FIELDIANA Anthropology Published by Field Museum of Natural History VOLUME 60 AKULIVIKCHUK: A NINETEENTH CENTURY ESKIMO VILLAGE ON THE NUSHAGAK RIVER, ALASKA JAMES W. VANSTONE OCTOBER 16, 1970 FIELDIANA: ANTHROPOLOGY A Continuation of the ANTHROPOLOGICAL SERIES of FIELD MUSEUM OF NATURAL HISTORY VOLUME 60 p MU ia of s _ 4" NATURAL &” History > Say "py peaaserwmsens’ |, FPHVD \c aamnemacesenn ii=il ee ce een, oro BY nanan CHICAGS FIELD MUSEUM OF NATURAL HISTORY CHICAGO, U.S.A. = AKULIVIKCHUK: A NINETEENTH CENTURY ESKIMO VILLAGE ON THE NUSHAGAK RIVER, ALASKA Frontispiece: Akulivikchuk in 1967. FIELDIANA Anthropology Published by Field Museum of Natural History VOLUME 60 AKULIVIKCHUK: A NINETEENTH CENTURY ESKIMO VILLAGE ON THE NUSHAGAK RIVER, ALASKA JAMES W. VANSTONE Curator, North American Archaeology and Ethnology OCTOBER 16, 1970 PUBLICATION 1108 PATRICIA M. WILLIAMS Managing Editor, Scientific Publications Library of Congress Catalog Card Number: 70-129465 PRINTED IN THE UNITED STATES OF AMERICA BY FIELD MUSEUM PRESS Contents HIST OF TMUUSTRATIONS — cy 5 en ee eS ee ie i, so. 4 SMe ice. URPIN tee 7e s G T SS OE Es ce Ra ina, od TR ORES x Come cer eae en Wil OC Pa er 9 te OW i 2 en ep a EE 5S INon=HNiskimo, pottery << . ... .....5 5522214. .s.8 1... . ~68 11. F2) 2G ea YY 0 Imported manufactured goods ...........2.2.2.2.2.2.2.. 78 Non-Eskimo pottery ...........2.2.2.2.2.242.0242442.4. 74 Ress, BOs ws ess Poe ne a fa ne ee vee OS ee ges ee eee el Mietalits 2235 8 ee es es ee 5 Miscellaneous materials . 2... 2... 2 CONTINUITY AND INNOVATION ..............2.2..42.42. 2. 92 INTERPRETATIONS AND CONCLUSIONS. ............... =. . 108 REFERENCES ............ 2... 2.248. So aes cee A oe SLO APPENDIX. Akulivikchuk trait list . 2... ......2.2.20202.2. 2. . 115 List of Illustrations mculvakenuk im V9G% <<. 2 aaa sca « G & GA Geo » 2 Frontispiece TEXT FIGURES — WiaptOr lanka: aba 2, bug tm o «fice 8d. Bo Bu. & 2b ow a eee. 168 Map of the Nushagak River region ................ «10 Map of the Akulivikechuk site ..........2.2.2.22224.2. 18 POISON eG, 1d, Shh ch Sew A GA Gee de Ee BY ey iene & By See 228 NSA ORE ne: ay oti avalic <8 Se o5y.cn Wp We. Sighs. Ste hy VET RCM ae Mer xa ae te Os Pes, OD HIROMAC TS Aaa. AS Se oP Uh “Gt eed pn. ee ce WO es Ese Boe ea ee MOORS ERR cao A Ue ae, ee as, RCA, hanks toe ee Mee eyes Ay Sa ek EROUS Ch OME UN AEE DURMERS TOMS Gg 45" 4! WG, vm, sak Sho ice Ge die a SA, Seal BOS CNG mR RSENS, ne POD cat es a ee Bee I ge A, ae aw SOD MU PEMB LOUIS le CVO AA eG A GA, Se Gch chen ie ahs nde ae Mahe G cm ot 1p Oe bo me CO Or co 60; yi OD GCS). CCHnt er ie oso) oe ee a eens AA Ue ee ee OS 12. Main chamber of kashgee in cross-section ..........2.2.~. «87 13. a) Composite harpoon head. b) Unbarbed, socketed projectile point . . 60 14. Bead types according to shapes, showing cross-sections of each . . . . 83 TABLES 1. House construction features .........2..2.2.2.2.2.2.2.2. 22 2. Animal bones recovered from the Akulivikchuk site... ...... 95 PLATES i Stone artltacts: 3. no 4 4 Gis es & Gb ew do we SS Be wo eo OAL 2% SvOne artiacts+. 4. « 6s ew S42 eae wae oe ee oe oe ee DEO LONEG APUACLS ak. des ia ae en & Gey Gh Sh ba Me ee a wm Gomee ee AS UOMCKAPGIACES) at's, 46 Spben--)ig- $12 be: OS a) doce a ae, BD SREPOUOMCRADCIMACES cite: Fc dsc idie ch -'Sp'> ai Lisa Ma. say etna lene eel Oe te aE OL Or OLONCRARUTACLS pwr Nal leks ate) 2) veer ip eget te ee ed ES a OMI CV CLGDEEVOUSE ge a Et eS RRP GAs eee Ae es Sa Go ed, ce ae Y5y5) or LIST OF ILLUSTRATIONS Bone and antler artifacts Bone and antler artifacts Bone and antler artifacts Artifacts of bone, antler and miscellaneous materials Wood and metal artifacts Non-Eskimo pottery Artifacts of metal and glass Artifacts of metal and miscellaneous materials PAGE 57 59 63 66 cl 76 87 89 Acknowledgements Excavations at Akulivikchuk were supported financially by a grant from the National Foundation for the Humanities (H67-0-51). Able assistance in the field was provided by Mr. Robert Lesch, at that time an undergraduate at the University of Illinois at Chicago Circle. In Alaska, the following individuals were particularly helpful be- cause of their interest in the work and willingness to assist the author in dealing with the many problems inevitably associated with archae- ological excavations in remote areas: Mr. and Mrs. Charles Lucier of Anchorage, Mr. and Mrs. C. J. Stovall and Mr. and Mrs. Alfred Andree of Dillingham, Alaska Department of Fish and Game em- ployees in Dillingham, Mr. Gust Johnson of New Koliganek, Mr. Paul Romie, Mr. Fred Hurley, and Mr. Luki Aleklok of Ekwok. The author is also grateful to several of his colleagues at Field Museum of Natural History for assistance during his study of the collections from the Akulivikchuk site. Dr. Glen Cole guided me toward a more accurate description of lithic materials and Dr. Alan Solem identified several mollusk shells that were recovered. The text figures and maps were drawn by Mr. Gustaf Dalstrom and the excellent photographs are the work of Mr. John Bayalis and Mr. Homer V. Holdren, Division of Photography. For critical comments and helpful suggestions during the prep- aration of this study, I wish to express my appreciation to Dr. Wendell H. Oswalt, University of California, Los Angeles. DENN Co (é) / So 0 SS S35 {n DYNS mare ee CX 9 Puv| NN aroor” N = Ny NS \ , \ = SD] DUWONIT 6K 4900p a SAOyDUijoy yoluy } 90uasIMO | 92Ua1M07 aS SS Preface The present study is one of a continuing series of publications dealing with the culture of the Nushagak River region in southwest- ern Alaska during the historic period (Fig. 1). Earlier publications in the series are a monograph on the ethnohistory of the region (VanStone, 1967), an annotated ethnohistorical bibliography (Van Stone, 1968a), and an archaeological report dealing with excavations at Tikchik, a nineteenth century settlement near the mouth of the Tikchik River (VanStone, 1968b). Field work was begun in the summer of 1964 with an archaeolog- ical survey of the Nushagak River and three of its major tributaries, the Wood, Nuyakuk, and lower Mulchatna rivers (Fig. 2). Forty- five sites belonging to the period of historic contact were located and many of them mapped. The survey was continued during the follow- ing three summers and at the present time there is information on a total of 58 historic sites in the area. Eventually a detailed report on nineteenth and early twentieth century settlement patterns through- out the river system will be prepared, but this requires an additional season of field work. The first summer of the survey suggested that throughout the period of historic contact there were centers of population along the upper and middle river. For purposes of the survey, the upper Nush- agak included Tikchik Lake, the Nuyakuk River, and the Nushagak between the mouths of the Nuyakuk and the Mulchatna, including the lower reaches of the latter. The Nushagak above the mouth of the Nuyakuk has not been included in any phase of this project be- cause of the virtual absence of reported historic settlements in that region. The middle river, a somewhat more populous center, was defined as encompassing that part of the Nushagak between the mouths of the Mulchatna and Kokwok Rivers. The river below this point, together with Nushagak Bay, may be regarded as constituting a third population center. Fig. 1. Map of Alaska. 10 FIELDIANA: ANTHROPOLOGY, VOLUME 60 Moy The Res 7 (New Koliganek Ahkokpak x PE lilakok SX7/T unravik <<) ¢ New Stuyahok } wok o/Agivavik WAkKulivikchuk aff{Kokwok, io Nushagak (Aleksandrovsh: Redoubt) So oe es ZS (i aL ahve a moo ole may, Fic. 2. Map of the Nushagak River region. Archaeological research was planned in each of these centers and initiated with excavation of the Tikchik site during the summer of 1965. The selection of a site in the middle river region proved more difficult. At least three localities were considered potentially suit- able, and there was also the possibility of working in several smaller settlements. Akulivikchuk, although not mentioned as frequently VANSTONE: AKULIVIKCHUK 11 in the historical sources as Kokwok and Agivavik, the other two large sites in the immediate area, had the advantage of having been aban- doned early enough so that a twentieth century occupation would not confuse the excavation picture. Agivavik also qualified in this regard, but appears to have had a prehistoric component. In addi- tion, much of this site, which is somewhat larger than Akulivikehuk, is covered with a heavy growth of spruce and willow trees. This would have made sufficiently extensive excavation virtually impos- sible with the crew and time available. In short, Akulivikchuk ap- peared a logical choice, and the most rewarding location for an archaeological approach to the nineteenth century in the middle river region. Excavations at Akulivikchuk were begun on June 15, 1967 and completed on August 19. In retrospect it must frankly be admitted that the site hardly lived up to expectations. The artifact collection is not extensive and the number of trade goods recovered was dis- appointingly small. The main value, perhaps, of the information contained in this study is that it provides important data for com- parison with recovered materials from the growing number of exca- vated historical sites in southwestern Alaska. Also, of course, it broadens considerably our perspective of Nushagak Eskimo culture in the nineteenth century and allows us to make at least a few gen- eralizations concerning culture change in the area during the early contact period. In the interests of consistency and to facilitate comparisons, the organization of the text follows as closely as possible other reports of historic site excavations in southwestern Alaska in which the author has participated. These include the above-mentioned Tik- chik study, an account of excavations at Crow Village in the middle Kuskokwim River region (Oswalt and VanStone, 1967), and a mono- graph on the nineteenth century Tanaina Indian village of Kijik on Lake Clark to the north and east of Iiamna Lake (VanStone and Townsend, 1970). The Historical Setting Informants who were questioned concerning the meaning of the name Akulivikchuk generally agreed that it was simply a place name. One person, however, suggested that the word may have been de- rived from agoonli which means “‘in between,” a reference to the small creek which formerly divided the site into two nearly equal parts. Throughout its recorded history, the settlement was occupied by Yupik speakers of the Western Eskimo language stock. This dialect was spoken in all the villages along the Alaskan coast from the vicin- ity of Nome southward to Bristol Bay and the western end of Tlamna Lake. The inland range of Yupik speakers was to the village of Paimiut on the Yukon River and the vicinity of Aniak on the Kus- kokwim River. The Eskimos of the Nushagak River region have the ethnic name of Kiatagmiut. This sub-group of Yupik speakers occupied, at the time of historic contact, the entire Nushagak River, excluding Nushagak Bay, and the area to the west as far as and in- cluding the Wood River Lakes and Tikchik Lakes (Oswalt, 1967, p. 6; VanStone, 1967, pp. 110-111). The cultural center of Yupik-speaking peoples was along the central Bering Sea coast. In this area the people were oriented toward a maritime economy in which the seal was the most impor- tant food animal. On the adjacent tundra, caribou were hunted and fishing for salmon was significant at the mouths of rivers and in cer- tain bays. The Yupik penetration of the Nushagak River system took place at some unknown time during the prehistoric period when the people presumably moved inland from the Bering Sea coast. Since, at that time, they already possessed a well-developed salmon fishing technology, these Eskimos were able to exploit effectively an inland environment like that along the Nushagak River and its trib- utaries where these fish are abundant. The first well-documented contact between Yupik-speaking peo- ples and Europeans took place in 1818 when a party of Russian- American Company employees was dispatched from Kodiak Island to explore the territory north of Bristol Bay. During these explora- 12 VANSTONE: AKULIVIKCHUK 13 tions a trading post, Aleksandrovski Redoubt, was established at the mouth of the Nushagak River as the first Company post north of the Alaska Peninsula. Using the redoubt as a base of operations Ivan Filipovich Vasiliev explored the Nushagak River for the Com- pany in 1829 and 18380. In the latter year he crossed over into the Kuskokwim drainage and descended that river to the coast. As a result of these explorations, a series of trading stations were built at various points along the middle Kuskokwim culminating, in 1841, with the construction of Kolmakovski Redoubt opposite the mouth of the Kwik River about 16 km. above Aniak. This station remained an important trading center until abandoned by the Russian-Amer- ican Company in 1866. The exploration of Bristol Bay and the Nushagak River, together with the founding of Aleksandrovski Redoubt, later to be called Nushagak by Anglo-Americans, played a vital role in opening up the interior regions of southwestern Alaska to the fur trade. Kolma- kovski Redoubt continued to be supplied from Aleksandrovski until 1845 and the route up the Nushagak River to its headwaters, across the divide and down the Holitna or Hoholitna to the Kuskokwim became a heavily traveled route with supplies going upriver into the Kuskokwim region and furs proceeding in the opposite direction.! Since Akulivikchuk was almost certainly one of the larger and more important settlements along the Nushagak River throughout all or most of its occupation, and since it was located on the direct route of these regular journeys between the Kuskokwim and Aleks- androvski Redoubt, the village could not fail to have been drawn increasingly into the flourishing fur trade that was rapidly develop- ing