results of this study reveal considerable uniformity of culture within the Canyon. A measure of this uniformity is the high proportion of Tait culture elements shared with the Lower Thompson. It was found that the Canyon groups shared 23 out of 27 traits, of which, on the one hand, 11 are not found on the coast, while 6 do not occur in the Northern Plateau. Thus, of the Tait elements showing distinct coastal or interior affinity, 85 percent are shared by the Lower Thompson* Results of the archaeological investigation (Part III) substantiate the ethnographic conclusions, although without archaeological data from  137  a Lower Thompson site, the findings are not strictly comparable. However, we may Infer that as. the Northern Plateau shared 78 percent of the listed Esilao characteristics, the proportion of traits that were probably shared by the Lower Thompson and Tait sites would be considerable® The study yielded no traits with exclusive Canyon distribution. Uniformity of Tait and Lower Thompson cultures lies In their closely parallel combinations of coastal and interior traits. The. reasons for the similarity of Canyon cultures have not been explored beyond mention of a common, ancient Salishan heritage, identical ecological base, and a shared isolation from.outside contacts®  In this  study, demonstrating the uniformity has been sufficient® Ethnographic and Archaeological Overlap-pings As a subsidiary objective, we sought what de Laguna (1960:2) has termed an overlapping of ethnographical and archaeological data. The same writer noted that because exchange between neighboring peoples tended to make the cultures within an area, similar to each other, the archaeologist might discover, in modern cultures of the region, clues that shed light on his own particular finds. Hence, for interpretation of our archaeological data, we have drawn on ethnographies for the two Canyon groups, finding, in the demonstrated uniformity of Canyon culture, justification for use of both Tait and Lower Thompson ethnographic sources in the archaeological analysis of a Tait site® Throughout Part III, attention has been drawn to the areas of ethnographic and archaeological overlap. In many other instances, the  138  archaeological data were seen to complement, the ethnographic information. Table..-XXXVII summarizes these areas of overlapping or complementary data. TABIE XXXVII INTEGRATION OF ETHNOGRAPHICAL AND ARCHAEOLOGICAL INVESTIGATION Trait 1. Pit house dwelling  Ethnographic Sources  Archaeological Investigation  X  X  a) Four uprights set 15" in ground  X  X  b) Central hearth  X  X  c) Inside wall angle of 70°  X  X  d) Sub-rectangular plan  X  e) Rocks anchoring rafters  -  X  f) Earth bench in house  »  X  g) Rocks at base of uprights  m»  h) Raised earth wall 2. Buckskin clothing a) Decorative copper tubes  »  Rare  X X Possibly common  X  X  -  X  X  X  X  x  X  X  X  X  a) Adze blades  X  X  b) Hand mauls  X  X  3 • Bark torches Stone vessels 5. Arrow points of chipped stone Arrow points of iron 6. Wood-working industry  '  .  • .  139  TABLE XXXVII (continued)  Trait  Ethnographic Sources  Archaeological Investigation  c) Stone point drills  X  X  d) Chipped stone scraper-knives  X  X  X  X  7. Stone-working industry a) Saws  X  b) Stone point drills  -  X  c) Carving tools  -  X  Smoking  X  X  a) Use of tubular pipe  X  X  b) Many pipe forms  -  X  c) Pipe manufacturing in Canyon  -  X  x:  X  x:  X  a) Copper pendants  X  X  b) Copper tube necklaces  X  X  c) Glass beads  "V V J  X  d) Use of ochre  X  X  9. Ground Slate Knives 10. Spinning of wool or other fibre lie Personal Adornment  We noted, in a passage quoted from de Laguna, that the ethnologist is unable to "grasp clearly just those aspects of the culture which may best be understood in their material embedments . . . " (1960:2). The archaeological investigation at Esilao has added considerably to our knowledge of such  140  aspects of;Canyon culture* In particular, the following implements not mentioned in the ethnographies were found during excavation bf the site: Various styles of chipped projectile points Boulder chip scraper-knives Choppers . ' Small carving chisels Several forms of pipes Abrasive stones and whetstones Stone saws Stone spindle whorls Stone pendants Hammerstones of 'elongated cobbles Bark roll torches Apart from adding to the tool inventory, we have furnished reconstruction of some steps in the manufacture of specific items of material cultureo The investigation has also added several details to our understanding of pit house construction® In summary, the excavation has: contributed much to our knowledge of Canyon culture and our