Quaternary International 297 (2013) 93e99

Contents lists available at SciVerse ScienceDirect

Quaternary International

journal homepage: www.elsevier.com/locate/quaint

Organization of at the / transition on the Plains of

Kristen Carlson, Leland Bement*

University of , United States article info abstract

Article history: This paper focuses on the development of large scale across the North American Great Available online 26 December 2012 Plains. Prehistoric hunters were not merely opportunistic. An understanding of topography, environ- ment, bison behavior, and migration patterns was necessary to perform complex, large scale bison kills. In turn, these kills required the existence of social complexity whereby multiple groups of hunters worked in unison toward a successful kill event. On the southern Plains of North America, evidence suggests large scale bison hunting arose as and other megafauna became extinct 11,000 radiocarbon years ago. We review this evidence in light of new site discoveries. Ó 2012 Elsevier Ltd and INQUA. All rights reserved.

1. Introduction This research falls under the overarching umbrella of landscape archaeological theory. Landscape theory can simply be defined as During a nearly 11,000 radiocarbon-year reign devoted to bison the fundamental nature in which people relate to the ecological hunting, Native American hunters added several large-scale kill spaces they occupy (Anschuetz et al., 2001). The application of techniques, including arroyo traps, dune traps, corrals, and jumps, cyclical nucleation under the larger theoretical framework of to their more typical small-scale, pond-side, single kill landscape theory provides a transformative way to view Paleo- repertoire. Many of these techniques were developed by the end of indian bison hunting that generates new hypotheses and expec- the Younger Dryas ca. 10,000 radiocarbon years before present (BP) tations, while acknowledging the substantial resource base these and all were widely employed by 5000 BP. The development of kill events produced. these various trap techniques stems from an understanding of North kill sites, once only thought to fulfill bison behavior. Arroyo traps are the earliest evidence for large-scale procurement and subsistence needs, are currently being re- kills in North America, jumps developed later. This paper provides assessed from a social perspective (Hayden and Dietler, 2001; a general discussion of herd manipulation and bison hunting in Speth, 2010). Often the bison kill is only one component of a site North America with the use of two case studies to demonstrate complex also containing a processing area, camp, and ritual area trends in large-game hunting. The following section provides (Zedeno, 2011). One possible interpretation of these communal theoretical background. complexes incorporates cyclical nucleation (rendezvous or aggre- In many parts of the world the development of specialized large- gations) and ritual observances with the execution of communal scale hunting of employing traps and containment bison hunts that employ large bison jumps, pounds, or arroyo traps systems (pounds) coincided with major environmental perturba- (Bement, 1999; Cooper, 2008; Bamforth, 2011; Carlson, 2011). tions (Krotova and Belan, 1993; Straus, 1997, p. 166). The magnitude Cyclical nucleation is the scheduled aggregation of multiple of environmental shifts altered vegetation structure and herd subsets of band society at a predetermined node or location behavior, leading to the development of migration patterns. Estab- (Schaedel, 1995; Turpin, 2004). Band amalgamation allowed other- lished migration patterns enabled hunters to predict and exploit wise dispersed groups to perform acts outside their usual capacities. seasonal increases in prey density (Wisehart, 2005; Graves, 2010). These acts comprised communal events such as large-scale hunting, Under these conditions, the were susceptible to and social networking including information exchange, feasting, manipulation. mate selection, trade relations, and social bond reinforcement. To participate in macroband activities, dispersed groups required a time and place to meet. A node was needed in a predictable location and at a time of the year when a resource surplus was * Corresponding author. University of Oklahoma, Oklahoma Archeological Survey, 111 E. Chesapeake, Rm 102, Norman, OK 73019, United States. available to support the accumulation of a large number of people. A E-mail address: [email protected] (L. Bement). herd of predictably migrating bison and a topographic such

1040-6182/$ e see front matter Ó 2012 Elsevier Ltd and INQUA. All rights reserved. http://dx.doi.org/10.1016/j.quaint.2012.12.026 94 K. Carlson, L. Bement / Quaternary International 297 (2013) 93e99 as an arroyo or jump provided this predictable node. This inter- turning northward on the Llano Estacado in west . Minor pretation is at odds with the alternative perspective that early large- amounts of Tecovas and Alibates chert from sources in the Texas game hunters were highly dispersed groups traveling across large panhandle signaled that central Texas Folsom groups visited those areas (Kelly and Todd, 1988), with few opportunities for aggregation source areas directly and refurbished their stores as Edwards (Hofman, 1994). This alternative perspective will be discussed chert supplies dwindled. The recovery of Folsom points made from further in subsequent sections of this paper. distant stone sources fit with the highly mobile lifestyle recon- Bison were a key component to the life of Plains hunters (Frison, struction for these early Paleoindians. 2004). Bison are herd animals, travel in groups, tend to pack The paucity of Alibates points and in central Texas together to avoid predators, and flee as a herd rather than scatter, assemblages, however, was enigmatic. The alternative argument for making them a prime target for human hunters. High bison pop- separate home ranges is also apparent in this same study. Hofman ulations were necessary for hunters to manipulate herds into traps. raised the possibility that a separate Alibates territory existed based The of mammoths and other large grazing animals may on the Folsom, New Mexico, site assemblage and dominance of have contributed to increased bison population levels and range Alibates chert in Oklahoma panhandle surface collections (1991,p. expansion by reducing competition for resources (Lewis et al., 343). The investigation of new Folsom sites in northwestern 2010). Environmental shifts may have had consequences for bison Oklahoma has identified some assemblages dominated by Alibates . Paleoenvironmental reconstructions for the chert (Table 1; Bement, 1999; Bement et al., 2012), possibly sup- Younger Dryas indicate the development of expansive grasslands porting the existence of a separate Alibates Folsom home range. A sufficient to support greater bison densities (Holloway, 1993; separate Alibates home range would account for the lack of Alibates Cummings, 1996; Connin et al., 1998; Larson et al., 2001; Hoppe, chert Folsom artifacts in central Texas. Alibates chert Folsom 2006; Meltzer, 2006; Bement et al., 2007; Holliday et al., 2011). material usually moves north, east, and west of the source area Bison behavior plays a crucial role in a successful hunt. Herd (Hofman, 1991), a pattern established during the Clovis Period composition varies seasonally (Brink, 2008). Cows and bulls (Holen, 2001, p. 177).