throughout southwestern Alaska. And yet there are no specific references to the village in any of the published or archival sources which I have examined dealing with exploration and trade during the Russian period. In 1841 the first Russian Orthodox church north of the Alaska Peninsula was established at Aleksandrovski Redoubt and mission- aries began to penetrate the Nushagak River country. Although very little is known concerning the exact nature of their contacts with the Eskimos of the interior regions, the earliest known reference to Akulivikchuk occurs in the vital statistics of the church at the redoubt. In 1843 a man from the village, presumably a visitor to 1 For a detailed account of explorations and trading activities in the Nushagak River region, see VanStone, 1967, chapters I, III. 14 FIELDIANA: ANTHROPOLOGY, VOLUME 60 the mission at Aleksandrovski, is listed as having been baptized. Residents of the settlement continue to be listed more or less regu- larly in the statistics until 1899, although as we will note presently, the village may have been abandoned somewhat earlier (Alaska Russian Church Archives, vital statistics, Nushagak, 1842-1931). These listings of Akulivikehuk residents in the church statistics are the only references to the village during the Russian period that have come to light. Because of this paucity of information, the sig- nificance of Russian trading and mission activity in the lives of the inhabitants of Akulivikchuk may be determined only within the general framework of Russian expansion into interior Alaska. For this information, the reader is referred to the author’s earlier publi- cations on the Nushagak River, its history and inhabitants (Van Stone, 1967; 1968a, b). It can be emphasized here, however, that as far as commercial relations were concerned, the number of Rus- sians and creoles occupied in trading activities was probably always small but they maintained control over desirable products, and the Eskimos could obtain these only through the exchange of fur. This simple fact explains more than anything else the amazing rapidity with which the Eskimos of southwestern Alaska were exposed to the fur trade. Between 1818 and 1840 the entire region was opened up and trading contacts were established throughout the heavily pop- ulated Yukon and Kuskokwim River systems as well as along the Nushagak and its tributaries. With reference to mission influence, it should be noted that al- though Akulivikchuk residents are mentioned frequently in the vital statistics of the Nushagak church, it seems virtually certain that missionaries did not visit the settlement regularly in spite of its rela- tively accessible location. Throughout the 1850’s and 1860’s, for example, the missionaries at Aleksandrovski were seldom able to visit the villages along the Nushagak more than twice a year. Thus, it is likely that only those settlements nearest the redoubt were strongly influenced by Christian teaching. Residents of Nushagak River villages were, at best, but marginal participants in the newly introduced faith, a situation that probably persisted as long as Akulivikehuk was occupied. For a long time after the transfer of Alaska from Russia to the United States in 1867, no attempt was made to explore the Nush- agak region. At the end of the nineteenth century the inland area north of Bristol Bay still remained comparatively unknown although it continued to be penetrated by missionaries, occasional trappers, VANSTONE: AKULIVIKCHUK 15 and traders. The assets of the Russian-American Company were purchased by Hutchinson, Kohl & Company of San Francisco. This firm was soon reorganized to form the Alaska Commercial Company which dominated trade in southwestern Alaska throughout the rest of the nineteenth century and well into the twentieth. The Russian Orthodox Church continued to send priests into the interior and was the major religious influence in the area as it is at the present time. The Moravian church established a mission near Nushagak in 1884 but it is unlikely that it exerted any influence on the residents of Akulivikchuk at this late date.’ Unfortunately, there is only one definite reference to Akulivik- chuk in the known late nineteenth century reports of missionaries, government officials or travellers. This is somewhat surprising con- sidering the settlement’s accessible location. Most of the villages on the Nushagak River were enumerated in the federal census of 1880 (Petroff, 1884) or 1890 (Porter, 1893). Akulivikchuk is correctly located on Petrofi’s map and a population listing of 72 is given in the text (1884, p. 17, map 1). This number seems reasonable in hight of the appearance of the site today. In the federal census of 1890 reference is made to a village named ‘‘Akgulurigiglak” in the Nushagak district, but it is not shown on the accompanying map. This settlement is listed as having a pop- ulation of 61 including 16 families living in five houses (Porter, 1898, p. 164). These figures again seem reasonable if the settlement in question is indeed Akulivikchuk. If not, the absence of the village from the 1890 census suggests that it had been abandoned by that time. When questioned on the subject, informants could only say that the settlement was abandoned sometime around the turn of the century. Reference to the village in the vital statistics of the Rus- sian Orthodox mission at Nushagak as late as 1899 could be mis- leading since it is not always clear whether the statistics refer to the place of a parishioner’s birth or to his residence at the time his name appears. The cause of the community’s abandonment cannot be deter- mined with certainty. Several informants mentioned that during one winter when many of the village children were playing in front of the settlement on the river ice, it suddenly gave way and a large number were drowned. It was said to be after this unfortunate acci- 2 For a detailed account of Moravian activities in the Nushagak River region, see VanStone, 1967, chapter II. 16 FIELDIANA: ANTHROPOLOGY, VOLUME 60 dent that Akulivikchuk was abandoned. Certainly, it is within the norm of Eskimo behavior to abandon a place of residence after it has been visited by serious illness or some other unhappy event. As we will note later, however, the emergence of other villages in the gen- eral area may also have played a role in the decision to abandon Akulivikehuk. In the summer of 1931 Ales Hrdli¢ka ascended the Nushagak River as far as Old Koliganek. The primary purpose of his trip was to collect skeletal material for the United States National Museum, but he also located and described a large number of archaeological sites, among which was Akulivikchuk. On his way up the river, Hrdliéka (1944, p. 357) notes that Akulivikchuk is “‘a very large old site on right bank, largest yet.’’ On the return trip he was even more impressed and described the site as follows: Reach another large old site two miles below Hurley’s [Ekwok]. Extends on both sides of a now dry small stream and then along a large flowing creek. Much larger than the site above Hurley’s but there is evidently an older part and a later. Scores of square pits of igloos some large, many small. Col- lectively extend along main river and creek for at least half a mile, igloos several deep. (p. 374) It is difficult to recognize the Akulivikchuk site from this description and it is clear that Hrdlitka greatly overestimated its size. This is understandable since the site is characterized by a cleared area that is large in relation to the actual number of house pits. The extremely tall grass growing on the site might also give the impression, to the casual visitor, that many structures are hidden. In addition, the high bank of the river at this point creates a false impression of con- siderable depth and makes the site look something like the large coastal middens with which Hrdlitka was familiar. Akulivikchuk is indeed one of the larger sites along the Nushagak River and its trib- utaries, but there is nothing to suggest that there are, or ever were, “‘seores of square pits of igloos.” Excavations The Akulivikchuk site is located on the west bank of the Nush- agak River approximately 5 km. below the village of Ekwok (59° 22’ N, 157°30’ W). Just above Akulivikchuk the river makes a big bend to the east and at this point there are several small islands which formerly were separated from the river bank by sloughs which connected with the main channel above and below the bend. At the present time, however, one of these sloughs is the main channel and the big bend in the river is in the process of becoming an oxbow lake. It will be a long time, though, before this process is complete and the bend is still the main route for barges and larger boats. The west bank of the Nushagak River just below Ekwok is low but rises again south of Akulivikchuk and is ten to twelve meters high at the site itself. The formerly occupied area appears as a large, cleared, relatively flat expanse of ground approximately 150 m. long and 75 m. deep. This area is covered with tall grass and divided about equally into two halves by a deep ravine that at one time con- tained a running stream (see frontispiece). At the peripheries of the site is a thick growth of small spruce and cottonwoods with the latter being more common near the river bank to the north and south, while the former predominate along the back of the site to the west; a few willows are also growing in the ravine. Trees are not encroach- ing on the site to any great extent and the cleared area of former occu- pation must be very nearly the same size today as it was when the settlement was abandoned. The cleared area to the south of the ravine is somewhat smaller than that to the north, and only a single house together with a num- ber of pits are located on this side. All the other houses, the single kashgee, and a large number of pits of varying sizes are on the north- ern side. The fact that the only good view down the river is from the north side may account for this distribution. This appears to have been a major factor in the orientation as well as the location of the structures since all those excavated on the northern side of the ravine face downriver with one exception (H-1) which opens toward the river. Le soul] peqjop ‘ueppru= J, ‘aabysny=y ‘)-T sosnoy=)-[-H{ :Aoy ‘asnoy po}JVAROXOUN 94BOIPUL ‘ays YNYOHLATNYY oy} Jo dey “¢ “DIq daniy yoboysny 18 VANSTONE: AKULIVIKCHUK 19 A series of approximately 15 pits were located at the upriver end of the site behind the excavated structures. These are presumed to have been primarily for the storage of fish or meat, but their exact purpose is unknown. There are also nine pits of the same general type at the southern end of the site. All pits on both sides measure approximately three meters by two meters and are one meter deep. Directly in front of Akulivikehuk the river is narrow. In fact, this point is one of the narrowest places along its course where the entire river is contained in one channel. The water is shallow along the shore in front of the site, but according to Ekwok informants, the silting is recent and not too long ago the river was deep close to shore and this was a very good place to set salmon nets. Even today a number of Ekwok families place their nets in the vicinity of Akul- vikchuk throughout the entire summer. Across the river from the site, the bank is low and covered with willows. Moose are said to appear frequently in this area and prefer to cross the river here be- cause it is relatively narrow. The slope leading up to the site from the river is quite steep, but the ravine that divides the cleared area is wide at the point where it meets the water’s edge and an easy ascent can be made by following the slope at this point. In fact, this was very likely the main trail leading to the houses at the time the site was occupied. In considering the natural advantages of Akulivikchuk as a place to live, we can mention the high ground, the presence of fresh water running in the ravine, a good hunting and fishing location, and a favorable downriver view of as much as three to five kilometers. Another factor that may have been taken into consideration by the residents was the availability of large timber for building, although there is little of that left in the area today. The purpose of the 1967 field season was to excavate the Akuli- vikchuk site as fully as possible. The first step involved stripping the sod layer from seven house pits and the single kashgee. An eighth house, located at the extreme northeast end of the site facing the river, was not excavated because of time limitations and the fact that several large cottonwood trees were growing in it. This house is in- dicated on the site map (Fig. 3) with dotted lines. A small but obvious midden deposit was found in front of the kashgee and here the sod was stripped from an area measuring eight meters in width and nine meters in length. There did not appear to be midden de- posits in front of any of the houses, although no actual testing was 20 FIELDIANA: ANTHROPOLOGY, VOLUME 60 attempted. It seems likely that much of the garbage and other de- bris disposed of by the villagers was simply thrown over the bank into the river. HOUSES Excavation of the houses at Akulivikchuk proved to be relatively easy once the difficult task of sod removal was completed. At the time work was begun on June 18 the sod was almost completely thawed and only occasionally was frost encountered in any of the house pits. Drainage from the excavations was always good, even after prolonged periods of wet weather. This was due at least in part to the fine-grained sand which underlies the occupational debris over the entire site. A number of generalizations about the seven houses excavated at Akulivikchuk are appropriate before discussing the details of each one in particular. First of all, it cannot be said with certainty that all houses were occupied