investigations have demonstrated that ethnographic and archaeological data do overlap at Esilao. Coastal and Interior Affinities: In the analysis of Tait ethnographic and archaeological data, we noted a significant alignment with the interior. The ethnographic study revealed that of the Tait characteristics with discernible interior or coastal affiliations (Table II), approximately 60 percent possessed greater interior affiliations. Of all the elements at the Tait village of Esilao showing either a marked interior or coastal affinity (Table XXXVI), 67 percent pointed to the interior. Among the most prominent interior traits present in the Canyon, specifically at the Coast Salish village of Esilao, were the pit house,  HI  the great importance of chipped stone implements, and the abundant tubular pipe remains. Possible evidence for another significant interior trait, buckskin clothing, lay in the boulder chip scraper-knives and chipped end-scrapers. The reasons behind the interior alignment are beyond the scope of this study. Affinities with the interior may indicate the beforementioned cultural heritage shared by Coast and Interior Salish groups. But there is evidence, too, of more recent contact with' the interior. The presence of quantities of blue glass trade beads at Esiiao is in striking contrast to their scarcity in the historic horizon at Stselax, a Goast Salish settlement at the mouth of the Fraser River. Aboriginal pipes seem also to be exceedingly rare in Fraser delta sites, while less than 110 miles up-river, there is definite evidence of pipe manufacture and use® In short, the picture emerging from our ethnographic and archaeological studies Is that of a Canyon culture more closely allied with the interior than with the coast. Sharing this interior-aligned Canyon culture are two linguistically distinct groups — the Lower Thompson and the Tait® The former are commonly regarded as an interior people, but the latter, as members of the Coast Salish linguistic division, are customarily linked with the coast® . Problems for Future Investigation: Our investigation has suggested a number of problems which must go unsolved in this enquiry: There is, first, the apparent similarity between Esiiao pit house No.l and the Lillooet form of semi-subterranean dwelling. The Tait and Lillooet  14.2  shared, of course, a common hunting territory boundary in the mountains between Harrison Lake and the Fraser Canyon, and there may have been fairly close contact between the groups® Bark roll torches and the small, unilaterally-spurred NBbl form chipped points have parallels on the northern part of the Northern Plateau and in Carrier territory®  The distribution of neither item is  adequately known® An affiliation with the Lower Columbia River region is indicated by the presence at Esilao of concave-edged SBb form chipped points® Much could be learned through investigation of sites in the intervening area ~ particularly those .lying along the eastern slopes of the Cascades Mountains and in the Otemgan and Similkameen Valleys® Finally, we may observe that much work must be done before the significance of trade goods is known. Such often-neglected contact items as glass trade beads can, with adequate study, be useful in dating of recent sites and for the reconstruction of trade relationships® All these are questions for further study. Our examination of two adjacent Canyon groups has revealed a significant uniformity of Canyon culture and has also-shown that this uniform culture displays greater Interior than coastal affinities. We have demonstrated as well the overlapping of ethnographical and archaeological data and shown the part archaeology can play in complementing ethnographic sources.  143  BIBLIOGRAPHY Barnett, Homer G. (1939). Culture element distributions; IX; The Gulf of Georgia Salish. University of California Anthropological Records, Vol. 1, No. 5« Berkeley. Boas, Franz (1894). The. Indian tribes of the Lower Fraser River. Report of the British Association for the Advancement of Science, Vo. 64. London. Borden, Charles E. (1950). Preliminary report on archaeological investigations in the Fraser delta region. Anthropology in British Columbia, No. 1 Victoria. (1951). Facts and problems of northwest coast prehistory. Anthropology In British Columbia, No. 2. Victoria. (1952). Results of archaeological investigations in central British Columbia. Anthropology in British Columbia, No. 3. Victoria. (1954). Some aspects of prehistoric coastal-interior relations in the Pacific Northwest. Anthropology in British Columbia, No.4. Victoria. . (1955). An ancient Coast Indian village In southern British Columbia. Indian