Long-distance transport of red ocher by Clovis foragers

Journal of Archaeological Science: Reports 25 (2019) 519–529 Contents lists available at ScienceDirect Journal of Archaeological Science: Reports journal homepage: www.elsevier.com/locate/jasrep Long-distance transport of red ocher by Clovis foragers a,⁎ b b b Sandra E. Zarzycka , Todd A. Surovell , Madeline E. Mackie , Spencer R. Pelton , Robert L. Kellyb, Paul Goldbergc,d, Janet Deweye, Meghan Kentb T a Department of Geography and the Environment 121700, University of North Texas, 1155 Union Cr., Denton, TX, USA Anthropology Department 3431, University of Wyoming, 1000 East University Ave., Laramie, WY 82071, USA Centre for Archaeological Science, SEALS, University of Wollongong, Wollongong, NSW 2522, Australia d Institute for Archaeological Sciences, University of Tübingen, Rümelinstr. 23, 72070 Tübingen, Germany e Department of Geology and Geophysics 3006, University of Wyoming, 1000 East University Ave., Laramie, WY 82071, USA b c A R TICL E INFO A BSTR A CT Keywords: Ocher Paleoindian Clovis Provenance ICP-OES Transport Red ocher (hematite) is a ferrous iron oxide mineral commonly used by Paleoindians in a variety of contexts, but its significance in mobile toolkits is not well understood. Here we demonstrate the importance of ocher to Paleoindians by determining the distance they were willing to carry it. The La Prele Mammoth site (48CO1401) is an approximately 13,000-year-old mammoth processing site with an associated campsite that contains scattered ocher nodules and a prominent ocher stain. The geochemical signature of La Prele ocher was established and compared to four natural ocher sources in Wyoming using inductively coupled plasma optical emission spectroscopy (ICP-OES). Ocher from La Prele was sourced to the Powars II ocher quarry approximately 100 km away. Paleoindians were willing to carry ocher for long distances, suggesting that red ocher was an important constituent of the Clovis mobile toolkit, while supporting the notion that Clovis maintained large territories. This geochemical analytical technique has potential for establishing geologic sources of ocher from other regions and time periods. 1. Introduction It is well-documented that members of one of the earliest, if not the earliest North American culture, Clovis, regularly transported chipped stone raw materials long distances, sometimes exceeding several hundred kilometers straight-line distance (Boulanger et al., 2015:551; Ellis, 2011:397; Gauthier et al., 2012: 2446; Hoard et al., 1992:663; Holen, 2010:299). Paleoindian archaeologists have commonly argued that such long distance transport is evidence of embedded procurement by foragers with large territories (Binford, 1979: 259; Goodyear, 1989:4–6; Speth et al., 2013:114). However, others note these studies are biased toward bifaces and projectile points (Bamforth, 2002:58; Bamforth, 2009:154) and suggest that long distance transport may have resulted from specialized forays or trade by groups operating within smaller territories (Speth et al., 2013:129). If the latter is correct, then our reconstruction of Clovis hunters as maintaining large territories may be exaggerated if not altogether wrong. Considering the potential bias imposed by using bifaces and projectile points to reconstruct the size of Clovis foraging territories, it may be useful to incorporate other artifact classes into the study of Clovis mobility. One such raw material commonly found in Clovis and other early Paleoindian sites is earthy ⁎ hematite, a soft ferrous iron oxide mineral (Fe2O3) commonly referred to as ‘red ocher’. Hematite pigments are one of the oldest known media for cultural expression (Henshilwood et al., 2002; Roper, 1991:296) and were commonly used in the Upper Paleolithic of the Old World (Haynes Jr, 1987:85–86; Erlandson et al., 1999:518; Roper, 1991:289; Roebroeks et al., 2012; Morrow, 2016:19). When people migrated to the New World from Asia, they almost certainly carried with them the tradition of using ocher for functional, aesthetic, and ritual purposes. Paleoindians used ocher in a variety of contexts, including in burials, such as Montana's Anzick