simultaneously. This is a question that must be dealt with, but it certainly cannot be resolved, or even adequately considered, until after the distribution of artifacts has been studied. Therefore, we will leave consideration of this problem for a later chapter. For the present it is sufficient to note that since the total period of occupancy at Akulivikchuk was almost certainly less than a century and since the dwellings are all structurally similar, we can consider them as a cluster of contemporary residences. Whatever changes may have been going on in the culture of the Akulivikchuk Eskimos throughout the period of occupation, they do not seem to be reflected to any great extent in the construction of the dwellings in which the villagers lived. In each instance the house builders at Akulivikchuk made an excavation in the ground slightly larger than the proposed structure. Thus all the houses were semi-subterranean dwellings. The species of wood employed in construction was, as far as can be determined, spruce. It is possible that some cottonwood logs may have been used, although spruce would generally have been much preferred because it does not rot as quickly. Entrance tunnels are perhaps the most consistent feature of Es- kimo dwellings everywhere and therefore it is not surprising that they were found attached to all the Akulivikchuk houses. The tunnel of only one excavated house (H-1) opens directly toward the river bank, while five (H-2, H-3, H-4, H-5, H-6) face downstream and one VANSTONE: AKULIVIKCHUK 21 (H-7) upstream. The floors of the tunnels of all houses are lower than the house floors, although in three cases (H-1, H-2, H-6) the difference is not great and no cold trap was formed. With one excep- tion (H-2), tunnels project into the main living area, although the degree of projection varies considerably. At the inner ends of the tunnels there are usually split logs or planks which serve to shore up the end and also form an entrance. In five houses (H-3, H-4, H-5, H-6, H-7) this takes the form of a row of horizontal planks, while one structure (H-1) has a semi-circular wooden sill at the inner end. Horizontal planks also form the outer ends of tunnels in six houses (H-1, H-2, H-3, H-5, H-6, H-7); vertical planks in this location are characteristic of one structure (H-4). Tunnel wall construction in four houses (H-1, H-5, H-6, H-7) consists of a series of parallel horizontal logs supported by short vertical ones, while in three (H-2, H-38, H-4) there are vertical wall planks. In each case these appear to have been held in place at the upper end by a long horizontal log. There are clear indications in two houses (H-1, H-4) that short, split logs were placed on top and at right angles to the tunnel wall logs to form the roof. In the same two structures there is evidence that sheets of birch bark had been placed over the tunnel roof logs before they were covered with sod or dirt. In general, it can be said that log preservation was much better in the tunnels than in the main rooms of the Akulivikehuk houses. However, all tunnel walls were, to some extent, compressed by pressure from the earth. Three of the Akulivikchuk houses (H-1, H-2, H-3) have entry- rooms or storm sheds at the outer end of the tunnels. Only one (H-1), however, contains a sufficient number of preserved logs to reveal features of construction. In all three houses the outer end of the tunnel extends into the entryroom and terminates in a pro- nounced step up. Another consistent feature of the three entryrooms is irregular floors that seem simply to be well-trodden, natural ground surfaces rather than actual prepared floors. It seems clear that en- tryrooms were not as well constructed as the rest of the houses. In over-all plan, five houses (H-1, H-3, H-4, H-5, H-7) are virtu- ally square, while two (H-2, H-6) are rectangular with the longer sides at right angles to the tunnel. Horizontal wall logs, presumably always held in place by vertical posts, are a consistent feature of all the houses, although one (H-6) has vertical wall planks along one side. A cribbed corner occurs in one structure (H-7), the cribbing logs being supported by two narrow, vertical posts. The floors of 22 FIELDIANA: ANTHROPOLOGY, VOLUME 60 TABLE 1.—House Construction Features House Feature H-1 H-2 H-3 H-4 H-5 H-6 H-7 Excavated foundation xX xX xX xX xX xX xX Spruce wall logs axe xX xX xX x xX xX Tunnel shallow floor, no cold trap xX xe 0 0 0 x 0 deep floor, cold trap 0 0 xX xX xX 0 xX horizontal wall logs x 0 0 0 xX xX xX vertical wall planks 0 x xX xX 0 0 0 short, split roof logs xe ? ts xe ? ys ? birch bark sheets at roof level xX 0 0 x 0 0 0 horizontal planks at inner end of tunnel 0 0 xX x xX xX xX horizontal planks at outer end of tunnel xX xX x 0 xX xX xX semi-circular wood sill at inner end of tunnel ».« 0 0 0 0 0 0 vertical planks at outer end of tunnel 0 0 0 xX 0 0 0 Tunnel entryrooms xX x x 0 0 0 0 House dimensions square x 0 ».¢ d€ x 0 D€ dimensions rectangular 0 xX 0 0 0 xX 0 horizontal wall logs xX x xX x x x xX vertical wall planks along one side 0 0 0 0 0 xX 0 cribbed corners 0 0 0 0 0 0 xX central hearth xX xX x xX xX xX xX pits in floor lined with birch bark 0 ».¢ 0 0 0 0 0 Benches back bench only 0 0 xX xX 0 xX 0 grass-covered bench 0 0 0 0 0 xX 0 dried grass on floor along side walls xX 0 0 0 0 0 0 Roof four-post-center ? xX xX ? xX xX us birch bark sheets at roof level 0 Xx 0 0 0 0 skins used in roof construction 0 ? 0 0 0 0 0 all the houses consist of a hard-packed, scaly, dark layer from which overlying material separated easily. Near the center of each house is a hearth characterized by discolored earth and a wood ash con- centration. In four houses (H-1, H-8, H-4, H-6) one or more fire- darkened rocks found in or near the fireplace suggest that at one time those hearths may have been clearly delineated by stone borders. Several structures have pits of varying depths in the floors, and one such pit (H-2) is lined with birch bark. Fic. 4. House 1. 23 24 FIELDIANA: ANTHROPOLOGY, VOLUME 60 Surprisingly, wall benches are not a characteristic feature of Aku- livikchuk houses. Three dwellings (H-8, H-4, H-6) have back benches only and in one of these (H-6) the bench was covered with dried grass. In house 1 there are no actual benches, but dried grass was found on the floor along the side walls. The manner in which roofs were supported is a rather consistent feature of the houses. Four structures (H-2, H-3, H-5, H-6) have the remains of vertical posts almost equidistant from the side walls. These posts presumably supported four horizontal beams with short roofing posts or planks stretching from these beams to the uppermost wall log on all four sides. This four-post-center type of roof con- struction may also have been present in the other three dwellings (H-1, H-4, H-7), but the presence of additional post fragments makes it possible that some other form of roofing, perhaps six-post-center, was used. At the level of the roof in one house (H-2) badly decayed sheets of birch bark were recovered suggesting that this material was placed between the roof logs and overlying sod as a waterproof pro- tective cover. At the roof level in the same structure were sizeable deposits of maggot cases indicating that caribou skins or moose hides may have been similarly used. House 1 (Fig. 4) The main room of this structure, the largest dwelling on the site, is virtually square with horizontal wall logs held in place by vertical supports. There appear to have been at least three and probably four wall logs along each wall, but no more than two were found. In the back corners are small piles of sterile sand which may have been intended for spreading on the floor when it became dirty. The floor itself is a hard-packed, scaly, dark layer, thick and distinct over the entire surface. A centrally-located fireplace appears to have had stone retainers. The manner in which the roof of this house was sup- ported must remain conjectural, although four-post-center construc- tion is a possibility. However, there are additional posts, the struc- tural significance of which is not clear. There are no benches in this house, although a covering of dried grass was found on the floor along the side walls in the area where benches might have been ex- pected to occur. The tunnel is deeply excavated but its floor is not far below that of the main chamber. At the inner end of the tunnel is a semi- circular wooden sill, but the step up here into the living area is less than 20 cm. Tunnel wall construction consists of five horizontal logs VANSTONE: AKULIVIKCHUK bo Or , METER Hie. 5. House 2. on each side held in place by opposite sets of three vertical supports. At the outer end of the tunnel is a steep step up into the entryroom, the floor of which is well above the floor of the main chamber. There are three short retainer logs at this end, the lowest of which is in the form of a narrow step. The tunnel floor is a hard-packed, dark layer but less thick than the floor in the living area. This structure was 26 FIELDIANA: ANTHROPOLOGY, VOLUME 60 one of two where tunnel roof construction could be determined with some degree of certainty. Short, split logs were placed horizontally with the flat side down. These appear to have been covered with strips of birch bark before a sod covering was added. The entry- room into which the tunnel extended is well defined by horizontal wall logs on the four sides, but the floor of this area is indistinct and somewhat lower in the center than on the sides. House 2 (Fig. 5) Wood preservation in this structure, although the poorest in any on the site, was sufficient to indicate the main features of construc- tion. Wall logs are located along the sides only and there is just one on each side. Nevertheless, it is clear that these walls were con- structed of horizontal logs held in place by vertical supports. The back and front walls had to be located by following the floor to its outer limits. It appears that the roof was supported by four center posts. During excavation many large strips of birch bark were en- countered at the roof level and these may have covered the roof logs under the sod. In places there were large deposits of maggot cases indicating that hides may also have been utilized in roof construction. The floor of the main chamber is a hard-packed, scaly, dark surface and there appeared to be several floor layers but these could not be separated consistently over the entire living area. Therefore, no attempt was made to separate artifacts from the various floors. Excavation of the floor was complicated considerably by the exist- ence of numerous pits, only one of which is shown in the house plan. This pit, approximately 40 cm. in diameter and 40 cm. deep, is cir- cular and lined with strips of birch bark. It contained bones and a few artifacts, mostly beads. A large, thick, central fireplace consists of discolored earth and charcoal fragments. The tunnel of this house is notable for its depth and the fact that there is no real cold trap. The floor of the main chamber slopes up slightly near the tunnel while the tunnel floor itself slopes gently downward and then rises slightly just before the step up into the entryroom. The tunnel floor is simply a continuation of the hard- packed floor in the main room. Log preservation is poor, but enough remains to indicate that the tunnel walls were lined with split vertical planks held in place by one or more horizontal logs. There was prob- ably one such log running along the top of the planks and another one in a similar position at the bottom. At the step up into the en- tryroom are four horizontal planks held in place on each side by a 28 FIELDIANA: ANTHROPOLOGY, VOLUME 60 vertical support. The entryroom itself contains no structural fea- tures of any kind and has a poorly defined, darkened floor that is low in the center and slopes up along the sides. Just at the tunnel en- trance is a slight, semi-circular depression. House 3 (Fig. 6) This generally well-preserved structure has a main room with the usual horizontal wall logs held in place by vertical supports. There were probably at least three logs on all four sides, but only a single one remains along each wall. The most striking feature of this house is a high bench which extends along the back wall. This bench is faced with two horizontal retaining logs which pass in front of the back two supports of the four which supported the roof. Floor in this structure is the usual hard-packed, dark layer and there are a num- ber of deep pockets of occupation debris along the side walls. The central fireplace contains a number of fire-cracked rocks as well as ash deposits and discolored earth. The tunnel of this house, which is quite deep and long, is charac- terized by extremely good wood preservation. The floor is a well- defined dark layer and the walls consist of vertically-placed split log planks supported by a single long log on each side at the top of the planks. These horizontal retainers are in turn supported by upright logs. The inner end of the tunnel extends a short way into the main chamber where there is a step up with two short retaining logs. At the opposite end the tunnel extends into the entryroom and in this area there are two steps, each one with a horizontal retaining log along the outer face. There is also a similar log at the top almost level with the floor of the entryroom. House 4 (Fig. 7) This is another well-preserved structure and its main living area, like those of other houses on the site, is characterized by horizontal wall logs held in place with vertical supports. There were probably four such horizontally-placed logs along each wall, but no more than three are preserved on any one wall. As in house 8, there is a high bench along the back wall which is faced with two large horizontal logs. This facing is retained by four posts, two or more of which may have been associated with roof support. Four- or six-post-center roof construction is possible for this house, but because of those posts along the bench and the seemingly indiscriminate location of others, no clear pattern emerges. There is a central fireplace containing a Fic. 7. House 4. 