Time, Vol. 2, No. 15. Vancouver. (1956). Results of two archaeological surveys in the East Kootenay region of British Columbia. Research Studies of the State College of Washington, Vol. 24. Pullman. (i960). DjRi 3, An early site in the Fraser Canyon, British Columbia National Museum of Canada, Bulletin 162. Ottawa. (1961)0 Fraser River archaeological project. Anthropology Papers, No. 1, National Museum of Canada. Ottawa. (1962). West coast crossties with Alaska. Prehistoric Cultural relations between the Arctic and Temperate zones of North America. Arctic Institute of North America, Technical Paper No. II. Montreal — Personal communication. Collier, Donald, A.E. Hudson, and A. Ford (1942). Archaeology of the upper Columbia region. University of Washington Publications In Anthropology, Vol. 9, No. 1. Seattle.  144  Cook, Captain James (1785). A voyage to the Pacific Ocean. Vol. 2 London. Creech, E.P. (1941). Similkameen Trails, I846 - 61. British Columbia Historical.,Quarterly, Vol. 5, No. 4. Victoria. Cressman, Luther S. (i960) Cultural sequences at the Dalles, Oregon. Transactions of the American Philosophical Society* Vol. 50 Part 10. Philadelphia. ' • ' ' Drucker, Philip (1943). Archaeological survey on the northern northwest coast. Bureau of American Ethnology, Bulletin 133, No. 20. Washington. (1950). Culture element distributions; XXVI; The northwest coast. University of California Anthropological Records, Vol. 9 No.3. Berkeley. • > (1955). Indians of the northwest coast. American Museum of Natural History, Handbook No. 10. New York. Duff, Wilson (1952). The Upper Stalo Indians of the Fraser Valley, British Columbia. . Anthropology in British Columbia, Memoir No. 1. Victoria. (1956). Prehistoric stone sculpture of the Fraser River and Gulf of Georgia. Anthropology in British Columbia, No. 5. Victoria. Foster, Douglas (n.d.). Post-contact goods from Stselax. Manuscript, term paper, University of British Columbia. Vancouver. Haeberlin, ^H.K., J.A. Teit, and II.H. Roberts (1928). Coiled basketry in British Columbia and surrounding region. Bureau of American Ethnology, 41st Annual Report. Washington. Hill-Tout, Charles (1900). Notes on the N'tlakapaMUQ of British Columbia, a branch of the great Salish stock of North America. Report of the British Association for the Advancement of Science, Vol. 69, 1899. London. Hutchison, Bruce (1950). The Fraser. Clark, Irwin. Toronto. Jenness, Diamond (i960). Indians of Canada. National Museum of Canada. Bulletin 65, No. 15. Ottawa. Xephart, Horace (1957). Camping and woodcraft. Macmillan, New York.  145  King, A.R. (1950). Cattle Point, a stratified site in the southern northwest coast region. Memoirs of the Society for American Archaeolopv No. 7, Menasha. ' de Laguna, Frederica (1956). Chugach prehistory; The archaeology of Prince William Sound, Alaska. University of Washington Publications in Anthropology, Vol. 13. Seattle. (i960) . The story of a Tlingit Community; A problem in the Relationship between archaeological, ethnological, and historical methods. Bureau of American Ethnology, Bulletin 172. Washington*  Personal communication. Lamb, W.K. (i960). Simon Fraser, letters and journals, 1806 - 1808. Macmillan, Toronto. Munro, J.A., and I. McT. Cowan (1947). A review of the bird fauna of British Columbia. British Columbia Provincial Museum, Ho. 2. Victoria• Osborne, Douglas (1957). Excavations in the McNary Reservoir Basin near Umatilla, Oregon. Bureau of American Ethnology, Bulletin 166, No.8. Washington. Province of British Columbia (1961). Climate of British Columbia. Department of Agriculture. Victoria. Ray, Verne F. (1939). Cultural relations in the plateau of northwestern America. Publications of the Fredrick Webb Hodge Anniversary Publication Fund, Vol. 3. Los Angeles. (1942). Culture element distributions; XXII: Plateau. University of California Anthropological Records, Vol. 8, No. 2. Berkeley. Sanger, David (1962). Archaeology of EeQw 1. Manuscript, M.A. thesis, University of British Columbia. Vancouver. Smith, Harlan I. (1899). Archaeology of Lytton, British Columbia. Memoirs of the American Museum of Natural History, Vol. 2, Part 3. New York. (1900). Archaeology of the Thompson River region, British Columbia. Memoirs of the American Museum of Natural History, Vol.2, Part 6. New York. (1903). Shell-heaps of the lower Fraser River, British Columbia. Memoirs of the American Museum of Natural History, Vol.4, Part 4. New York.  146  (1907). Archaeology of the Gulf of Georgia and Puget Sound. Memoirs of the American Museum of Natural History, Vol. Part 6. New York. ... (1910). The archaeology of the Yakima Valley. Anthropological Papers of. the