site, Colorado's Gordon Creek, New Mexico's Arch Lake, and Minnesota's Browns Valley (Roper, 1991:291; Morrow, 2016:39–42), non-mortuary ritual or artistic expression (Roper, 1991:292; Frison and Stanford, 1982; Morrow, 2016:49, Stafford et al., 2003:88), and in domestic spaces, perhaps for treating leather clothing (Bement, 1999; Tankersley et al., 1995; Ruth, 2013; Frison and Stanford, 1982). Paleoindian artistic expression is most commonly conveyed through utilitarian objects, such as stone tools (Morrow, 2016:19), although organic objects that may have been used for artistic expression are rare in Paleoindian assemblages. Ocherpainted animal bones and ocher-covered stone tools show that Paleoindians used ocher as a medium for artistic expression or ritual at the Corresponding author. E-mail address: sanzarzycka@gmail.com (S.E. Zarzycka). https://doi.org/10.1016/j.jasrep.2019.05.001 Received 13 January 2019; Received in revised form 24 April 2019; Accepted 2 May 2019 2352-409X/ © 2019 Published by Elsevier Ltd. Journal of Archaeological Science: Reports 25 (2019) 519–529 S.E. Zarzycka, et al. Sheaman site, the Cooper bison kill site, and the Powars II site (Roper, 1991:292; Frison and Stanford, 1982:144–145; Frison et al., 2018:9; Morrow, 2016:27; Bement, 1999; Tankersley et al., 1995:185; Stafford et al., 2003:88). Lastly, hematite was used in domestic contexts by Paleoindians. Red ocher occurs as nodules in Paleoindian campsites, as smears on grinding slabs, and as stains covering living surfaces (Bement, 1999:179; Tankersley et al., 1995:185; Ruth, 2013; Frison and Stanford, 1982). Large stains of ocher on floors are argued to be evidence of ocher-stained hides used as floor-coverings (Ruth, 2013:224). Despite its common occurrence in Paleoindian sites, ocher has seldom been a topic of Paleoindian mobility studies due to the difficulty of sourcing the mineral. One previous study demonstrated Paleoindian ocher transport over a local distance of several miles (Tankersley et al., 1995), but it has yet to be determined if ocher was also transported over equally long distances alongside lithic raw material. For example, was the ocher covering the Anzick burial transported with the distant Phosphoria and porcellanite projectile points in the Anzick assemblage (Jones, 1996), or procured from a more local source? Should we expect ocher found in domestic or utilitarian contexts to differ in provenance from ocher found in sacred contexts? How might ocher provenance alter or confirm our view of Paleoindian mobility? Toward answering such questions, we use geochemical sourcing to study the provenance of a large red ocher stain found in a domestic context at the La Prele Mammoth site. We first describe the La Prele Mammoth site, focusing on the evidence for ocher use, contextualized with micromorphological analysis. Second, we discuss possible geologic sources of ocher. We then describe the analytical method, ICP-OES, to geochemically source the site's ocher. Lastly, we discuss the significance of the analyses to Paleoindian mobility and transport studies. Fig. 2. Map of Block B showing the locations of the hearth feature and ocher stain relative to piece-plotted nodules of ocher (small circles). at comparable stratigraphic positions in context with the mammoth remains, and proboscidean proteins recovered from artifacts in both Block A and B support their contemporaneity despite previous questions (Byers, 2002). Dates on the site vary, likely due to contaminants from local biogeochemistry (Mackie et al., 2016:24). But the recovery of a Clovis point in 2017, and recent dates both confirm that the site is Clovis and dates to approximately 12,925 cal years BP (Deviese et al., 2018; Mackie et al., 2016). We currently interpret the site as a mammoth kill with an associated campsite occupied by Clovis foragers for a brief duration. The focus here will be one of the activity areas, Block B, located approximately 12 m south of the mammoth remains (Fig. 1). Excavation of Block B revealed a large ocher stain associated with hearth feature, at least seven flake tools, at least three bone needles, a bone bead, bison remains, and several