29 30 FIELDIANA: ANTHROPOLOGY, VOLUME 60 large number of small fire-cracked rocks. The floor of the main cham- ber is a relatively even layer of the familiar hard-packed, scaly, black material. The long tunnel extends into the main living area and then goes off at a slight angle. Wall construction consists of vertical planks held in place by a single horizontal log along each side at the top. These in turn are held in place by vertical supports. This tunnel is quite deep and has a hard-packed, dark floor. At the inner end is a wooden sill that sits flush with the floor and in front of which are three short horizontal retaining logs that form a single narrow step. At the outer end of the tunnel are four vertical planks. In front of these are two short horizontal logs which also form a narrow step. In this structure there is clear evidence for the type of tunnel roof construction previously noted for house 1 and probably characteristic of all the structures. House 5 (Fig. 8) This is asquare house, the main chamber of which is characterized by the usual horizontal wall log construction. There were apparently four logs along each wall, although this number was preserved on only two sides of the main room. The roof was probably supported by four posts, although one of these appears to be considerably out of line. A central fireplace contains much charcoal and discolored earth, while the floor itself is no different from that described for the other structures. It should be noted, however, that the excavation for this house is shallower than that for any of the others, and the overburden less thick. The tunnel, which is short, extends well into the main chamber and is very well preserved. There are four horizontal logs along each wall held in place by vertical supports at the inner end. As retainers at the step up into the main chamber, there are three short horizontal logs. One is flush with the house floor and extends outward on either side of the tunnel. The second below it protrudes slightly and the third enough to forma step. At the outer end of the tunnel are four horizontal split logs forming a retaining wall. The floor is a well- defined dark layer, lowest in the center but sloping up slightly at either end. House 6 (Fig. 9) This house, cut deeply into the bank of the draw, is also the most deeply excavated of all the structures. There was a great deal of VANSTONE: AKULIVIKCHUK 31 ee ee ee | IMETER Fic. 8. House 5. overburden to be removed before features of construction were re- vealed. Horizontal wall logs, held in place by vertical supports, are present along three walls of the main chamber. Only three logs were used along the back wall, while there were apparently four along each side. The front wall on both sides of the tunnel consists of ver- tical split logs dug into the floor at the bottom and supported by a horizontal log at the top. The split logs are badly decayed but they appear to be approximately 15 em. wide and set in place with the flat side facing inward. The central fireplace in this structure con- 32 FIELDIANA: ANTHROPOLOGY, VOLUME 60 =) ——~ io) 0 fe) @) u ——— O} fe y) | (\ — 000 ° | to oO | fe | | Z | | | i ! a | | {| | \ | \ g <) \\ |O —s> | | | | _——SS————————— loc >| ,||SSSSSsesesse)| . (p foo | | | | Be a ee | | = = 0 no j | } le) fo) | re) | 5 8 | | =) = : {METER Fic. 9. House 6. tains the usual charcoal, discolored earth and a few fire-cracked rocks. Along the rear wall is a low, narrow bench rising no more than 15 cm. above the floor. It appears to have extended out as far as the back two posts of the four that supported the roof. The floor is the usual hard-packed, scaly, black layer and this surface also occurs on the bench along with indications that dried grass may have been laid over it. In places the floor surface extends into deep pockets, par- ticularly along the sides near the wall logs. The tunnel of this house, like that of most of the others, extends well into the main chamber. On either side there are horizontal logs VANSTONE: AKULIVIKCHUK 33 supported by vertical ones, and separate short logs form that part of the tunnel inside the living areas. This tunnel extends no more than 25 em. below the house floor level and, therefore, hardly can be said to form a cold trap. There is a single log retainer at the inner end, while at the outer end there is a similar arrangement involving two short logs. The tunnel floor is the same hard-packed, dark layer as the house floor. House 7 (Fig. 10) Like the structure just described, this house is also cut well back into the bank. However, wood preservation is very poor and wall logs are located along one wall only. These indicate horizontal log wall construction with the logs being held in place by vertical sup- ports. One corner apparently was cribbed, perhaps for the purpose of utilizing shorter wall logs. The cribbing was supported by small, vertically-placed stakes. The floor is a hard-packed, grey to black layer which in places is thick and scaly but elsewhere exceedingly thin and difficult to follow. It is also very uneven. The manner of roof construction is virtually impossible to determine because of the large number of posts and their seemingly random placement. A central fireplace consists of burned earth and charcoal fragments. Log preservation in the tunnel of this dwelling is much better than elsewhere. There are four horizontal logs on either side held in place by vertical supports. The tunnel is quite deep and extends into the main chamber where there are three short horizontal retain- ing logs. At the outer end, two such retainers are still in place and there may have been a third. When this structure was excavated, the tunnel logs were found to be pushed well toward the center and the drawing restores them to their original position. The floor is the same well-defined, hard-packed, dark layer that was characteristic of much of the floor in the main chamber. KASHGEE As might be expected, by far the most imposing structure in the village is the centrally located kashgee (K), or ceremonial house (Fig. 11). It is impressive not only because of its size, but also its depth. Unlike kashgee 1 at Tikehik (VanStone, 1968b, pp. 253-255), the exea- vation is directly down into the ground rather than cut into a steep hillside. Wood preservation in the Akulivikchuk kashgee was re- markably good and nearly all features of construction can be deter- 34 FIELDIANA: ANTHROPOLOGY, VOLUME 60 fe) N NN ee ————S) si — \|b —— lo} 12) fo) (e) e 4 —, ra = ° I METER 9 aS je) Fic. 10. House 7. mined with some degree of certainty. The main chamber is charac- terized by horizontal wall logs, probably held in place by vertical supports although only two of the latter were encountered (Fig. 12). Three of the four corners are cribbed, but the actual number of wall logs along each side of the main chamber cannot be determined defi- nitely. There were apparently at least four since this number was preserved along one wall. Single, long wall logs appear to have been used almost exclusively, although the condition and preservation of the remaining logs is such as not to exclude the possibility that two 36 FIELDIANA: ANTHROPOLOGY, VOLUME 60 were joined at some point along a wall. There are wide bench areas on all four sides and all seem to have been covered with wooden planks, the only example of this kind of bench covering at the site. The two side benches have only a pair of very wide planks near the walls, while the front and back benches are covered with a number of horizontally placed planks. The deep center of the main chamber has a log framework that runs all around it, probably four horizontal logs on the two short sides and just one each on the long sides where a large number of sizeable rocks appear to be a structural feature. These rocks, which may also have been used during sweat baths in the kashgee, seem to have served as fill to help support the front and back benches. The center area between the piles of large rocks is filled with a thick, sterile layer of charcoal and ash, and the retaining logs are badly charred. In fact, some of the wall logs are blackened too, suggest- ing that the structure was used extensively as a bathhouse. The kashgee tunnel enters the main chamber at the level of the front bench. Wood preservation in this area is poor, but it appears that there were vertical planks along the walls. These may have been held in place by horizontal supports as in some of the house tunnels. On the other hand, they may have been placed in such a manner as to be self-supporting. The tunnel enters the entryroom at floor level which is approximately the same as bench level in the main chamber. The entryroom presents something of a puzzle. Its walls consist of horizontal logs, probably three on a side and presumably supported in position by vertical posts. The floor of the entryroom tunnel runs into the center of the room and is framed at the inner end by a log at floor level on the end and two wide planks in the same position on either side. At this inner end of the tunnel are a number of large, fire-darkened rocks and a small hearth. One possible explanation for this arrangement is that the entryroom might itself have been used on occasion as a bathhouse. At the outer end of the tunnel is a single horizontal retainer. Floor in the entryroom is a dark grey layer, but in the entryroom tunnel it is distinct, hard-packed and dark. In the regular tunnel the floor is also hard-packed and dark. In fact, even on the plank-covered benches of the main chamber a certain amount of scaly black layer was found over and in between the planks. The entryroom tunnel extends out at an angle and has horizontal wall logs. It is difficult to determine the manner in which the main chamber of the kashgee was roofed, but it is virtually certain that center posts “UOIJDIS-SSOID UL aabyspy JO JaquiVyo UIeJY “ZI ‘DI i WILAW | O 000 ‘e) 37 38 FIELDIANA: ANTHROPOLOGY, VOLUME 60 were not involved. Concentrations of collapsed short logs in two cor- ners suggests cribbing, a method of roof construction for ceremonial houses that has been reported elsewhere in southwestern Alaska (Nelson, 1899, p. 250; Lantis, 1946, fig. 5). It will be recalled that cribbed wall logs occurred in three corners and this may be an indi- eation of cribbing that extended upward as part of the roof struc- ture. Both tunnels were apparently roofed with short, horizontally arranged split logs placed with the flat side down. MIDDEN A careful examination of the Akulivikchuk site at the beginning of the field season revealed a single obvious midden deposit directly in front of the kashgee (T-1). Prior to excavation it appeared as a rec- tangular, grass-covered hump. Sod was stripped from an area 8 m. wide and 9 m. long and excavation was begun in arbitrary 15 em. levels in hopes of distinguishing significant differences in the nature of artifacts and artifact distribution at various levels. After the entire cleared area was excavated to a depth of 15 em. below the sod level, it became clear that approximately half of it, that half closest to the kashgee entryroom, was not midden debris at all. People using the kashgee had apparently walked forward and tossed their rubbish over a small, natural bank. Therefore, actual midden material occurred only in the lower half of the cleared section and further excavations were confined to this smaller area. Here the midden was deepest in the center where excavations were extended to a depth of slightly more than 1 m., and thinned out at the sides where the sterile under- lying sand was encountered at approximately 40 cm. As expected, the midden consisted almost exclusively of bone and antler fragments along with considerable charcoal and pieces of wood. Most of this material was doubtless the by-product of baths and the various manufactures that took place in the kashgee. The artifact collection is not extensive and it proved impossible to distinguish significant artifact differences by level. Therefore, the small col- lection of midden artifacts has been considered as a single unit with materials from the structures on the site. In conclusion it should be noted that there were apparently no residential middens at Akulivikchuk and although extensive testing was not attempted, it is most unlikely that cultural debris exceeds 60 em. in depth anywhere on the site except in front of the kashgee. Even here, where the accumulation was clearly visible before exca- vation, the extent of the deposit is modest indeed. It seems likely, VANSTONE: AKULIVIKCHUK 39 as we have noted, that the residents of the village threw most of their debris either over the riverbank or into the ravine where it was even- tually washed into the river. What little did accumulate around the houses would perhaps have been dispersed by dogs and by the people themselves. The total span of occupation at Akulivikehuk was not sufficient for the accumulation of any significant amount of midden debris around the houses. This has also been found to be true for other nineteenth century village sites in southwestern Alaska (Os- walt and VanStone, 1967, pp. 28-24; VanStone, 1968b, p. 260). The kashgee would be an exception because of its use as a workshop by the men of the village. Collections In this chapter the artifacts recovered from the Akulivikchuk site will be described under two headings: locally manufactured goods and imported manufactured goods. Within these two broad cate- gories, further subdivisions will be made on the basis of the material from which the various artifacts were made. Although no exhaustive comparative treatment will be attempted, some comparative data derived from those published and unpublished reports dealing with sites which are closest, both spatially and temporally, to Akulivik- chuk will be included with the descriptions when considered relevant. Comparative generalizations and conclusions derived from them will be reserved for a later chapter. LOCALLY MANUFACTURED GOODS Under this heading are included those artifacts presumably manu- factured at the settlement, or at nearby settlements, by Eskimos. This would also include, of course, any objects made by Eskimos and traded to the people of Akulivikchuk, although it should be stated at the outset that proof of such trade is lacking. The artifacts to be described here, then, are what we might call traditional Eskimo forms and they are made from materials readily available in the local environment such as stone, bone and antler, birch bark, ete. Both the forms and the materials were, of course, known to pre-contact Eskimos of the Nushagak River region and therefore we can say that the artifacts about to be described represent a continuity of material culture stretching back into the prehistoric period. But there are two other categories of artifacts to be dealt with here. The first includes forms made locally from exotic materials introduced into the area by Europeans and may be characterized as an attempt to per- petuate traditional forms in new materials. The second represents forms that were direct products of the contact situation but have been manufactured from traditional, locally available materials. 