Table 1 Lithic composition of the Beaver River Complex chipped stone assemblages.

Site/component Alibates chert Edwards chert Morrison Quartzite Smoky Hill Jasper Petrified wood Other Jake Bluff e Clovis 55.6%(5) 33.3% (3) 11.1% (1) Jake Bluff e Folsom 66.7% (2) 33.3%(1) Cooper e Lower 42.9% (3) 57.1% (4) Cooper e- Middle 9.1% (1) 81.8% (9) 9.1% (1) Cooper e Upper 68.4% (13) 21% (4) 5.3% (1) 5.3% (1) Badger Hole 75% (6) 12.5% (1) 12.5% (1)

converge during the rutting season. The rest of the year, cows and The idea that the amalgamation of disparate lithic source bulls split into bull herds and cow/calf herds. The rut occurs from materials in a site assemblage may indicate the aggregation of July through early August. Calving occurs in April or May after groups from different regions has been previously proposed a nine-month gestation period. Animal condition also varies (Wilmsen and Roberts, 1978; Reher and Frison, 1980; Jodry, 1999). A seasonally (Brink, 2008). Females carry the most fat throughout the case for the simultaneous occupation of the same site by different year, making them the prime hunting target. Bulls carry the most Folsom groups based on lithic stone types can be made at the fat up to the rutting season, but in that season hormones render the Lindenmeier site in Colorado (Wilmsen and Roberts, 1978; LaBelle, meat undesirable (Audubon and Bachman, 1846; Brink, 2008). 2012). This has been postulated by Jodry (1999) as well at Stewart’s The final factor of the hunting system is the hunters. Group Cattle Guard site in the San Luis Valley of Colorado. aggregations may have been necessary to amass the necessary The joining of otherwise autonomous groups has implications manpower to carry out these complex drive systems (Frison, 2004; for the social organization of early Paleoindians during the Folsom Brink, 2008; Kornfeld et al., 2010). The rise of social complexity was period and is considered in the following discussion. Archaeologi- seen in the emergence of expedient leaders for specific tasks. cally, aggregation sites contain larger-than-normal assem- Increased population density promoted scalar stress associated blages. These assemblages are often accompanied by a higher with dissemination of information among large numbers of people diversity or richness of cultural items, exotic materials, curated and (Johnson, 1982). Emergent hierarchies to handle information recycled tools, and artifacts in late-stage reduction (Hofman, 1994; transference and organization of people developed (Fawcett, 1987; Veth, 2006, p. 270). Further discussion on aggregation and lithic Cooper, 2008). materials will be discussed below; but first the organization of the Early large-game hunter mobility patterns have been recon- structure of bison jumps on the northwestern Plains is presented as structed from lithic raw material distributions across the landscape. an example linking social organization with bison behavior in The interpretations of these distributions vary. One interpretation large-scale kill events. is that highly mobile groups directly procured various stone sources A comparison of bison jumps across the northwestern Plains as needed along seasonal rounds. The other interpretation posits and an arroyo trap complex on the southern Plains provides an raw materials from multiple sources indicate the aggregation of opportunity for understanding multiple factors pertinent in the regionally distinct groups utilizing single sources per group. manipulation of large herds for large-scale bison kills. The success The case for the direct procurement of various sources by of a large-scale bison kill relies on the knowledge and ability of a single group was summarized by Jack Hofman (1991). Hofman hunters to intercept herds of bison and maneuver those herds to hypothesized that Folsom groups utilizing the Edwards Plateau kill areas. The following section outlines the importance of herd cherts from central Texas followed an eastewest annual round, manipulation and topographic understanding among large-game K. Carlson, L. Bement / Quaternary International 297 (2013) 93e99 95 hunters on the northwestern Plains employing jumps. The drive lanes near the or in some cases hiding behind them in southern Plains kill complex section addresses herd movement and order to keep the bison within the drive lanes. As the bison ran migration patterns. through the drive lane, the hunters may have flapped hides and shouted to direct the bison through the lane (Schaeffer, 1978). 