hundred flakes (Fig. 2). Excavators plotted over 1600 pieces of solid ocher within and around the stain (Fig. 2, Fig. 3a). The ocher stain includes pink to red sediments forming an oval shape, covering an approximately 2 × 1.5 m area with an average thickness of 1.1. La Prele Mammoth site The La Prele Mammoth site (48CO1401) is located along La Prele Creek, a perennial tributary of the North Platte River outside of Douglas, Wyoming. The initial excavations in 1987 (herein, ‘Block A’) uncovered the partial remains of a subadult Colombian Mammoth (Mammuthus columbi) along with a stone tool, hammerstone, and debitage (Byers, 2002; Mackie et al., 2016:2). Renewed investigations at the site between 2014 and 2017 expanded the site's boundaries considerably and identified two additional artifact concentrations south (Block B) and west (Block C) of Block A (Fig. 1). All artifacts are located Fig. 1. Site map showing excavation areas. Block A contains the mammoth remains. Block B contains the large ocher stain and the great majority of the ocher. 520 Journal of Archaeological Science: Reports 25 (2019) 519–529 S.E. Zarzycka, et al. Fig. 3. a. A sample of ocher nodules recovered from excavations. b. Photograph of the ocher-stained portion of Block B looking south. Color has been enhanced using Dstretch software. Nails are spaced 1 m apart. Fig. 4. a. Ocher stain in profile. Excavation wall is 1 m in width. b. Scan of thin section in reflected light from the ocher-rich layer. Note the diffuse concentration of the reddish material in the upper left, as well as rounded grains of ochre, especially in the upper part of the thin section. Thin section measures 50 × 75 mm. c. Photomicrograph of sample LPM-2 from within the ocher layer, showing rounded sand-sized pieces of ochre in the upper part and very fine-grained silt-sized ochrous material distributed within the sandy silty matrix. PPL. 15 to 20 cm (Fig. 3b). The eastern edge of the ocher stain was truncated by stream erosion (Fig. 4a). Micromorphological analysis of thin sections from above and within the ocher-rich layer showed the ocher as either discrete sand-sized detrital clasts of hematite, or reddish, dusty, silt-sized grains distributed throughout the matrix. In these dusty domains rich hematite can be locally concentrated (Fig. 4b, c). The clastic nature of these grains demonstrates that they are not the result of secondary hematite formation. In addition, the overall lenticular and very well-defined shape of the ocher-rich layer and the lack of internal bedding (Fig. 4a) show that the layer was not part of the active fluvial system, sensu strictu. In all, the micromorphological data show no evidence that the ochre is a result of groundwater or vertical translocation of fine particles or sand-size grains down the profile and thus, the ocher is not likely created by natural hydrogeological processes. The exact functional association of the ocher stain is unknown but, based on the diversity of material present and artifact types, we hypothesize that this area is likely a domestic space and therefore, the ocher may have served some domestic function. The bone needles, choppers, and scrapers found within excavation Block B suggests ocher was used within the same context as hide processing activities. Ocher has been used in hide processing ethnographically by several North American and African groups (Ruth, 2013:224). Large stains of ocher on floors are thought to be remains of ocher-stained hides used as floor-coverings (Ruth, 2013:224). The shape and density of ocher in Block B could be the result of hide-processing or other domestic activities, however this is merely speculation. Ocher has also been identified in other portions of the site (Block A and C), but in smaller quantities than what was excavated in Block B. The Block B flake tools and vast majority of debitage were produced from varieties of chert and quartzite likely procured from bedrock sources in the Hartville Uplift, 100 km downstream along the North Platte (Mackie et al., 2016:34). Thus, we hypothesized that the Block B ocher may have been quarried from this region as well, specifically from the Powars II ocher quarry