American Museum of Natural History, Vol.6, Part 1. New York. Smith, Marian W. (194-7). House types of the middle Fraser River. American Antiquity, Vol. 12, No.4» Menasha. (1950). Archaeology of the Columbia-Fraser region. Memoirs of the Society for American Archaeology, No. 6. Menasha. Strange, James (1928). Journal and narrative of the commercial expedition from Bombay to the north-west coast of America. Ed. A.V. Venkatarama Ayyar, Records of Ft. St. George. Madras. Strong, W.D. (1935)• An introduction to Nebraska archaeology. Smithsonian Miscellaneous Collection, Vol. 93, No. 10. Washington. Strong, W.B., W.E. Schenck, and J.H. Steward (1930). Archaeology of the Dalles-Deschutes region. University of California Publications in American Archaeology and Ethnology, Vol. 29, Ho. 1. Berkeley. Suttles, Wayne P. (i960). Affinal ties, subsistence and prestige among the Coast Salish. American Anthropologist, No. 62. Menasha. Taylor, W.W. (1948). A study of archaeology. American Anthropological Association, Memoir No. 69. Memsha.. Teit, James (1900). The Thompson Indians of British Columbia. Memoirs of the American Museum of Natural History, Vol.2, Part 4,. New York. (1906). The Lillooet Indians. Memoirs of the American Museum of Natural History, Vol. 4, Part 5. New York. (1909). The Shuswap. Memoirs of the American Museum of Natural History, Vol. 4, Part 7. New York. (1917). Thompson tales. Folk-tales of Salishan and Sahaptin tribes. American Folk-Lore Society, Memoir No. 11. New York. Willey, G.R., and P. Phillips (1958). Method and Theory In American Archaeology. University of Chicago Press, Chicago. Wissler, Clark (1923). Man and culture. Thomas Y. Crowell, New York.  147  PLATE I: Chipped points and drill from upper grey deposit. 1-2, NAbl; 3, NAb2; 4, NBa; 5-6, KBalj 7, "NBbl; 8-9, NBb2; 10, NBcj 11-12, miscellaneous; 13, Ste; 14, SBalj 15-16, SBb; 17, SBc; 18-19, SCal; 20-22, SCa2; 23-24, SCa3; 25-26, SCb2; 27, SCc; 28, Chipped stone drill.  148  PLATE II: Chipped points from lower grey deposit. 1-2, m.b2j 3 NBalj 5-6, NBblj 7-8, NBc; 9-10, miscellaneous; 11, S&a; 12, S&b SBa; 14-16, SBb; 17-18, SCalj 19-23, SCa2; 24, SCa3; 25, SCb2.  149  PLATE Ills 1, chipped end-scraperj 2, chipped scraper-knifej 3, problematical steatite object (No. 6375); 4, ground slate knife (No. 2506); 5, chipped slate knife. All from upper grey deposit.  150  PLATE IVi Pipe fragments from lower grey deposit. 1-4, bowl fragments; 5, fragment of unfinished conical pipe; 6, fragment of unfinished cigarshaped pipe; 7, stem fragment; 8-9, stem end piece fragments; 10, stem fragment; 11, stem end piece fragment, possibly blank for pipe decoration as in Plate V:16; 12-17, stem end fragments.  151  PLATE ¥j Pipe fragments from upper grey deposit. 1, fragment of unfinished, flaring bowl; 2-3, fragments of bowl blanks; pentagonal bowl fragments; 5-7, bowl fragments; 8, stem fragments; 9, complete conical pipe; 10, bowl of conical pipe similar to 9j 11, stem fragment; 12-15, stem ends; 16, base of bowl, stem and decorated stem eoi.  152  PLATE VI; 1, Spindle whorl fragment; 2,4, boulder chip scraper-knives 3, piece of steatite showing method of sawing and breaking. All from i^jper grey deposit.  153  PLATE VIIs Saws and abraders. 1, fragment of possible slate file; 2-3, schist saws; 4, abrasive stone; 5, whetstone. Ill from upper grey deposit.  154  PLATE VIIIs 1, fragment of hand maul from lower grey deposit; 2, decorated stone vessel fragment; 3, hammerstone; 4, hammerstone possibly of type used to block out slate artifacts. 2-4 from upper grey deposit.  155  PLATE U s Pendants, beads and problematical objects. 1, decorated steatite pendant (No. 8533); 2, possible steatite pendant (No. 1199); 3, pendant (No. 5418); 4, pendant (No. 7367); 5, decorated problematical object (No. 8691); 6, pendant fragment (No. 2584); 7, large, spherical light blue glass trade beads; 8, oval blue glass trade bead; 9, small, spherical light blue glass trade beads (one at right is opaque); 10, dark blue translucent glass trade beads, 11, cylindrical, light blue glass trade beads; 12, problematical carved steatite object; 13, disc, bead; 14, bead blanks, 15, problematical steatite object (No. 5354); 16, possible zoomorphic pendant (No. 7135); 1-11, upper grey deposit; 12-16, lower grey deposit.  156  PLATE X: Metal objects from upper grey deposit. 1-2, iron points; 3-9, copper tubes; 10, copper ring; 11, copper bracelet; 12, brass object, possibly pendant; 13-14, copper tubes; 15-19, small pieces of sheet copper 20, concave copper pendant (No. 8 9 9 7 ) ; 21, copper jingle; 22-24, sheet copper pendants.  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items