2. Herd manipulation on northwestern Plains jumps Ongoing research however has dismissed this theory (Brink, 2008). Ethnographic accounts indicate that cairns were likely used in fl Herd manipulation is apparent on the northwestern Plains conjunction with twigs, brush, and ags creating a path of moving where -marked drive lanes delineate preplanned routes objects that a bison with poor eye sight could not distinguish from leading to jump precipices. Aggregated groups of large-game a real person. This method of cairn use requires far fewer individ- hunters would gather typically in the late fall to carry out uals in order to drive bison to the kill point. complex bison jumps (Verbicky-Todd, 1984; Frison, 2004; Brink, Bison will not stampede over an obvious cliff. The drive lane ’ 2008; Zedeno, 2011). These bison jump systems appear in the topography must therefore obscure the bison s view of the drop-off archaeological record for over 5000 years. People of the north- point. If the herd detects the trap or cliff early, the animals will shift western Plains employed communal bison hunting strategies that away, posing a substantial risk to hunters. In a successful jump the required inter-band cooperation, topographic understanding, and inertia of the herd pushes the lead animals over the cliff after which a sophisticated knowledge of animal behavior. A bison jump is some hunters gather at the base to slaughter bison that survive the e a communal hunting method employed by prehistoric hunters that fall. Successful jumps require herds of 50 100 animals (Frison, involves running the lead animals in a stampeding bison herd over 1987, p. 196). The development and function of these hunting a cliff. Bison jump sites occur from , Canada, to Texas, United methods suggests not only high levels of coordinated human States. Bonfire Shelter in Texas (Fig. 1) could be the oldest known activity, but also a deep understanding of bison behavior as well as bison jump site at 10,230 160 BP (TX153; Dibble and Lorrain, complex decision-making by hunters in response to bison behavior 1968:33), predating Head-Smashed-In (in Alberta, Canada) by (Brink and Rollans, 1990:164). 5000 years (Bement, 1986; Byerly et al., 2005; Brink, 2008). In any large-scale hunt, bison must first be moved from their 2.1. Methods grazing basin, often referred to as a milling area, to the kill area. In the case of arroyo traps, the steep arroyo walls marked the path to In a study of eight bison jumps spanning a distance of 1280 km move animals from the milling area to the trap. In the case of jumps (800 miles) across the northwestern Plains similarities arose that hunters marked the paths, referred to as drive lanes, with rock provide insight into the way large-scale kills operated (Fig. 1; cairns. Prehistoric rock cairns create a map of drive lanes across the Carlson, 2011). All eight sites have milling areas, drive lanes, and landscape. For example, at Head-Smashed-In, prehistoric hunters cliffs. The cliff is hidden from view by the topography in all eight constructed more than 500 drive lane cairns to help guide bison cases. The relatively short drive lanes when mapped and assessed herds to the critical jump point (Reeves, 1978, p. 154). by Least Cost Path Analysis, illustrate herd manipulation from The working of the cairns lining the drive lanes is a source of milling area to jump. ongoing debate among Plains archaeologists. The initial assump- Least Cost Path Analysis is applied through a geographic infor- tion was that people positioned themselves along either side of the mation systems (GIS) program such as Esri GIS. A digital elevation model (DEM) is used as a base map. This DEM is then ranked with a cost based on topographic slope. Low relief topography incurs a lower ranking while high relief topography gets a higher ranking. The Least Cost Path Analysis will locate the route outlined by lowest ranking values between two points to determine a path of least resistance. With the use of Least Cost Path Analysis Carlson (2011) determined that six of the eight jumps compared had marked drive lanes that followed a path of least cost across the landscape (Fig. 2). The drive lane systems at the other two sites were too poorly preserved for GIS analysis. The conclusion of Least Cost Path Analysis indicates that prehistoric hunters applied extensive knowledge of the landscape to aid in moving bison. Hunters carried out these complex and difficult drives by using the topography to their advantage in drive lane construction.

2.2. Conclusions

After conducting GIS analysis, visiting eight sites, and surveying the landscape, Carlson (2011) concluded that bison drive lanes were determined by the use of natural topography in conjunction with knowledge of bison behavior. Cairn lines occurred where control of the bison appears to be of the most importance in the drive, for example, in tight turns, and along ridge edges. The cairns draw a clear map of where the bison needed to run in order to have a successful jump. The use of modern geographic information systems illustrates that prehistoric hunters on the northwestern Plains had a sophisticated knowledge of topography and animal behavior. Cairn-lined terrain maps were built to guide large bison Fig. 1. Map of site locations included in this study. The Bighorn National Recreation Area (4) contains three sites. The Beaver River Complex (6) includes the Jake Bluff, herds at a stampede over a cliff sometimes no wider than 20 m. The Cooper, and Badger Hole sites. cliffs were ingeniously chosen with bison behavior in mind, hiding 96 K. Carlson, L. Bement / Quaternary International 297 (2013) 93e99

Fig. 3. Site location map of the Beaver River Complex sites.