near Hartville, Wyoming (Frison et al., 2018). The ocher source at Powars II is known to have been used by early Paleoindians, both because it has been previously tied to Paleoindian sites through sourcing (Tankersley et al., 1995) and because the site contains an abundance of Paleoindian artifacts, including Clovis projectile points (Frison et al., 2018). Given the Block B chipped stone raw material source and the known significance of Powars II as an ocher source, Powars II seems a likely place of origin for the La Prele ocher. However, archaeologists currently have little reason to assume that ocher was transported in the same way as chipped stone tools. It is entirely possible, and perhaps likely, for ocher to have been transported to the La Prele site from other nearby ocher sources or distant sources 521 Journal of Archaeological Science: Reports 25 (2019) 519–529 S.E. Zarzycka, et al. Table 1 Ocher sample information and depositional context. Sample source Sample ID Sample type Geologic formation Age of formation Depositional context Powars Powars Powars Powars Powars Powars Powars Powars Powars Powars Rawlins Rawlins Rawlins Rawlins Rawlins Rawlins Rawlins Rawlins Rawlins Rawlins La Prele B Block La Prele B Block La Prele B Block La Prele B Block La Prele B Block La Prele C Block La Prele C Block La Prele Creek La Prele Creek La Prele Trench Sheep Mountain P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 RR1 RR2 RR3 RR4 RR5 RR6 RR7 RR8 RR9 RR10 FS 1849 FS 2133 FS 3785 FS 3937 FS 4412 FS 6615 FS 6629 LC1 LC2 LP Trench SM Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Archaeological Sample Archaeological Sample Archaeological Sample Archaeological Sample Archaeological Sample Archaeological Sample Archaeological Sample Possible Geologic Source Possible Geologic Source Possible Geologic Source Possible Geologic Source Good Fortune Schist Good Fortune Schist Good Fortune Schist Good Fortune Schist Good Fortune Schist Good Fortune Schist Good Fortune Schist Good Fortune Schist Good Fortune Schist Good Fortune Schist Flathead Sandstone and Madison Limestone Flathead Sandstone and Madison Limestone Flathead Sandstone and Madison Limestone Flathead Sandstone and Madison Limestone Flathead Sandstone and Madison Limestone Flathead Sandstone and Madison Limestone Flathead Sandstone and Madison Limestone Flathead Sandstone and Madison Limestone Flathead Sandstone and Madison Limestone Flathead Sandstone and Madison Limestone Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Casper and Amsden Formation Sandstone Precambrian Precambrian Precambrian Precambrian Precambrian Precambrian Precambrian Precambrian Precambrian Precambrian Cambrian, Mississippian Cambrian, Mississippian Cambrian, Mississippian Cambrian, Mississippian Cambrian, Mississippian Cambrian, Mississippian Cambrian, Mississippian Cambrian, Mississippian Cambrian, Mississippian Cambrian, Mississippian Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Pennsylvanian, Mississippian Metamorphic Metamorphic Metamorphic Metamorphic Metamorphic Metamorphic Metamorphic Metamorphic Metamorphic Metamorphic Sedimentary Sedimentary Sedimentary Sedimentary Sedimentary Sedimentary Sedimentary Sedimentary Sedimentary Sedimentary Redeposited alluvial Redeposited alluvial Redeposited alluvial Redeposited alluvial Redeposited alluvial Redeposited alluvial Redeposited alluvial Redeposited alluvial Redeposited alluvial Redeposited alluvial Sedimentary cobbles cobbles cobbles cobbles cobbles cobbles cobbles cobbles cobbles cobbles and Sheep Mountain, were chosen because of their proximity to the site and to detect natural occurrences of ocher at the La Prele Mammoth site. like Sheep Mountain or the Rawlins Red Paint Mine (Fig. 4). We used geochemical and statistical analyses to determine the source of the La Prele ocher. 