40 PLATE 1. Stone Artifacts. 1—9. Whetstones (pp. 42-43); 10. Projectile point (p. 42); 11. Whetstone (p. 43); 12. End scraper blade (p. 42); 13. Snub-nosed scraper (p. 42); 14-16. Whetstones (p. 43). 41 A2 FIELDIANA: ANTHROPOLOGY, VOLUME 60 Chipped and Ground Stone These two methods of working stone are considered together partly because there are virtually no true chipped stone implements in the collection, but also because many stone artifacts show that both chipping and grinding techniques were utilized in their manu- facture. Evidence for the chipping of flinty materials at Akulivikechuk is extremely limited. There is one large nodule of blue chert, approxi- mately 7.5 em. long and 5 cm. wide, from which a number of flakes have been removed, and eight flakes of the same material, two of which show some retouching. In addition, four flakes of opaque chalcedony were recovered. There are only three finished flint arti- facts: a fragment of opaque chalcedony with a steep, carefully pre- pared working edge that may have been hafted as a snub-nosed scraper (Pl. 1, 13); a blue chert flake retouched along one edge, per- haps as a hafted end scraper blade (Pl. 1,12), and the tip of a small pro- jectile point of blue chert (Pl. 1,10). Although chipped flint artifacts are characteristically early in Alaska, they are also widely distrib- uted in late sites including two nineteenth century Eskimo settle- ments in southwestern Alaska (Oswalt and VanStone, 1967, pp. 26- 27, pl. 1, e-f, h-i; VanStone, 1968b, pl. 1, 1-7). There seems little reason to doubt, therefore, that the small number of flint artifacts and chips from Akulivikchuk are indigenous and contemporary with the rest of the collection. A single hammerstone is simply a large water-worn pebble of ba- saltic material which has had a number of rough flakes removed through use. It is 15 em. long and 8 cm. wide near the distal end. There are 40 whetstones in the Akulivikchuk collection and for descriptive purposes these have been divided into three types based on the nature of the stone from which they are made. Those speci- mens belonging to type 1, of which there are two, are made of very fine-grained filite (Pl. 1, 4). Both are rectangular, water-worn peb- bles that have simply been picked up and used as whetstones and they show wear on one surface only. The smallest of the two imple- ments is illustrated; the larger is of a similar shape and is 15 em. long, The seven whetstones of type 2 have been shaped of very fine- grained siltstone (Pl. 1, 2-3). Three specimens show wear on all sur- faces and tend to be worked to a rectangular form. The illustrated specimens are typical in size, although one broken whetstone of this VANSTONE: AKULIVIKCHUK 43 type was obviously much larger, the fragment measuring 14 em. in length and 4.5 em. in width. By far the majority of whetstones, 31 in number, belong to type 3. They are made of fine- to medium-grained sandstone and a variety of sizes and shapes is represented. The very fine-grained specimens, of which there are 14, tend to be small, light colored and show wear on all surfaces. Of these, all are fragmentary and the illustrated specimens (PI. 1, 1, 5-7, 11) show the range of shapes and sizes. The fine- to medium-grained sandstone specimens belonging to type 3 show a much greater variety of size and shape. All but two are fragmentary and as a group they tend to be broad, flat and worked on two or more surfaces. The range in length of these frag- ments is from 4 to 17 em. and in width from 2 to 10 em. The illus- trated specimens are typical and include the two complete whetstones of this type (Pl. 1, 8-9, 14-16; Pl. 2,12). Three specimens belonging to type 3 have narrow V-shaped grooves in them which indicate cut- ting with a stone saw (Pl. 1, 5). Although no such saws have been identified in the collection, it is likely that this implement was used for the initial shaping of some whetstones and doubtless for many other stone working purposes as well. A single type 3 whetstone (Pl. 1, 16) has reddish colored marks on one side indicating that it was used as a surface on which to grind red ochre. All the whetstones described were held in the hand and presum- ably used for shaping and sharpening small stone, bone and metal objects. In addition to these artifacts, the collection also includes 11 large, flat grinding slabs of medium-grained sandstone which have been roughly shaped in a rectangular form and show wear on one sur- face only. The largest of these is 33 cm. in length and 14 em. wide at its widest point. Most specimens, however, are approximately 20 by 15 cm. It is probable that some of these flat slabs were used for the grinding, polishing and sharpening of large adz and knife blades. Informants indicated, however, that the more fine-grained specimens were simply stones which women used to sharpen their ulus when splitting fish. Such stones are used in a similar manner in the Nushagak River villages today, the women whetting their knives occasionally as they work. Ten fragmentary slate end blades all exhibit finely ground sur- faces and bilateral cutting edges. Of this number, four large tip fragments represent more than half of the complete blades and are characterized by a hollow-ground groove running parallel to the en- 4 5 6 4 ; cenhmeters PLATE 2. Stone Artifacts. 1. End blade blank (p. 45); 2. End blade (p. 45); 3. End blade blank (p. 45); 4-6. End blades (p. 45); 7-8. End blade blanks (p.45); 9. End blade (p. 45); 10-11. Ulu blades (p. 45); 12. Whetstone (p. 43); 13-14. Ulu blade blanks (p. 45). 44 VANSTONE: AKULIVIKCHUK 45 tire length of the blade (PI. 2, 2, 5-6, 9). Of the remaining six frag- ments, four show the hollow-ground groove. In addition, there is a single blade that appears to be complete except for final polishing (Pl. 2, 4). It has a flat base where primary chipping is visible and the hollow-ground grooving, presumably the final step in the manu- facturing process, had just begun. This specimen, as well as all the fragments just described, are of a size to suggest their use as arrow or lance blades. The 15 slate end blade blanks vary in length from 3 to 7.5 em.; some are probably fragmentary but it seems certain that all would have been finished as arrow points (Pl. 2, 1, 3, 7-8). The nature of the workmanship exhibited by these blanks suggests that a piece of slate was chipped to roughly the desired size and shape, and then finished by grinding first the flat surfaces and then the working edges and base. Ground slate end blades with hollow-ground grooves are reported from a number of sites in southwestern Alaska including Tikehik (VanStone, 1968b, pp. 267-268, 270), Crow Village (Oswalt and VanStone, 1967, p. 29, pl. 2, g-h), Togiak at the mouth of the river of that name (Kowta, nd.) and along the Yukon River (deLaguna, 1947, p. 151, pl. XX VI, 36-41). This form is probably related to blades with a very thin and sharply cut triangular mid-section such as have been found at the Pavik (Paugvik) site at the mouth of the Naknek River (Larsen, 1950, p. 178, fig. 55, A 14, 15) and on Kodiak Island (Clark, 1966, fig. 9, p). Another related form which is tri- angular and hollow-ground occurs at the Uyak site on Kodiak Island (Heizer, 1956, pl. 46, n). None of the eight slate ulu blade fragments in the collection is complete enough to suggest the shape of the entire blade or to indi- cate the manner of hafting. All have convex cutting edges, there being some degree of variability in this convexity. The two largest fragments are illustrated (Pl. 2, 10-11), one of which has a beveled cutting edge and a drilled hole, possibly as an aid to hafting. The five slate ulu blade blanks are all complete or nearly so. They are roughly flaked and the final form of the blade is clearly suggested. In addition to being finished with a convex cutting edge, it is likely that none of these blades would have been tanged, but would rather have had flat or convex proximal surfaces for insertion into the han- dle (Ply 2513-14): 46 FIELDIANA: ANTHROPOLOGY, VOLUME 60 By far the largest category of stone implements is that of scraper or planing adz blades of which there are 538. For convenience of de- scription, these have been divided into four types on the basis of the nature of the stone from which they were made. The largest number of complete and fragmentary specimens belong to type 1 and are made of medium- to fine-grained sandstone. Of the 34, 16 are com- plete or nearly so, and 18 fragmentary. The type 1 blades are crudely fashioned except for the working edges which are finely ground and V-shaped in cross-section; 12 are slightly concave on one surface. Of the complete or nearly complete specimens, all but two taper toward the proximal end for insertion into an antler or wooden han- dle (Pl. 3, 1-8, 5-7, 9-10, 13-14); the other two have straight sides (Pl. 3, 11, 12). Nine of the fragmentary blades also show this taper while the remainder are not complete enough to indicate their shape. Complete blades of type 1 vary in length from 5 to 13 em., while the fragments are from 8 to 8 cm. long. Seraper or planing adz blades belonging to type 2 are, for the most part, similar in form to those just described but they are made from a slightly metamorphosed sedimentary rock like a schist that has a pronounced metallic sheen and a tendency to exfoliate in hori- zontal layers. There are 13 specimens of this type, nine of which are complete or nearly so. All are roughly worked except for the cutting or scraping edge and taper toward the proximal end (PI. 3, 4,8). Three of the fragmentary blades also show this taper. All the type 2 blades have working edges that are roughly V-shaped in cross- section, five being slightly convex on one side. A single specimen has a pronounced beveled edge on both sides. The complete blades range in length from 6.5 to 11 em. The four blades belonging to type 3 are all complete and made from a very fine-grained siltstone (Pl. 4, 1, 3-4, 8). They are some- what more carefully finished than those of types 1 and 2 and all have V-shaped working edges. Three specimens taper toward the prox- imal end and one of these is much flatter than the others (Pl. 4, 4). The fourth blade is unique in the collection being carefully worked and polished over its entire surface and with a working edge at each end (Pl. 4, 8). Such a blade might have been hafted without an antler head so that either working end could be used, or the two working edges could have been used successively, the blade being turned in the head as necessity required. Type 4 blades, two in number, are of basalt and taper toward the proximal end. One specimen shows a minimum of workmanship and PLATE 3. Stone Artifacts. 1-14. Seraper or planing adz blades (p. 46). 47 48 FIELDIANA: ANTHROPOLOGY, VOLUME 60 is essentially just a basaltic chip with a thin, ground, V-shaped work- ing edge (Pl. 4, 9). The other is a fragment that appears to have been only roughly worked over most of its surface. It has a polished and steeply beveled, V-shaped working edge (PI. 4,5). This blade is convex along one side and flat on the other except where the bevel occurs. Seraper or planing adz blades similar to those from Akulivikchuk have been described from a large number of archaeological sites in southwestern Alaska and are also present in ethnographic collections. In the past a number of writers, myself included, have attempted to differentiate between skin scraping blades and planing adz blades on the basis of the material from which they were made. Those made of a hard material could be either scraper or adz blades, but adz blades were never made of soft stone (deLaguna, 1947, p. 186; Os- walt, 1952, pp. 57-58, 61; VanStone, 1968b, pp. 270-271). While there is doubtless some validity to this distinction, it has been de- cided in the present context not to distinguish between the two forms on this basis, particularly since it is virtually impossible to draw a dividing line between “‘hard”’ and “‘soft’”’ stone. Any attempt to do so would be meaningless. Apart from any distinction based on ma- terial, it would seem that the majority of the blades described above were used in small adzes as wood working tools. Of the total of 53 specimens, 30 were found either in the kashgee or in the midden (T-1) directly in front of it. Wood working adzes would have been used by men, most likely in the kashgee, while a skin scraper was a woman’s tool. Those blades most likely to have been used as scraper blades may be the beveled specimens and possibly those that are broad and flat. The scraper or planing adz blade blanks in the collection can be described on the basis of the same four types as the complete speci- mens. Of the 17 blanks belonging to type 1, all but two are complete enough to show that the completed implement would have tapered toward the proximal end. Most of the blanks have been worked sufficiently so that little remained to be done except grind and polish the working edges. A variety of sizes is illustrated (PI. 4, 2, 7, 11, 13). A single specimen was apparently used secondarily as a grinding slab for some bluish substance, traces of which can be seen on the surface near the distal end (Pl. 4, 11). Three of the five type 2 blanks are unbroken and do not differ substantially in shape from type 1 (PI. 4, 10). The two type 3 specimens appear to be finished except for grind- ing the distal ends (Pl. 4, 6, 12), while the single type 4 basalt blank PLATE 4. Stone Artifacts. 