consisted of a cow/calf herd of at least 22 individuals (Table 2) driven up the arroyo to the knickpoint. A single black bear was also killed by the Clovis hunters. The kill took place during early fall. The associated projectile points, flake , and resharpening flakes are made of lithic raw material from the Texas panhandle and far western Oklahoma panhandle at distances of 250 km and 350 km, respectively (Table 1). In addition to the Clovis kill, a Folsom layer occurs 1 m above the Clovis kill. Small fragments of bison bone are scattered within the deposit and the lithic assemblage consists of one Folsom , several resharpening flakes, and caliche cobbles of undetermined use. The lithic material comes from central Texas at a distance of 400 km (Table 1; Bement and Fig. 2. Overlay results of the Least Cost Path Analysis, mapped position of rock cairns, and DEM base map for the TE Ranch site in Wyoming. The prehistoric drive lane Carter, 2010). follows the path of least resistance between the bison milling area and jump site.

Table 2 Summary attributes of the Beaver River Complex sites. the precipice from view until the last 10e15 m of the drive. Site Radiocarbon MNI Seasonality Herd Topography is a key element in these successful bison jumps, but date demographics bison behavior clearly was taken into consideration during the Jake Bluff Clovis 10,838 17 22 Late Summer/early fall Cow/calf entire process from milling area to cliff. Cooper-upper 10,505 40 29 Late summer/early fall Cow/calf Cooper-middle 10,535 40 29 Late summer/early fall Cow/calf Cooper-lower 10,600 45 20 Late summer/early fall Cow/calf 3. The Beaver River Complex on the southern Plains Badger Hole 10,300 25 10 Late summer/early fall Cow/calf

On the southern Plains, arroyo traps replace jumps as the most common large-scale bison killing system (Cooper, 2008). On the 3.2. Cooper site southern Plains the drive lane systems (arroyo walls) are absent, having washed away after thousands of years of erosion. What A distance of 400 m separates Jake Bluff from the three kills in remains are remnants of intact bonebeds. The Beaver River the Cooper site arroyo (Bement, 1999). The Cooper bonebeds are Complex illustrates herd manipulation in a different setting and characterized by dense accumulations of overlapping articulated time period than northwestern Plains jumps. Arroyo traps are the skeletons. These bones are adjacent to the knickpoint or headward earliest-known form of large-scale bison hunting dating back to the end of the paleoarroyo. Orientation of the articulated skeletons Clovis period (Bement, 1999; Kornfeld et al., 2010). Three arroyo indicates the bison were herded up the arroyo to the dead-end traps along the Beaver River comprise the Beaver River Complex knickpoint where the lead animals attempted to turn around, (Bement et al., 2012). All three employ steep-walled dead-end only to be stalled by the animals coming up the arroyo. The three fl arroyos that open onto the oodplain of the Beaver River (Fig. 3). bonebeds contained minimum numbers of individuals (MNI) of 29, 29, and 20, respectively (Table 2). Each group consisted of cows, 3.1. Jake Bluff site calves, and juveniles. All three kills occurred in late summer/early fall. The selective butchering targeted the hump and ribs of animals The earliest site of the Beaver River Complex is the Clovis-age of all ages (Johnson and Bement, 2009). The combined lithic Jake Bluff site (Fig. 3; Bement and Carter, 2010). The activity at assemblage of projectile points, flake , and resharpening Jake Bluff dates to 10,838 17 BP (Bement and Carter, 2010), post- flakes consisted of raw materials from central Texas, Texas dating all known Clovis kills and earning Jake Bluff panhandle, northwest Kansas, and possibly the Colorado foothills a position as one of the latest Clovis sites in North America (Waters at distances of 450 km, 250 km, 300 km, and 600 km, respectively and Stafford, 2007; Holliday and Meltzer, 2010). The Clovis kill (Table 1). K. Carlson, L. Bement / Quaternary International 297 (2013) 93e99 97