2. Ocher samples 2.2.1. Powars II ocher mine Ocher has been discovered in various Paleoindian archaeological sites, however the Powars II site is the only confirmed Paleoindian aged ocher mine (Stafford et al., 2003:71; Tankersley et al., 1995:186; Frison et al., 2018:2). The Powars II site is located within the Sunrise Mine outside of Hartville, Wyoming (Fig. 5). Evidence for Paleoindian ocher quarrying include 2 m mine tailing interspersed with thousands of Paleoindian artifacts, including bones used for digging, hammerstones, and flakes with damage potentially incurred as a result of scraping ocher from nodules (Stafford, 1990; Stafford et al., 2003:74; Frison et al., 2018:14). Mining at Powars II occurred between approximately 13,000 and 11,000 years ago based on diagnostic projectile points recovered from the site spanning the Clovis through Hell Gap cultural complexes. Published dates from the site are from a redeposited context and slightly post-date the site's occupation, suggesting contamination of the dated bone or mixing with later occupations (Frison et al., 2018). Ongoing work at the site is focused on refining its chronology of use. Ocher from Sunrise Mine varies from specular to earthy, with colors ranging from dusky red to pale reddish brown (Tankersley et al., 1995:188). Specular hematite has a metallic luster and is typically silver, gray or black in color. The earthy, softer variety is referred to as red ocher, and is the variety most commonly found in archaeological sites. The red ocher from Powars II is located within the Precambrian Good Fortune Schist formation (Table 1; Southerland and Cola, 2015:20). This metamorphically-derived hematite is readily available along the surface, but deposits were also mined. In either case, Powars II was an important ocher quarrying site for Paleoindians. It is likely that excursions to the area were planned events, during which large quantities of ocher could be quarried (Stafford et al., 2003:86). The Powars II site is located 108 km downriver along the North Platte 2.1. Ocher samples from the La Prele Mammoth site All ocher encountered in situ was mapped and collected from Blocks B and C (Fig. 1), resulting in over 1600 mapped ocher pieces. All other ocher pieces were collected from the 1/16 inch water screen, resulting in over 37,000 total nodules. Of the ocher recovered from Block B, we chose five samples for geochemical analysis (Table 1). All five ocher samples were large nodules located within the ocher stain. Many ocher nodules found within Blocks B and C were too small to analyze individually using ICP-OES (see Fig. 3a). Hence, individual ocher samples were chosen based on size. Ocher within the stain is soft and smearable with a dark, rusty red hue similar in color and texture to both the Rawlins Red and Powars II samples. Excavation of Block C in 2017 revealed the presence of ocher as well, but here the ocher was sparse and scattered, with no clear stain like that of Block B. This ocher is dissimilar in texture to the ocher found in Block B and is most similar in texture and color to naturally occurring hematite found within La Prele Creek. Ocher from Block C is coarser-grained and appears to be derived from highly oxidized sandstone alluvial cobbles. Ocher sampled from Block C includes field specimens 6615 and 6629 (Table 1). 2.2. Ocher sources in Wyoming Samples from five possible geological sources of ocher were collected for geochemical comparison to the La Prele Mammoth archaeological ocher. The two non-local sources, Powars II and Rawlins Red, were selected because of their well-known use in prehistoric and historic time periods. Three local sources, La Prele Creek, La Prele Trench, 522 Journal of Archaeological Science: Reports 25 (2019) 519–529 S.E. Zarzycka, et al. Fig. 5. Map of the study area showing locations of the La Prele Mammoth site and locations of geologic sources included in the study. terrace several meters from the excavation blocks. Within this trench, a large nodule of ocher was sampled from the south stratigraphic profile (Fig. 1). This ocher sample is labeled La Prele Trench (Table 1). The hematite is similar in texture to the Powars II source and is fine grained and highly smearable but much lighter in color. Although it is likely that ocher within Block B is the result of human activity due its defined oval-shape (Fig. 3b), the possibility of naturally derived ocher from redeposited alluvial cobbles cannot be excluded. It is entirely likely that hematitic nodules from the Casper or Amsden formations located upstream from the La Prele Mammoth site, could have been redeposited by La Prele Creek, as the site is dominated by overbank alluvial deposits. Thus, two samples were taken from the gravel bars along La Prele Creek; these samples are labeled the La Prele Creek samples (Table 1). The La Prele Creek, La Prele Trench, and Sheep Mountain samples constitute possible local geologic sources of ocher. (85 km straight-line distance) from the La Prele Mammoth site (Fig. 5). 