1. Scraper or planing adz blade (p.46); 2. Scraper or planing adz blade blank (p. 48); 3-5. Scraper or planing adz blades (pp. 46-48); 6—7. Scraper or planing adz blade blanks (p. 48); 8-9. Scraper or planing adz blades (pp. 46, 48); 10-18. Scraper or planing adz blade blanks (p. 48). 49 50 FIELDIANA: ANTHROPOLOGY, VOLUME 60 is little more than a roughly-flaked nodule, the identification of which should be considered tentative (Pl. 5, 6). It will be noted from the trait list that the majority of the blanks, like the complete blades, were recovered from the kashgee or its midden. The double-grooved splitting adz head is represented in the Akuli- vikehuk collection by three fragments and one unfinished specimen, all made from basalt. Two of the fragments are large and appear to have had their distal ends broken through use. Both are relatively unworked except where the lashing grooves have been pecked out in such a manner as to leave a pronounced unworked ridge between them. One specimen has a rather large flake removed from the prox- imal end suggesting that a blow was struck at this point (PI. 5, 7). The third fragment is very small and can be identified as part of a splitting adz only because part of one groove is visible. The unfin- ished specimen would apparently have been worked over most of its surface. Shallow grooves are visible and the working edge has been roughly shaped (Pl. 5, 9). The double-grooved splitting adz has a fairly wide distribution in late sites in southwestern Alaska, being found at Crow Village (Oswalt and VanStone, 1967, p. 29, pl. 2, k), Nanvak Bay (Ackerman, 1964, p. 32, fig. 15, 2), Kachemak Bay (deLaguna, 1934, pp. 56-57, pl. 18, 2), along the Yukon River (de Laguna, 1947, pp. 121, 150, pl. X, pl. XXV, 22, 25; Nelson; 1899) pl. XX XIX, 3), Prince William Sound (deLaguna, 1956, pp. 110- 117, pls. 10-11), and on Kodiak Island (Heizer, 1956, p. 44, fig. 26). A curious object similar in size to the splitting adz heads and also of basalt is what appears to be a form of wedge (Pl. 5, 8). The im- plement has been worked rather carefully on one side in such a man- ner that it tapers to a wedge-shaped tip at the distal end. The opposite end is completely unworked. Another unusual and fairly abundant category of artifacts in the collection has been identified as paint mortars or palletes, or possibly pigment or tobacco grinders. Actually, most of these are not artifacts at all in the accepted sense, but rather natural forms that were used, or were collected for the purpose of being used, for the above-men- tioned purposes. Three of these are large, irregularly shaped boul- ders of fine-grained sedimentary rock from which large flakes or spalls have been removed, probably as a result of frost action or heat, to leave depressions or flattened areas. On two specimens this flattened area is darkened from use and a third clearly shows the remains of red ochre (Pl. 6, 5). The two larger specimens, not illus- trated, measure 20 by 24 em. and 20 by 10 cm., respectively. PLATE 5. Stone Artifacts. 1-38. Mortars, grinders or palletes (p. 52); 4-5. Bullet molds (p. 56); 6. Scraper or planing adz blade blank (p. 50); 7. Splitting adz head (p. 50); 8. Wedge (?) (p. 50); 9. Splitting adz head (p. 50). 52 FIELDIANA: ANTHROPOLOGY, VOLUME 60 The above-mentioned spalls have also been utilized for similar purposes. Most of these are probably the result of fire action and are perhaps from rocks that formed house fireplaces. Often these spalls are extremely deep with steep sides and make excellent con- tainers. There are eight objects of this type, most made of sedimen- tary material but at least one of igneous rock. The three illustrated specimens are typical, one showing a considerable deposit of red ochre (PI. 5,1-3). These specimens range in length from 4.5 to 14 em. Four of these mortars or grinders are small pieces of sedimentary conglomerate out of which an inclusion has weathered. None show indications of use, but all were almost certainly collected with the idea of use in mind (Pl. 6, 3). Two small sandstone pebbles with recessed surfaces (Pl. 6, 6) and another fragment of fine-grained sandstone with circular recessed surface are also included in this group. The latter shows a thick deposit of red ochre (PI. 6, 2). The only specimens in this category which appear to have been worked by man are a small, rectangular pebble with a circular depres- sion ground into one surface (Pl. 6, 4), a much larger, irregularly shaped metamorphic boulder with a large, deeply ground cavity to the sides of which adhere particles of an unknown substance (PI. 6, 7) and a more carefully worked oblong slab with a ground flat surface and traces of red ochre (Pl. 6, 1). In addition to being used as paint palletes or pigment grinders for red ochre and similar substances, it is possible that objects sim- ilar to these might also be used for grinding birch fungus for adding to tobacco. Eskimos in the area today frequently mix their tobacco in this manner. Similar grinders or mortars occur in other sites in southwestern Alaska, notably on the Yukon (deLaguna, 1947, p. 221, pl. XXI, 2, 4), at Crow Village (Oswalt and VanStone, 1967, p. 20), Tikchik (VanStone, 1968b, p. 271, pl. 3, 2), and in Prince Wiliam Sound (deLaguna, 1956, pp. 141-148, pls. 22, 3; 23, 6). Specimens from other sites probably have been described as lamps. A flat, roughly worked, oval object of a coarse-grained meta- morphic rock has been tentatively identified as a platter, although it may also have served as a grinder. It is slightly discolored on one surface but otherwise shows no signs of use. The specimen is 25 cm. long and 18 em. wide. Two stone lamps were recovered, one of which is complete and the other a large fragment representing about three-quarters of an entirelamp. The complete specimen, made of medium-grained sand- PLATE 6. Stone Artifacts. 1-7. Mortars, grinders or palletes (pp. 50, 52 53 54 FIELDIANA: ANTHROPOLOGY, VOLUME 60 stone, is characterized by a wide and flat rim, a relatively shallow bowl, evenly rounded exterior and wide, concentric wick channel cut out of the flat rim (Pl. 7,2). This type of lamp, as pointed out by Hough (1898, p. 1054, pl. 20) is typical of Kodiak Island. Heizer also notes that Hrdliéka collected two lamps of this type from living Koniag and the form occurs exclusively in the upper level of the Uyak site (Heizer, 1956, p. 33, pl. 26, c). Similar lamps are illus- trated by deLaguna (1934, pl. 24, 1; 1956, pl. 24, 4) from Kachemak Bay and Prince William Sound; a fragment of a lamp of this type occurs in the collection from Crow Village (Oswalt and VanStone, 1967, pp. 30-81, pl. 2, o). The fragmentary lamp is a large, ovoid specimen characterized by a narrow, beveled rim, broad, flat bottom, and a longitudinal groove in the center of the bowl (Pl. 7,1). The three fragments of this lamp were found in two different houses but in the trait list the specimen is recorded as coming from house 4. This Akulivikchuk specimen somewhat resembles the Type IB2 form from the Uyak site where it occurs in the lower level (Heizer, 1956, pp. 34-85) and lamps with longitudinal grooves also occur in Kachemak Bay where deLaguna (1934, p. 65) has noted that its purpose was probably to direct the flow of oil to the wick. Since pottery lamps were well established throughout the Nusha- gak and Kuskokwim river regions in the late prehistoric and early historic periods, the finding of stone specimens at Crow Village was explained by suggesting that they were not indigenous to the site, but had been traded into the area, perhaps from Kodiak Island or other stone lamp areas to the south (Oswalt and VanStone, 1967, p. 31). Since the Crow Village report was written, however, oval stone lamps with flat rims have also been found at Kolmakovski Redoubt (Oswalt, personal communication), at Tikchik (VanStone, 1968b, p. 272, pl. 8, 1), and at the mouth of the Togiak River (Kowta, nd). In addition, the author has recently discovered that Gordon (1906, pl. 26, fig. 6) refers to lamps similar to the complete specimen described above as being in use as far north as Norton Sound at the beginning of the twentieth century. Also recently noted is the fact that Zagoskin (1967, p. 228) refers to the use of stone lamps by the peoples of interior southwestern Alaska at the time of his explorations in 1842-1844. It would seem, therefore, that stone lamps probably spread north into the pottery lamp area in the late prehistoric period from their center in the Kachemak Bay— Kodiak Island region. The author knows of no stone lamps reported PLATE 7. Stone Artifacts. 1. Stone lamp fragment (p. 54); 2. Stone lamp (p. 54). 55 56 FIELDIANA: ANTHROPOLOGY, VOLUME 60 from the Kuskokwim-—Nushagak area that do not resemble types illustrated either by Heizer (1956) or deLaguna (1934, 1956). The collection also contains three halves of bullet molds made from medium- to fine-grained sandstone. These bullet mold sections each consist of a prepared flat surface into which has been ground a circular depression. One specimen is very fragmentary but the other two are complete and their depressions are 1.4 cm. and 1.6 cm. in diameter, respectively. At one end of the depressions are grooves which, when the identical other halves of the molds were fitted to these, would permit the lead to be poured in (PI. 5, 4-5). The two halves of each of these molds would, of course, be lashed together before molten lead was poured and then after the metal had hard- ened, the halves could be separated and the completed balls removed. The point where the two halves of the mold joined would leave a raised burr around the circumference of the lead balls and these burrs are actually visible, to a greater or lesser degree, on lead balls re- covered from the site. It is possible that these stone molds were originally fitted into wooden or antler handles resembling those in ethnographic collections (see particularly Nelson, 1899, pl. LXITII, 8). Locally made bullet molds similar to these stone specimens from Akulivikchuk are likely to be found in increasing numbers in south- western Alaska and elsewhere as more sites belonging to the historic period are excavated. They have already been recovered from Crow Village (Oswalt and VanStone, 1967, p. 31, pl. 2, n) and the Tanaina Indian settlement of Kijik on Lake Clark (VanStone and Town- send, 1970, p. 62, pl. 11, 2). Bone and Antler Artifacts of bone and antler are common in the Akulivikchuk collection, there being 145 identifiable implements made from these materials along with two unidentified objects and a large number of worked fragments. Of the 14 complete and fragmentary net weights, 13 are made of antler and one of bone. All are roughly rectangular in outline with laterally drilled holes at each end for attachment to the net. With one exception, the antler weights are made from split sections of the material (Pl. 8,15). The single exception is more carefully shaped than the others, having a rounded, triangular form in cross-section, narrow at the top and thick at the bottom (PI. 8, 14). An unfinished specimen appears to be complete except that the line holes have not been drilled. The complete antler weights range in length from PLATE 8. Bone and Antler Artifacts. 1-3. Fishing ice picks (p. 58); 4. Sal- mon harpoon socket piece (p. 58); 5-9. Salmon harpoon dart heads (p. 58); 10. Lure-hook (p. 58); 11. Salmon harpoon socket piece (p. 58); 12. Salmon harpoon dart head (p. 58); 18-15. Net weights (pp. 56, 58). 57 58 FIELDIANA: ANTHROPOLOGY, VOLUME 60 14.5 to 23 em. The bone specimen has been fashioned from a sec- tion of moose or caribou rib. The line holes are broken out (PI. 8, 13). The single lure-hook is complete and consists of an antler shank in a stylized fish shape with a bent nail in the distal end as a barb. The proximal end has a small, drilled suspension hole that runs par- allel to the barb (PI. 8, 10). Three bone salmon harpoon socket pieces (P\. 