3.3. Badger Hole site of signature elements is known. On the southern Plains, the area signatures are provided by analysis of tooth enamel from modern The third arroyo trap is found at the Folsom-age Badger Hole site bison born and raised within small ranches in disparate locations. (Bement et al., 2012). It is located 300 m upstream from Jake Bluff By using modern bison teeth, trace element signatures were and 700 m upstream from Cooper. The Badger Hole bison demonstrated to be incorporated in bison tooth enamel, thereby (MNI ¼ 10; Table 2) are not at the knickpoint of the arroyo, but circumventing the expensive and drawn out arduous task of instead are closer to the mouth of the arroyo. The knickpoint of the tracking the movement of each trace element through the ecolog- arroyo has been washed away by later cut and fill sequences and ical system/food chain. The modern teeth provide a geo-reference only a small remnant of the original kill remains intact. The animals against which the prehistoric samples can be compared and indicate a cow/calf herd was slaughtered during the end of plotted to approximate movement. The fourth premolar enamel summer/early fall, mirroring the seasons at Jake Bluff and Cooper. forms over the course of approximately 12 months while the third At Badger Hole, the skeletons do not overlap. The difference molar enamel forms over 18 months (Niven et al., 2004; Graves, between Badger Hole and Cooper may be the result of the position of 2010; Widga et al., 2010; Julien et al., 2012) making it the ideal the excavated areas in relation to the knickpoint. Carcasses jammed sample for comparison. against the knickpoint provide little room for subsequent butchering State-of-the-art laser ablation time of flight inductively coupled activities. Bison farther down-arroyo from the knickpoint where the plasma mass spectrometry of tooth enamel sampled down the arroyo broadens, may not be so densely packed, allowing butchers tooth at set intervals starting at the crown provides chemical more mobility around the carcasses. The presence of broken skulls characterization of the areas traversed by the animals during the and spirally fractured long bones at Badger Hole may signify that time of tooth enamel amelogenesis. Down-tooth sampling provides butchers were better able to access the carcasses and more fully a time stamp: 5 mm increments reflect every 2 months (Graves, butcher the animals. Cutmarks, however, are only found in the hump 2010), 1.5 mm increments reflect every 2 weeks (Graves and and rib areas, identical to the distribution in the Cooper site kill Bement, 2012) on the mandibular p4. episodes which is consistent with the selective butchering patterns In an effort to determine the extent of bison mobility during early discussed elsewhere (Johnson and Bement, 2009). Paleoindian times on the southern Plains, Graves (2010) matched The stone assemblage of four Folsom points, two flake knife trace element signatures obtained from known tooth enamel fragments, and several resharpening flakes is dominated by stone developmental stages with the distribution of select geographically from the Texas panhandle with lesser amounts of central Texas and restricted trace elements. The results: Bison mobility patterns western Oklahoma panhandle sources also included. These source ranged from resident/restricted movement covering ca. 100 km areas lie 250 km, 400 km, and 300 km, respectively, from the during Clovis times (10,850 BP), to migrations across an area with Badger Hole site (Table 1). a 300 km diameter during middle Folsom times (ca. 10,500 BP); and finally covering a maximum migration diameter of 600 km by late 3.4. Interpretation of the Beaver River Complex Folsom times (10,300 BP) (Fig. 4). Migration followed east-west trajectories; calving during spring in the west, rutting during The development of the arroyo trap technique in this area of the summer in transit from west to east, wintering in the east and then southern Plains was possible because of the existence of suitable short, steep-walled arroyos, large numbers of bison e at least seasonally e and the requisite numbers of hunters with the knowledge of arroyo trap hunting methods. The presence of arti- facts from distant lithic raw material sources raises the possibility that hunters from different regions came together to conduct large-scale bison hunts (Bement et al., 2012), similar to the group aggregation apparent in northwestern jumps. The Beaver River Complex provides the opportunity to under- stand the development of herd migration as it relates to large-scale hunting. In order to track bison herd movement the use of stable isotope analysis and trace element analysis were applied (Chisholm et al., 1986; Graves, 2010; Julien et al., 2012). A resident herd was indicated if all individuals yield a tight cluster of stable d13C and d15N results (Chisholm et al., 1986). A migrating herd would display greater variability in stable d13C and d15N isotopic results than a resident herd (Chisholm et al., 1986; Tieszen, 1998; Hoppe and Koch, 2007; Julien et al., 2012). Just as geologists collect hand samples as they walk across distinct geologic formations, ungulates pick up trace amounts of elements as they graze across the landscape. Bison ingest trace elements dissolved in water, adhering to plants, or directly through geophagy (ingestion of sediment). Some of these elements are incorporated in tooth enamel during amelogenesis (Hillson, 2005). This minute geologic signal can be extracted from the enamel in down-tooth increments representing known time slices, and, when tied to known geographical source areas, provide a map of the areas traversed by the animal during enamel formation (with a slight Fig. 4. Map outlining the migration areas for sites of the Beaver River Complex at time delay from consumption to amelogenesis). different times. The bison skull location marks the convergence of hunters carrying The down-tooth results can be used to map the movement of lithic materials from distant sources (shaded areas), the late summer/early fall position the individual across the landscape if the geographical distribution of migrating bison herds, and suitable arroyos for bison traps. 98 K. Carlson, L. Bement / Quaternary International 297 (2013) 93e99 completing the cycle back to the west for spring calving (Fig. 4). At ability to manipulate bison groups from milling areas to the kill its greatest migration during late Folsom times, calving occurred areas. The southern Plains Beaver River Complex is proposed to be in north central New Mexico, rut in the Oklahoma and Texas an example of a node where various early Paleoindian groups Panhandles, and wintering in eastern Oklahoma/southeastern converged to conduct communal bison kills. This cursory view of Kansas; a one-way straight-line distance of 600 km (Graves and northern and southern Plains bison hunting strategies sets the Bement, 2012). stage for expanding the investigation into large-scale hunting The use of the arroyo bison trap technique along the Beaver adaptations in other areas of the world and incorporating other River began by 10,838 17 BP at Jake Bluff. Cooper kills date to species. 10,600 45 BP at the lower kill, 10,535 40 BP at the middle kill, and 10,505 40 BP at the upper kill level (Johnson and Bement, Acknowledgements 2009). Badger Hole now extends the arroyo bison trap hunting complex to at least 10,300 25 BP. The 500 radiocarbon-year This article was presented in 2012 at the 77th Society for duration of the Beaver River bison hunting complex transitions American Archaeology annual meeting in Memphis, Tennessee, in from Clovis to Folsom periods. This suggests that conditions per- the Worldwide Large-Scale Trapping and Hunting of Ungulates in sisted from late Clovis through Folsom times, enabling arroyo trap Past Societies symposium. Thanks to Guy Bar-Oz and Dani Nadel kills to continue in this region. Folsom people elaborated on the for organizing and graciously conducting this session. The ’ Clovis adaptation to make arroyo kills one of Folsom hunters comments of two anonymous reviewers improved the content of hallmark attributes. this manuscript. The seasonality of the Beaver River Complex consistently indi- cated these kills occurred in the late summer/early fall. The References repeated late summer/early fall season for the Beaver River