2.2.2. Rawlins Red Rawlins Red Paint mine is located approximately 224 km upriver along the North Platte (183 km straight-line distance) from the La Prele site (Fig. 5). This source is located just north of Rawlins, Wyoming and is rumored to be the original paint source for the Brooklyn Bridge in the late 1800s (Lovering, 1929). The Rawlins Red source is one of the most well known in Wyoming and was used throughout the historic period. Though Rawlins Red does not have evidence of prehistoric ocher mining like Powars, it contains abundant earthy and specular hematite similar to Powars II. Ocher in Red Rawlins Paint mine is located within the upper half of the Cambrian Flathead Sandstone formation, and the lower portion of the Mississippian age Madison Limestone (Table 1; Southerland and Cola, 2015:54). Oolitic chert from the Green River Formation outcrops is located approximately 60 km southwest of Rawlins, Wyoming. A flake of similar oolitic chert was found within Block C of the La Prele excavation is consistent with cherts found outside of Rawlins, Wyoming. 3. Archaeometric background Hematite (Fe2O3) may contain trace elements that reflect the local depositional environment. Like any crystal lattice mineral, hematite is subject to chemical imperfections as a result of mixing with other rocks, clay, and sand (Dana, 1932:484; Tankersley, 1995:186). Common inclusions are quartz (SiO2), goethite (FeOOH), kaolinite (Al2Si2O5(OH)4), and various oxides (e.g., ZnO or Al2O3). The presence of such imperfections is reflected in trace element concentrations that identify the unique chemical signature or “fingerprint” of an ocher sample (Moyo et al., 2016:23; Popelka-Filcoff et al., 2007, 2008, 2012). The provenance of ocher can be determined by comparing the chemical signature of unprovenanced samples to the signatures of known geological sources. Past ocher provenance studies have used a variety of geochemical analyses but they have largely focused on determining major element distribution patterns rather than trace element signatures (Cavallo et al., 2017a, 2017b; Bernatchez, 2008; Gialanella et al., 2011; Moyo et al., 2016; Tankersley, 1995). 2.2.3. Local sources There are also local sources of ocher near La Prele. The Sheep Mountain hematite source is located outside of Douglas, Wyoming (Fig. 5). Sheep Mountain is the closest known source to the La Prele site, at approximately 29 km south along the North Platte River. The ocher from Sheep Mountain has a very different texture than ocher from Rawlins or Powars II. It is an earthy hematite, with a far coarser texture and lighter hue. Ocher from Sheep Mountain is located within the Casper and Amsden Sandstone Formations (Southerland and Cola, 2015:64). The Casper Formation was deposited in the Pennsylvanian, while the Amsden Formation dates to the Mississippian (Sando et al., 1975). With the landowner's permission, samples of ocher were taken from the area for comparison. During the 2017 field season, two geoarchaeological trenches were excavated by backhoe. One of these trenches was located on the alluvial 523 Journal of Archaeological Science: Reports 25 (2019) 519–529 S.E. Zarzycka, et al. Table 2 ICP-OES results of major elements. All concentrations are corrected for dilution, and reported in ppm. Sample source Sample ID Ca ppm Fe ppm Powars Powars Powars Powars Powars Powars Powars Powars Powars Powars Rawlins Rawlins Rawlins Rawlins Rawlins Rawlins Rawlins Rawlins Rawlins Rawlins La Prele B Block La Prele B Block La Prele B Block La Prele B Block La Prele B Block La Prele C Block La Prele C Block La Prele Creek La Prele Creek La Prele Trench Sheep Mountain Detection limit (ppb) P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 RR1 RR2 RR3 RR4 RR5 RR6 RR7 RR8 RR9 RR10 FS 1849 FS 2133 FS 3785 FS 3937 FS 4412 FS 6615 FS 6629 LC1 LC2 LP Trench SM 1942.0 1622.4 1848.4 948.0 5260.0 1476.0 2352.0 894.0 842.8 510.0 2480.0 8792.0 12,060.0 4592.0 2009.2 1122.4 696.4 5828.0 3488.4 1818.8 2214.8 3836.0 21,288.0 1736.8 872.0 188.4 129.2 167.6 184.8 14,080.0 2714.8 8 33,848.0 44,520.0 33,032.0 31,344.0 53,760.0 27,380.0 21,388.0 60,160.0 45,360.0 9584.0 16,064.0 16,088.0 22,600.0 21,712.0 22,200.0 41,280.0 8844.0 48,360.0 17,352.0 49,840.0 22,492.0 47,840.0 29,804.0 49,800.0 63,400.0 1030.8 15,440.0 17,312.0 24,168.0 545.2 3973.2 2 Mg ppm Na ppm Si ppm 368.8 432.0 488.8 180.0 71.2 432.4 769.6 150.0 225.2 64.4 948.8 4636.0 4488.0 1104.4 620.4 282.0 564.0 480.0 1121.2 602.4 166.0 106.4 345.2 214.4 119.2 134.4 232.8 90.0 153.2 7336.0 2224.8 1 51.2 61.2 51.6 55.2 53.2 63.2 66.8 58.4 128 81.6 109.2 86.8 73.6 83.2 64.8 130.4 82.8 421.6 57.6 116.8 120 160.4 142.8 126.8 113.2 161.6 246 66.8 174.4 232.8 120.4 9 202.0 174.8 168.4 158.0 161.6 167.6 238.8 156.4 184 241.2 500.4 200.4 232.0 537.2 244.8 365.6 200.8 219.2 162.4 587.2 96.0 172.4 140.4 152.8 139.2 338.4 332.8 142.4 222.8 358.4 759.6 25 in provenance studies, ICP-OES1 was considered the best option due to its short sample preparation time, small sample size requirement, cost effectiveness, and ability to detect trace elements in the ppb range. A pilot study of geochemistry of La Prele ocher used both ICP-OES and XRD analyses (Kent, 2017). Though XRD was useful in distinguishing geologic sources of ocher, ICP-OES had lower detection limits. We refined the sample preparation methods from the pilot study then expanded the project to include more geological and archaeological samples. Geochemical analyses such as X-ray diffraction (XRD), particle-induced X-ray emission (PIXE), X-ray powder diffraction (XRPD), and energy dispersive X-ray fluorescence (ED-XRF) are useful in determining the bulk geochemistry of ocher in the parts per million (ppm) range. Many ocher provenance studies have shown promising results using the above analyses (Cavallo et al., 2017a, 2017b; Bernatchez, 2008; Gialanella et al., 2011; Moyo et al., 2016), including two specifically relevant to Powars II (Erlandson et al., 1999; Tankersley, 1995). Archaeological ocher samples from Nelson Bay Cave, South Africa were successfully differentiated using XRD and PIXE analyses and highlighted future implications of provenance studies against geologic deposits (Bernatchez, 2008:3). Use of XRD analysis has also proved highly effective in provenance studies in Italy where it was used to test whether red ocher from archaeological sites was hematite (Fe2O3), or heataltered goethite (FeOOH) (Cavallo et al., 2017a, 2017b:763; Gialanella et al., 2011:953). Tankersley et al. (1995:192) used XRPD to infer mineral composition and were able to source ocher from a Paleoindian occupation in Hell Gap, Wyoming to the Powars II ocher mine. Geochemical techniques that can determine bulk geochemistry in the parts per billion (ppb) range depending upon sample preparation methods include but are not limited to: inductively coupled plasma optical emission spectroscopy (ICP-OES), inductively coupled plasma mass spectrometry (ICP-MS), and instrumental neutron activation analysis (INAA). Using INAA, Eiselt et al. (2011:3027) successfully determined the provenance of ocher on pottery. ICP-OES analysis was useful for differentiating the chemical signatures and provenance of ocher in Blombos Cave (Moyo et al., 2016:23). A first attempt to use ICP-MS to establish ocher provenance was also successful (Green and Walting, 2007), with trace element data supporting the results of the major element concentrations derived using X-ray analytical techniques (Green and Walting, 2007:851). We used ICP-OES to determine the provenance of red ocher from the La Prele Mammoth site. Although other analytical techniques are useful 4. Materials and methods 4.1. Sample selection When selecting samples from the possible geologic ocher sources (Table 1) we attempted to represent the variability of the parent deposits. We collected samples from multiple locations