8, 4,11) are all broken at the proximal end but a socketed tang is suggested for one specimen (Pl. 8,11). Each is drilled at the distal end to receive the dart head and two had small wooden insets in this hole which presumably served to wedge the tang of the dart head in place. Six pointed sections of antler are identified as fishing ice picks, although this identification should be considered tentative for at least some of these specimens. Five are simple split sections of antler worked to a point at one end (Pl. 8,1, 2). One is scored along the outer surface near the proximal end, presumably as an aid to hafting to a wooden shaft (Pl. 8,2). The sixth is more carefully shaped, be- ing worked to a roughly rectangular shape at the proximal end (Pl. 8,3). It may have been intended for use as a knife. The 15 antler salmon harpoon dart heads have been divided into three types on the basis of the shape of the line hole and the position- ing of the barbs. Nine specimens belong to type 1 and are charac- terized by a round, centrally located line hole and asymmetrically placed barbs ranging in number from one to three. Three have sharp shoulders and plain conical tangs (Pl. 8, 5), while the others have sloping shoulders and wedge-shaped tangs (Pl. 8, 8, 12). On one of these the tang is irregularly shaped and forms a pointed spur (Pl. 8,7). The five type 2 dart heads also are asymmetrically barbed with the number of barbs ranging from one to three, but they have rectangular, centrally located line holes. Two specimens have sharp shoulders and plain, conical tangs (Pl. 8, 9). One of these has a deeply incised line running upward toward the tip from the line hole. A very fragmentary specimen has a sloping shoulder and plain, con- ical tang, while another fragment is shoulderless. A third fragment is broken at the proximal end. There is only one specimen in type 3. It has two barbs on each side near the tip, rounded, oblong line hole, and a sloping shoulder with conical tang (PI. 8, 6). Two badly weathered antler objects are tentatively identified as side prongs for fish spears. One is a simple leister prong, asymmetri- cally barbed with a pointed tang. The other, probably unfinished as it is barbless, has an irregularly shaped tang designed to fit into a 20 PLATE 9. Bone and Antler Artifacts. 1. Fish spear point (p. 60); 2. Center prong for fish spear (p. 60); 38. Side prong for fish spear (p. 58); 4. Center prong for fish spear (p. 60); 5-7. Composite harpoon heads (pp. 60-61); 8. Un- barbed, socketed projectile point (p. 61); 9. Blunt arrowhead (p. 62); 10-14. Unbarbed, socketed projectile points (p. 61); 15. Arrowhead (p. 61); 16-17. Blunt arrowheads (p. 62); 18. Splitting wedge (p. 62); 19-20. Mesh gauges (p. 61). 59 60 FIELDIANA: ANTHROPOLOGY, VOLUME 60 A Fic. 13. a. Composite harpoon head. b. Unbarbed, socketed projectile point. slot near the proximal end of a wooden shaft (Pl. 9, 3). Two barbless prongs of antler may be center prongs for the two- or three-pronged fish spear. One is shoulderless and slopes to a pointed proximal end (Pl. 9,4). The other is more rounded and has a slot at the proximal end to facilitate hafting (Pl. 9, 2). The identification of this latter specimen should be considered particularly tentative. A fish spear point of antler has three barbs along one side and is tangless (PI. 9, 1); the specimen may be broken at the proximal end. A similar point was recovered from the Tikchik site (VanStone, 1968b, p. 275, pl. 4, 13). Both would presumably have been hafted to a long shaft; a spear of this type is illustrated by Nelson (1899, Pl. LX VII, 4). What have been referred to in the archaeological literature as composite harpoon heads are represented in the Akulivikchuk collec- tion by four finished and two unfinished halves. The term appears VANSTONE: AKULIVIKCHUK 61 first to have been used by Drucker (1948, p. 39, fig. 4) who found these two-piece spurred points along the northern Northwest Coast. Heizer (1956, p. 64, fig. 40) also reports such points as occurring in the lower levels of the Uyak site. The complete halves have short sockets and broad lashing grooves that encompass virtually the en- tire head except for the spur and the tip (Pl. 9, 6-7; fig. 18a). It seems likely that these heads were used with a very thin, narrow metal blade inserted between the halves, or perhaps with no blade at all. On the Northwest Coast and Kodiak Island, these harpoon heads were associated with sea mammal hunting and it may be that the implement was used by the residents of Akulivikchuk for a sim- ilar purpose during their trips to Nushagak Bay. The larger speci- mens do not seem to be adapted for fishing, but there is a small one that might have been so used (PI. 9, 5). This form has not hitherto been reported from sites in southwestern Alaska. Except for the previously described net weights, the only objects in the collection associated with netting are two complete and one fragmentary mesh gauges of antler. The two complete specimens have gauging distances of 9.3 cm. and 4.4 cm., respectively (Pl. 9, 19-20). Itis likely that the larger gauge was used in the manufacture of gill nets for salmon fishing, while the smaller may have been used in making seins or dip nets. There are four antler artifacts which have been identified as ar- rowheads but unfortunately they are all so fragmentary and poorly preserved that it is difficult to note their characteristics. One frag- ment seems to have had a single barb, sharp shoulder and conical tang (Pl. 9, 15), while another apparently had a number of small barbs along one side, a sloping shoulder, and conical tang. The re- maining specimens are so fragmentary that nothing definite can be said about them and their identification as arrowhead fragments should be considered very tentative. A very distinctive form is an unbarbed, socketed projectile point which has been described by Heizer (1956, pp. 64-65, figs.41-42) as typical of Kodiak Island and considered by him to be a local spe- cialization. There are six of these, all of antler, in the Akulivikehuk collection and they can be divided into two types. Three belong to type 1 and are characterized by a simple sharpened tip and a square- cut base with a short, wide, open socket (Pl. 9, 10, 13-14). The type 2 projectile points are in every way similar to type 1 except that they have a recessed lashing slot along one side (PI. 9, 8, 11-12, fig. 13b). 62 FIELDIANA: ANTHROPOLOGY, VOLUME 60 A socketed projectile point similar to these was discovered at the Tikchik site (VanStone, 1968b, p. 275, pl. 4, 12) where it was errone- ously identified as a fish spear point. Aside from this lone occurrence, the form has not previously been reported from anywhere in Alaska except Kodiak Island. Like the arrowheads described above, the five antler artifacts that have been identified as blunt arrowheads are also unfinished and poorly preserved. Two specimens have sharp shoulders and rela- tively conical tangs (PI. 9, 16-17), while a third is faceted and broken at the proximal end (PI. 9, 9). Of the remaining two, both very frag- mentary, one appears to be wedge-shaped at the distal end with a shouldered, conical tang, while the other is shoulderless and slopes to a blunted point at the proximal end. The 25 complete antler splitting wedges vary in length from 7 to 18.5 cm. and average about 12 em. With one exception, all are sec- tions of antler blunt at one end and tapering to a wedge-shaped tip at the other (Pl. 9, 18; Pl. 10, 11,15). None shows signs of extensive use. The single exception is made from a piece of split antler and has apparently been used as a beaming tool as it has a concave, worn edge along one side (Pl. 10, 12). There are also five antler wedge tip fragments. The single fragmentary skin scraper has been made from a eari- bou metatarsal cut longitudinally to form a sharp scraping edge which shows signs of use (Pl. 10, 16). The first type of awl, represented by three specimens, is made from slivers of antler, blunt or irregular at one end and sharpened to a point at the other (Pl. 10,10). The three awls of the second type are made of bone. Two specimens are fragments of caribou ribs worked to a point at one end (PI. 10, 4), while the other is a leg bone of some small animal with a narrow, flat distal end (Pl. 10, 5). The two crooked knife handles are formed from slightly curved pieces of bone. One specimen has the metal blade still in place (Pl. 10, 7). A curious feature of this handle is that the blade slit is 10.5 em. long or 3.5 em. longer than the blade. Perhaps the blade was pushed forward as it wore away at the distal end. The second handle, with a blade slit 7 cm. in length, also has another slit through the body of the handle parallel to and nearly directly opposite the blade sht (Pl. 19, 6). Extending from this slit toward the distal end of the implement are two engraved lines which may have been in- tended as decoration. The function of this second slit, which passes through the handle, is unknown. centimeters. PLATE 10. Bone and Antler Artifacts. 1. Storyknife (p. 64); 2. Pendant (p. 65); 38. Unidentified (p. 65); 4-5. Awls (p. 62); 6-7. Crooked knife han- dles (p. 62); 8. Engraving tool handle (p. 64); 9.Spoon (p.64); 10. Awl (p. 62); 11-12. Splitting wedges (p. 62); 13. Adz head (p. 64); 14. Spoon (p. 64); 15. Splitting wedge (p. 62); 16. Skin scraper (p. 62). 63 64 FIELDIANA: ANTHROPOLOGY, VOLUME 60 A tentatively identified artifact is an antler engraving tool han- dle. The specimen, which is badly weathered, does not appear to be complete as no slot is visible at the distal end for an animal tooth or metal blade. There are, however, a series of four raised bands running around the implement toward the distal end, a type of dec- oration that is characteristic of Alaskan Eskimo engraving tools ak, 10, S)e Although many stone adz blades have been described, the only evidence for the manner in which these blades were hafted is the pres- ence of a single, unfinished antler adz head. This is the characteristic form found in Eskimo sites throughout Alaska and in ethnographic collections. It is a section of antler with a blade slit at one end and pronounced lashing knob at the other. The inner side is flat to re- ceive a wood or antler handle which would be lashed to the head (Pl. 10, 13). The blade shit in this particular specimen is very nar- row and shallow suggesting that the head was intended to receive a small metal blade. There is one complete antler spoon which has a long, narrow bowl and ashort, straight handle (Pl. 10, 14). Another specimen is incom- plete but apparently had a shorter, narrower bowl and a longer han- dle which narrows toward the center and flares at the proximal end (Pl. 10, 9). A third spoon is represented by a fragment of a deep, oblong bowl. Two storyknives are made of curved, flat sections of antler which widen at the distal end to form a knife-like blade. Both are virtually the same length, but one is much better preserved than the other. The lower half of this specimen is decorated with two parallel lines running along one side. Inside these lines, an inverted chevron de- sign occurs at irregular intervals (PI. 10,1). It is curious that a story- knife with an almost identical decorative motif was recovered from the Tikchik site (VanStone, 1968b, p. 279). Ethnographically, the telling of stories illustrated by means of a storyknife appears to be confined to Eskimo girls living in southwestern Alaska. The north- ern distributional limit is in the vicinity of St. Michael, while to the south such knives are absent on Kodiak Island (Oswalt, 1964, p. 310). Archaeologically, storyknives have been recovered from Crow Vil- lage (Oswalt and VanStone, 1967, p. 40, pl. 7, j-1) and Hooper Bay Village (Oswalt, 1952, pp. 69-70, 80) as well as Tikehik. An abundant artifact type in the Akulivikchuk collection is sled shoe sections, of which there are nine of antler and 16 of bone. The VANSTONE: AKULIVIKCHUK 65 antler sections are all drilled with holes for pegging to the runner (Pl. 11, 6-7). These holes vary from 5 mm. to 1.3 em. in diameter and are irregularly spaced. The bone specimens are also character- ized by irregularly spaced pegging holes drilled 6 mm. to 1.3 em. in diameter, but one section shows, in addition to a pegging hole, two pairs of parallel holes joined by a groove for lashing the shoe to the sled runner (Pl. 11,1). The width of the unbroken sled shoe sections of both materials varies from 2.3 to 5.5 em. and many are thin, indi- cating considerable wear. The antler sections tend to be shorter, narrower, and thinner in cross-section than those made of bone. A single antler kayak shoe section was recovered. Such shoes were attached to the bow and stern of a kayak so that the cover would not tear as the boat was drawn up on the beach. This specimen has three drilled holes for pegging to the kayak frame (Pl. 11, 8). A bear tooth has a drilled hole at the proximal end indicating its use as a pendant or as part of a necklace (Pl. 10, 2). There are two seemingly complete antler artifacts that cannot be identified. One is a curved section of antler of uniform width that is concave on one surface and convex on the other (Pl. 11, 9), while the other is a ring with a pointed projection rising from one side (PI-10.3). In addition to the identified and unidentified artifacts described above, a total of 288 sections of cut antler and 1,073 pieces of cut bone without any articular surface present were counted and dis- carded in the field. Clay In other historic Eskimo sites in southwestern Alaska excavated in recent years, it has been noted that the tradition of pottery making was seen in its final stages (Oswalt and VanStone, 1967, p. 74; VanStone, 1968b, p. 312). This is even more the case at Akulivik- chuk where only two potsherds and 12 lamp fragments were recov- ered. In connection with the Crow Village and Tikchik sites, it has been suggested that the acceptance of metal and crockery containers by the Eskimo rendered clay cooking and storage vessels obsolete, but that lamps may have continued in use for a longer period of time due to the difficulties involved in obtaining a regular supply of coal oil. In any event, it seems clear that in the Nushagak River region, as well as along the Kuskokwim River to the north, pottery making had virtually ceased by the end of the nineteenth century. centimeters { 4 2 SS ee PLATE 11. Artifacts of Bone, Antler and Miscellaneous Materials. 1. Sled shoe section (p. 65); 2. Scraper (p. 68); 3. Labret (p. 68); 4. Drilled fragment of non-Eskimo pottery (p. 68); 5. Scraper (p. 68); 6-7. Sled shoe sections (p. 65); 8. Kayak shoe section (p. 65); 9. Unidentified (p. 65); 10. Pottery lamp fragment (p. 67); 11. Net float (p. 68); 12-13. Basket fragments (p. 67). 