Complex and other large-scale Folsom-age bison kills in the region Anschuetz, K.F., Wilshusen, R.H., Scheick, C.L., 2001. An archaeology of landscapes: (Bement, 2003) suggests patterning in the movement of bison, perspectives and directions. Journal of Archaeological Research 9 (2), 157e211. mobility strategies of hunting groups, or both made bison hunting Audubon, J.J., Bachman, Rev. J., 1846. The Quadrapeds of North America, vol. 2. V.G. predictable and reliable in this area. That all three sites date to the Audubon, New York. Bamforth, D.B., 2011. Origin Stories, archaeological evidence, and Postclovis Paleo- Younger Dryas chronozone following the initial cold snap ca. indian bison hunting on the . American Antiquity 76 (1), 24e40. 10,900 BP but before the ending warm-up ca. 9800 BP indicates the Bement, L.C., Carter, B.J., 2010. Jake Bluff: Clovis bison hunting on the southern e conditions for the development and continuation of this style of Plains of North America. American Antiquity 75 (4), 907 933. fi Bement, L.C., Carter, B.J., Varney, R.A., Cummings, L.S., Sudbury, J.B., 2007. Paleo- bison hunting persisted for hundreds of years. Continued re ne- environmental reconstruction and bio-stratigraphy, Oklahoma Panhandle, USA. ment of the paleoenvironmental reconstruction for this area will Quaternary International 169e170, 29e50. elucidate the context of this adaptation. Initial results from stable Bement, L.C., Carter, B.J., Jelley, P., Carlson, K., Fine, S., 2012. Badger Hole: towards defining a Folsom bison hunting complex along the Beaver River, OK. Plains carbon and nitrogen isotope analysis of bison bone collagen and Anthropologist 57, 53e62. trace element analysis of bison tooth enamel suggest the grassland Bement, L., 1986. Excavation of the Late Pleistocene Deposits of Bonfire Shelter, structure and bison mobility patterns evolved during this interval, 41VV218, Val Verde County, Texas. In: Archaeology Series 1. Texas Archaeo- logical Survey, University of Texas, Austin. leading to development of set migration patterns (Graves, 2010; Bement, L.C., 1999. Bison Hunting at Cooper Site: Where Lightning Bolts Drew Graves and Bement, 2012). These hunters may have planned their Thundering Herds. University of Oklahoma Press, Norman. hunts to intercept the migrating herds (Fig. 4). Bement, L.C., 2003. Constructing the Cooper Model of Folsom bison kills on the southern Plains. Great Plains Research 13, 27e41. In summary, the Beaver River bison hunting complex consists of Brink, J.W., Rollans, M., 1990. Thoughts on the structure and function of drive lane large (>20 bison), late summer/early fall, arroyo traps manned by systems at communal buffalo jumps. In: Davis, L.B., Reeves, B.O.K. (Eds.), groups from various southern Plains regions or alternatively Hunters of the Recent Past. Unwin Hyman, London, pp. 152e167. a single group carrying multiple raw material types from western Brink, J.W., 2008. Imagining Head-smashed-in, Aboriginal Buffalo Hunting on the Northern Plains. AU Press, Athabasca University, Edmonton, Alberta. Oklahoma panhandle, Texas panhandle, and central Texas. Occa- Byerly, R., Cooper, J.R., Meltzer, D.J., Hill, M.E., Labelle, J.M., 2005. On Bonfire Shelter sional inclusion of individuals or the visitation of lithic sources at (Texas) as a Paleoindian bison jump: an assessment using GIS and zooarch- e even greater distance within the central Plains is suggested by the aeology. American Antiquity 70 (4), 595 629. Carlson, K.A., 2011. Prehistoric Bison Procurement: Human Agency and Drive Lane occasional occurrence of tools or points made from sources in Topography on the Northwestern Plains. Unpublished M.A. thesis, Department northwestern Kansas and eastern Colorado (Bement, 1999). of Anthropology, Northern Arizona University. The Beaver River Complex has the highest-density Folsom site Chisholm, B., Driver, J.C., Dube, S., Schwarcs, H.P., 1986. Assessment of prehistoric bison foraging and movement patterns via stable carbon isotopic analysis. concentration in this area of the southern Plains. Together these Plains Anthropologist 113, 193e205. sites belie a Folsom bison hunting complex that is structured on Connin, S., Betancourt, J., Quade, J., 1998. Late Pleistocene C4 plant dominance and bison concentration, arroyo formation, and seasonally redundant summer rainfall in the southwestern United States from isotopic study of herbivore teeth. Quaternary Research 50, 179e193. intercept patterns by Folsom hunters. Cooper, J., 2008. Bison Hunting and Late Prehistoric Human Subsistence Economies in the Great Plains. Unpublished Ph.D. dissertation. Department of Anthro- 4. Concluding remarks pology, Southern Methodist University, Dallas, Texas. Cummings, L.S., 1996. Paleoenvironmental Interpretations for the Mill Iron Site: stratigraphic pollen and phytolith analysis. In: Frison, G.C. (Ed.), The Mill Iron In conclusion the archaeological record of bison kill systems is Site. University of New Mexico Press, Albuquerque, pp. 177e193. complex. The interplay between environment, landform, bison Dibble, D.S., Lorrain, D., 1968. Bonfire Shelter: A Stratified Bison Kill Site, Val Verde mobility, and hunter mobility are just a few of the variables that County, Texas. Miscellaneous Papers No. 1. Texas Memorial Museum, Austin. Fawcett, W.B., Jr., 1987. Communal Hunts, Human Aggregations, Social Variation, must be considered before large-scale bison hunting adaptations and Climatic Change: Bison Utilization by Prehistoric Inhabitants of the Great can be understood. Based on this research it is reasonable to Plains. Unpublished Ph.D. dissertation, Department of Anthropology, University assume that Paleoindian and later hunters could predictably of Massachusetts, Amherst. Frison, G., 1987. Prehistoric hunting strategies. In: Nitecki, M.H., Nitecki, D.V. (Eds.), intercept migrating bison. By combining hunting band aggrega- The Evolution of Human Hunting. Plenum Press, New York, pp. 177e223. tions, seasonal movement of bison, and locations or nodes con- Frison, G., 2004. Survival by Hunting, Prehistoric Human Predators and Animal Prey. taining the requisite landforms for traps or jumps, Paleoindian and University of California Press, Berkeley. Graves, A., Bement, L., 2012. Trace Element Analysis Results from Badger Hole, later hunters developed successful large-scale hunting adaptations. Northwest Oklahoma. Data on File at the Oklahoma Archeological Survey, The success of this adaptation ultimately rested on the hunters’ Norman, Oklahoma. K. Carlson, L. Bement / Quaternary International 297 (2013) 93e99 99

Graves, A., 2010. Investigation Resource Structure and Human Mobility: an Example Larson, R., Todd, L., Kelly, E., Welker, J., 2001. Carbon stable isotopic analysis of bison from Folsom-aged Bison Kill Sites on the U.S. Southern Great Plains. Unpub- dentition. Great Plains Research 11, 25e64. lished dissertation, Department of Anthropology. University of Oklahoma. Lewis, P.J., Johnson, E., Buchanan, B., Churchill, S.E., 2010. The impact of changing Hayden, B., Dietler, M., 2001. Feasts: Archaeological and Ethnographic Perspectives grasslands on late Quaternary bison of the southern Plains. Quaternary Inter- on Food, Politics, and Power. Smithsonian Institution Press, Washington and national 217, 117e130. London. Meltzer, D.J., 2006. Folsom, New Archaeological Investigations of a Classic Paleo- Hillson, S., 2005. Teeth. Cambridge University Press, Cambridge. indian Bison Kill. University of California Press, Berkeley. Hofman, J.L., 1991. Folsom land use: projectile point variability as a key to mobility. Niven, L.B., Egeland, C.P., Todd, L.C., 2004. An inter-site comparison of enamel In: Montet-White, A., Holen, S. (Eds.), Raw Material Economies Among hypoplasia occurrence in bison: implications for paleoecology and modeling Prehistoric HuntereGatherers, 19. University of Kansas Publications in Late Plains Archaic subsistence. Journal of Archaeological Science 31, 1783e Anthropology, Lawrence, KS, pp. 335e355. 1794. Hofman, J.L., 1994. Paleoindian aggregations on the Great Plains. Journal of Reeves, B.O.K., 1978. Head-Smashed-In: 5500 years of bison jumping in the Alberta Anthropological Archaeology 13, 341e370. Plains. Memoir 14. Plains Anthropology 23, 151e174. Holen, S.R., 2001. Clovis Mobility and Lithic Procurement on the Central Great Plains Reher, C.A., Frison, G.C., 1980. The Vore site, 48CK302, a stratified in of North America. Unpublished Ph.D. dissertation, Department of Anthropology, the Wyoming . Plains Anthropologist. Memoir 16. University of Kansas, Lawrence. Schaedel, R.P., 1995. The temporal variants of proto-state societies. In: Kradin, N.N., Holliday, V.T., Meltzer, D.J., 2010. The 12.9-ka ET impact hypothesis and North Lynsha, V.A. (Eds.), Alternative Pathways to Early States. Dal’nauka, Vladivostok, American Paleoindians. Current Anthropology 51 (5), 575e607. pp. 47e53. Holliday, V.T., Meltzer, D.J., Mandel, R., 2011. Stratigraphy of the Younger Dryas Schaeffer, C.E., 1978. The bison drive of the Blackfeet Indians. In: Davis, L.B., Chronozone and paleoenvironmental implications: central and southern Great Wilson, M. (Eds.), Bison Procurement and Utilization: a Symposium. Plains Plains. Quaternary International 242, 520e533. Anthropologist, vol. 14, pp. 243e248. Holloway, R., 1993. Pollen Analysis of Ferndale Bog, Oklahoma. Report to the Speth, J.D., 2010. The Paleoanthropology and Archaeology of Big-Game Hunting: Institute of Applied Sciences. University of North Texas, Denton, Texas. Protein, Fat, or Politics? Springer, New York. Hoppe, K.A., Koch, P.L., 2007. Reconstructing the migration patterns of Late Straus, L.G., 1997. Late glacial hunters along the French Pyrenees. In: Pleistocene from northern Florida, USA. Quaternary Research 68, Jackson, L.J., Thacker, P.T. (Eds.), Caribou and Reindeer Hunters of the Northern 347e352. Hemisphere. Avebury Press, Aldershot, pp. 165e184. Hoppe, K.A., 2006. Correlation between the oxygen isotope ratio of North American Tieszen, L.L., 1998. Stable isotopic determination of seasonal dietary patterns