66 VANSTONE: AKULIVIKCHUK 67 Neither of the two recovered potsherds from Akulivikehuk is a lip fragment and it is impossible to learn anything concerning vessel size or shape from them. One is plain and the other shows the Yukon Line-Dot form of decoration (Oswalt, 1955, p. 37). This form of sur- face treatment occurs at both Crow Village and Tikehik where it is associated with the so-called situla shape. Vessels of this shape are widely spread in Alaska from Hotham Inlet south to Kodiak Island. Both Akulivikchuk sherds are tempered with a coarse inorganic ma- terial and are 7 mm. thick. Lamps in the collection are represented by one large fragment and 11 small ones including three rim sherds. The large fragment, which represents about one-quarter of a complete, saucer-shaped lamp, has an encircling line outside the rim, five lines just inside the bowl, and a spoke-like design of two lines each which, presumably, divided the center of the bowl into quarters (Pl. 11, 10). There is also a single, or possibly a double, circular engraved line at the very center of the vessel. The saucer-shaped lamp with encircling lines near the rim and a cross motif toward the center is widespread in southwestern Alaska and seems to have made its appearance in the region no earlier than 1830 (Oswalt, 1958, p. 22). Of the remaining lamp sherds, all but three rim sherds show two encircling lines outside the rim and the others, one. All the fragments are tempered with gravel. Bark Containers of birch bark are commonly associated with interior Eskimo settlements in southwestern Alaska and since preservation was generally poor at the Akulivikchuk site, the few basket frag- ments recovered undoubtedly give an unrealistic impression of the extent to which the material was actually used. Only three frag- ments clearly suggest one manner in which such baskets were con- structed. These represent containers which were made from a single piece of bark folded at two ends to form the basket and then stitched, probably with spruce root (Pl. 11, 12). The remaining fragments were probably also parts of baskets, but this cannot be determined with any degree of certainty. All of them have a row of sizeable, widely-spaced holes along one edge for sewing with spruce root. However, one fragment has very closely-spaced, small sewing holes suggesting that a finer material, perhaps finely-braided grass, was used (Pl. 11, 13). 68 FIELDIANA: ANTHROPOLOGY, VOLUME 60 The only other bark artifacts in the collections are two fragmen- tary net floats made of cottonwood bark. Both are roughly rectangu- lar in outline with laterally-gouged, square holes at each end for attachment to the net (Pl. 11,11). In cross-section, these floats have a rounded triangular form, narrow at the top and thick at the bot- tom. Like the birch bark baskets mentioned above, these floats are doubtless a form that was used extensively at Akulivikchuk but is virtually absent from the collection because of poor preservation at the site. Glass Two pieces of dark green bottle glass and a single fragment of a clear bottle have been extensively retouched for use as scrapers (Pl. 11, 2,5). All are between 6 and 8 mm. in thickness. Chipped glass scrapers of this type have been reported from several historic sites in Alaska and the form is also common elsewhere in North America. A fragment of window glass 2 mm. thick and 2.7 em. in diameter has been chipped to a round shape for an unknown purpose. Wood The only two wooden artifacts in the collection were recovered from the midden directly in front of the kashgee. Both are illus- trated because it has been impossible to identify them (PI. 12, 1-2). The narrow, flat object rounded at one end and witha flattened, rounded tip at the other may be a peg, or possibly a trap or snare part. Non-Eskimo Pottery A single rim sherd of ironstone china showing the familiar “‘wil- low” transfer print has a small hole drilled through it (Pl. 11, 4). An undecorated fragment of ironstone china has been worked into a round shape for use as a labret (Pl. 11, 3). It has been grooved in such a way that each surface forms a lip to hold the labret in place. A similar pottery labret was recovered from the Crow Village site (Oswalt and VanStone, 1967, p. 51, pl. 12,m). Metal In identifying locally manufactured metal artifacts from other historical sites in southwestern Alaska, it has been noted that the indigenous nature of such artifacts is sometimes difficult to deter- mine because heavily rusted metal objects may look very similar whether they were made in the village by Eskimos or imported as VANSTONE: AKULIVIKCHUK 69 finished trade goods (Oswalt and VanStone, 1967, pp. 48-49; Van Stone, 1968b, pp. 284-285). The situation at Akulivikchuk, how- ever, is somewhat clearer in this regard and the author is reasonably certain that all those specimens described below were, indeed, made in the village and, as at Crow Village and Tikchik, they clearly illus- trate the Eskimo reaction to the availability of a new raw material. This new and exotic material was apparently most readily available in the form of tinned steel plate normally used in the manufacture of tin cans. Wrought and cast iron as well as small amounts of brass were apparently also available. In the past when describing indigenously manufactured metal artifacts, it has been assumed that metal, particularly can metal, could be cut and worked more easily than stone and other indigenous materials (Oswalt and VanStone, 1967, p. 49; VanStone, 1968b, p. 285). One reason for this assumption was that such metal arti- facts greatly outnumbered those of stone at both Crow Village and Tikchik. This is not the case at Akulivikchuk, however, and it will be noted that of the few comparable forms, only metal ulu blades outnumber their stone counterparts. Of course, this may be due directly to the variety or abundance of metal at the various sites, but it would also seem that the two materials were not generally considered interchangeable. That is, it was considered possible to substitute metal for stone for some forms, but this substitution was rarely, if ever, attempted with others. More often, probably, metal lent itself to use for new forms or as a substitute for other materials such as bark and wood, or for decorative purposes. One of the most typical artifact types found in sites belonging to the historic period in Alaska are blunt arrowheads which have been constructed by fitting spent rifle cartridges over the ends of thin, wooden shafts. A small section of shaft protrudes from the two Akuli- vikchuk specimens (Pl. 12, 5). Both cartridge cases are of the 44 caliber rimfire type and show the double firing pin marks of the Henry Repeating Rifle, invented in 1860, or of the Model 1866 Win- chester, manufactured until 1898. One case has a plain base while the other is headstamped with a raised letter ‘‘H.’’ The raised “‘H,”’ as opposed to a sunken one, dates this particular cartridge, which was made into the 1930’s, as pre-1900 (Logan, 1959, p. 68; Barnes, 1965, p. 280; Smith, 1955, p. 8; Williamson, 1952, p. 460). Two lead musket balls ave virtually the same size, being approxi- mately 1.5 cm. in diameter (Pl. 12,11). Both have raised burrs run- 70 FIELDIANA: ANTHROPOLOGY, VOLUME 60 ning medially around them indicating the division of the two halves of the mold in which they were made. In addition to these complete balls, there is also a single fragment of melted lead in the collection. Four complete and one unfinished end bladed knife blades we recovered at the site, three of which are 14.8, 15.5, and 8.8 em. in length, respectively. The complete specimens appear to have been made from narrow, rectangular sections of metal which perhaps were originally sections of barrel hoops. The only workmanship evident is a V-shaped cut at one end and the longitudinal sharpening of one side (Pl. 12, 3, 10). A fourth has a more carefully shaped blade with the remnants of an antler handle attached (Pl. 12, 7). The unfin- ished specimen, 22.3 cm. long, is made of a somewhat heavier mate- rial and has been roughly cut on one side along about half its length. There are four ulu blades, flat across the top with semi-lunar edges and all cut from the sides, tops or bottoms of cans. Three are sim- ilar in size to the largest illustrated specimen and one of these has a very narrow, rectangular antler handle in place (Pl. 12, 14). This handle has a narrow groove running along each side parallel to the cutting edge of the blade. The fourth blade is similar to the others but much smaller (Pl. 12, 6). Six crooked knife blades are similar in shape and do not vary much in size from the illustrated examples. With one exception, all curve up at the distal end, narrow slightly at the proximal end, and tend to be made of somewhat heavier metal than the previously described end bladed knife blades and ulu blades. The exception resembles the others except that it is wider at the proximal than at the distal end and has a hole through it which doubtless aided in hafting the blade (2s AS BES). Two artifacts that have been tentatively identified as adz or skin scraper blades are quite different in appearance. One has a convex scraping edge and tapers toward the proximal end. The sides have been folded over, presumably to strengthen the blade (Pl. 12, 17). The other resembles a small ulu blade except that it is broader in proportion to its length (Pl. 12, 18). A fragment of heavy, narrow, rectangular metal has been worked to a point at one end and may have been used as an awl (Pl. 12, 4). An interesting indigenously constructed metal artifact that has been reported from both the Crow Village and Tikchik sites (Os- walt and VanStone, 1967, p. 50, pl. 12, 1; VanStone, 1968b, p. 287, pl. 7, 1) is a crude dish or small container made from pieces of can PLATE 12. Wood and Metal Artifacts. 1-2. Unidentified (p. 68); 3. End bladed knife blade (p. 70); 4. Awl (p. 70); 5. Blunt arrowhead (p. 69); 6. Ulu blade (p. 70); 7. End bladed knife blade (p. 70); 8-9. Crooked knife blades (p. 70); 10. End bladed knife blade (p. 70); 11. Musket ball (p. 69); 12. Reinforce- ment piece (p. 73); 138. Pendant (p. 73); 14. Ulu (p. 70); 15. Scoop (p.72); 16. Sled shoe section (?) (p.72); 17-18. Adz or skin scraper blades (p. 70). (pl G2 FIELDIANA: ANTHROPOLOGY, VOLUME 60 siding which have been folded at the corners to form a shallow, roughly rectangular container. There are three of these in the Akuli- vikchuk collection and they are all badly corroded and very frag- mentary. Closely related to these dishes are two scoops, also made of can metal which has been bent in such a manner as to form a round hole at the proximal end into which a short, wooden handle might have been fitted. The other end is crimped and folded to form a shallow bowl (Pl. 12, 15). The largest specimen, not illustrated, is 21 cm. long. Such scoops, which were also recovered from the Tikchik site (VanStone, 1968b, p. 287, pl. 7, 9), might have been used for flour or sugar. An iron rod, rectangular in cross-section and 47 em. in length, has been bent so as to form a circular hook at one end and a small eye at the other. Such an implement was probably used as a pothook. A rectangular strip of metal with a single hole centrally located near one end has been identified as a sled shoe section (Pl. 12, 16). A shoe of this type would probably be attached to the runner with short, flat-headed nails, in the same manner as the previously de- scribed bone and antler shoes. There are also four sections of rec- tangular metal approximately the same width as the illustrated specimen but shghtly longer. These may have been cut for finishing as sled shoe sections, but they are just as likely to have been pre- pared for manufacture into other types of artifacts. A type of necklace, parts of which are represented in the collec- tion, was made from beads and spent rifle cartridges, each with a small hole drilled through the head. Three are 44 Henry rimfire cartridges, two with plain bases and one with a raised letter ““H’’ in the center of the head. All these show double firing pin marks like those 44 caliber cartridge cases previously described. The fourth cartridge case is a 38 long rimfire with plain base, a cartridge that was very popular for a period of 50 or 60 years following the close of the Civil War. It was chambered to many different makes of weap- ons, usually inexpensive rifles, and some revolvers (Barnes, 1965, p. 278; Datig, 1958, vol. II, p. 150). Necklaces with bead separators made from spent cartridge cases would appear to be common in his- toric sites in southwestern Alaska as they have been reported at Crow Village (Oswalt and VanStone, 1967, p. 51), Tikchik (Van Stone, 1968b, p. 287, pl. 7, 18), and the Kijik site on Lake Clark (VanStone and Townsend, 1970, p. 71, pl. 14, 10). VANSTONE: AKULIVIKCHUK 73 The center panel has been cut from the lid of a can for percussion caps and three small, closely spaced holes drilled along the edge, pre- sumably so that the object could be worn as a pendant (Pl. 12, 13). On the lid is a stamped heraldic design. Above this design and run- ning around the edge is the following two-word inscription, one of which is only partly legible: ““——-—MILLS / LONDON.” The three drilled holes are directly over these words. Below the design is the single word ““ELEY’S” and below that in three successive lines are the words ““_DOUBLE WATERPROOF / CENTRAL FIRE PERCUSSION CAPS.” Eley Brothers is a British firm which manu- factured percussion caps in the nineteenth century and is still in business (Russell, 1962, p. 248). An Eley cap box with a somewhat different inscription and of a smaller size was recovered at the Crow Village site (Oswalt and VanStone, 1967, p. 64). Four cut can fragments have a number of holes around the sides and may have been used as reinforcement pieces. These fragments vary in length from 4 to 10 em. and in width from 2.5 to 8 em. Such metal reinforcement pieces are often used in the Nushagak River re- gion today to strengthen the cracked shaft of an oar, or to repair wooden artifacts of various kinds.