in bison teeth and local waters: implication for paleoclimatic reconstructions. bison at four preserves across the Great Plains. In: Irby, L., Knight, J. (Eds.), Earth and Planetary Science Letters 244 (1e2), 408e417. International Symposium on Bison Ecology and Management in North America. Jodry, M.A.B., 1999. Folsom Technological and Socioeconomic Strategies: Views State University Press, Bozeman, pp. 130e140. form Stewart’s Cattle Guard and the Upper Rio Grande Basin, Colorado. Turpin, S.A., 2004. Cyclical nucleation and sacred space: at the center. In: Unpublished Ph.D. dissertation, Department of Anthropology, American Berghaus, G. (Ed.), New Perspectives on . Praeger, Westport, University, Washington, D.C. Conn, pp. 51e64. Johnson, E., Bement, L.C., 2009. Bison butchery at Cooper, a Folsom site on the Verbicky-Todd, E., 1984. Communal Buffalo Hunting Among the . southern Plains. Journal of Archaeological Science 36, 1430e1446. Occasional Paper No. 24. Archaeological Survey of Alberta, Edmonton. Johnson, G.A., 1982. Organizational structure and scalar stress. In: Renfrew, C., Veth, P., 2006. Cycles of aridity and human mobility: risk minimization among late Rowlands, M.J., Segraves, B.A. (Eds.), Theory and Explanation in Archaeology, Pleistocene foragers of the Western Desert, Australia. In: Sellet, F., Greaves, R., The Southhampton Conference. Academic Press, London, pp. 389e422. Yu, P. (Eds.), Archaeology and Ethnoarchaeology of Mobility. University Press of Julien, M., Bocherens, H., Burke, A., Drucker, D.G., Patou-Mathis, M., Krotova, O., Florida, Gainesville, pp. 262e282. Pean, S., 2012. Were European migratory? 18O, 13C and Sr intra- Waters, M.R., Stafford Jr., T.W., 2007. Redefining the age of Clovis: implications for tooth isotopic variations applied to a palaeoethological reconstruction. the peopling of the Americas. Science 315, 1122e1126. Quaternary International 271, 106e119. Widga, C., Walker, J.D., Stockli, L.D., 2010. Middle Holocene bison diet and mobility Kelly, R., Todd, L., 1988. Coming into the country: early Paleoindian hunting and in the eastern Great Plains (USA) based on d13C, d18O, and 87Sr/86Sr analyses of mobility. American Antiquity 53, 231e244. tooth enamel carbonate. Quaternary Research 73, 449e463. Kornfeld, M., Frison, G.C., Larson, M.L., 2010. Prehistoric Hunter-eGatherers of the Wilmsen, E.N., Roberts, F.H.H., 1978. Lindenmeier, 1934e1974 Concluding Report on High Plains and Rockies. Left Coast Press, Walnut Creek, California. Investigations. Smithsonian Contributions to Anthropology Number 24. Krotova, A.A., Belan, N.G., 1993. Amvrosievka, a unique upper site in Smithsonian Institution Press, Washington, D.C. eastern Europe. In: Soffer, O., Praslov, N.D. (Eds.), From Kostenki to Clovis: Wisehart, A.W., 2005. A Cultural Study of the Bad Pass Trail in the , e Paleo-Indian Adaptations. Plenum Press, New York, Montana and Wyoming. Unpublished M.A. thesis, University of Montana, pp. 125e142. Missoula, Montana. LaBelle, J.M., 2012. Hunteregatherer adaptations of the Central Plains and Rocky Zedeno, M.N., 2011. Kutoyis Archaeology Project and Tribal Historic Preservation Mountains of western North America. In: Eren, M.I. (Ed.), HuntereGatherer Office, Blackfeet Tribe. If We Build It, Will They Come? Planning Bison Hunting Behavior, Human Response During the Younger Dryas. Left Coast Press, Inc., on the Two Medicine River, Montana. Poster Presented at the 69th Plains Walnut Creek, CA, pp. 139e164. Conference, Tucson, Arizona.