at each source site along stratigraphic layers. We collected large, unconsolidated hematite nodules were collected using a trowel and the location of each sample was recorded using a GPS (Table 1). During excavation of Blocks B and C of the La Prele Mammoth site, we plotted the location of each ocher nodule was piece plotted using a total station. We selected only seven archaeological samples for ICP-OES analysis due to the destructive nature of the sample preparation process. We analyzed a total of seven archaeological samples from La Prele Mammoth site and twenty-three samples from possible geologic sources were analyzed using ICP-OES (Table 2). 1 Inductively coupled plasma optical emission spectroscopy (ICP-OES) is a geochemical analytical technique which requires aqueous samples. The sample is introduced to plasma, causing ionization. Elements concentration are detected based on their specific emission spectra. Each element's emission line is then quantified based on the intensity of these emission spectra. 524 Journal of Archaeological Science: Reports 25 (2019) 519–529 S.E. Zarzycka, et al. Table 3 ICP-OES results of trace elemental concentrations. All concentrations are corrected for dilution, and reported in ppb. Critical elements highlighted in gray. Sample Source Sample ID Al ppb B ppb Ba ppb Co ppb Cr ppb Cu ppb K ppb Mn ppb Nb ppb Ni ppb Pb ppb Sr ppb Ti ppb Zn ppb Powars P1 817600.0 13584.0 6604.0 7640.0 1700 7528.0 381360.0 68600.0 10216.0 30392.0 <LOD 7168.0 51800.0 65200.0 Powars P2 1075200.0 13308.0 13928.0 7060.0 2824 12864.0 466800.0 57440.0 12888.0 39876.0 <LOD 7308.0 64280.0 62800.0 Powars P3 771600.0 10252.0 13988.0 5624.0 2119 22476.0 312120.0 50080.0 9288.0 30312.0 <LOD 5496.0 51840.0 50400.0 Powars P4 571600.0 8904.0 7564.0 3432.4 932.4 17676.0 169880.0 37764.0 10548.0 19856.0 <LOD 2498.0 62920.0 42400.0 Powars P5 357280.0 12308.0 10724.0 14112.0 <LOD 22624.0 74400.0 139480.0 23020.0 59120.0 <LOD 2627.2 85000.0 174000.0 Powars P6 682800.0 8692.0 10912.0 4072.0 1899 5728.0 187640.0 35240.0 10996.0 21352.0 <LOD 4448.0 67160.0 38400.0 Powars P7 935600.0 17976.0 24632.0 9112.0 1917 48800.0 241240.0 48120.0 7808.0 21752.0 6932.0 5232.0 71560.0 54400.0 Powars P8 502000.0 11036.0 8064.0 8132.0 1929 4308.0 131560.0 65280.0 16816.0 48120.0 <LOD 5608.0 41520.0 67200.0 Powars P9 508400.0 11776.0 5624.0 7376.0 1434 11692.0 195280.0 54600.0 14052.0 38056.0 <LOD 6640.0 35608.0 55600.0 Powars P10 267800.0 11544.0 3503.2 2064.4 7312 11352.0 166160.0 1933.2 2435.6 4548.0 <LOD 1482.4 62880.0 2400.0 Rawlins RR1 741600.0 6240.0 26844.0 1640.8 16288 418.0 318240.0 Rawlins RR2 1782800.0 20376.0 32736.0 2811.2 23952 31128.0 Rawlins RR3 1603600.0 17576.0 11112.0 2782.0 13480 Rawlins RR4 3263200.0 22660.0 26868.0 2940.8 4984 Rawlins RR5 3224000.0 27704.0 30204.0 3213.6 Rawlins RR6 584800.0 13688.0 15984.0 3712.4 Rawlins RR7 1414400.0 21768.0 7948.0 Rawlins RR8 1109200.0 15652.0 Rawlins RR9 821200.0 11576.0 Rawlins RR10 617200.0 La Prele B Block FS 1849 34756.0 4804.0 3162.4 12884.0 3735.2 138200.0 800.0 820400.0 180320.0 3641.6 5600.0 13472.0 4572.0 79360.0 <LOD 16740.0 767200.0 101760.0 3641.6 5996.0 9992.0 6188.0 84480.0 <LOD 8324.0 1104400.0 36408.0 5344.0 5264.0 10656.0 10632.0 171760.0 8000.0 23968 30424.0 1335600.0 16976.0 5632.0 6952.0 9708.0 5752.0 135120.0 48000.0 24752 8056.0 169000.0 19704.0 6128.0 5136.0 8896.0 4780.0 199720.0 8400.0 2152.0 30508 14112.0 583600.0 20204.0 11080.0 4012.0 7444.0 2593.2 72320.0 16000.0 13432.0 3849.2 16896 9980.0 260440.0 88360.0 11080.0 6824.0 15660.0 4480.0 82760.0 8800.0 15816.0 1162.8 31328 13992.0 181400.0 36732.0 2402.0 4028.0 10880.0 4024.0 74000.0 3600.0 5352.0 16640.0 11772.0 11900 70840.0 197760.0 71080.0 13388.0 11836.0 9180.0 2050.4 109760.0 5200.0 288560.0 1503.2 10836.0 2847.2 <LOD 3446.0 44640.0 17488.0 4304.0 10868.0 <LOD 4652.0 10596.0 6800.0 La Prele B Block FS 2133 282320.0 3461.6 8920.0 7644.0 <LOD 14280.0 39576.0 75200.0 17568.0 34488.0 <LOD 4928.0 27708.0 60400.0 La Prele B Block FS 3785 521600.0 13432.0 26712.0 47040.0 878 96920.0 117120.0 133240.0 11532.0 101000.0 9460.0 15696.0 34620.0 632000.0 La Prele B Block FS 3937 474800.0 4872.0 37884.0 2903.6 <LOD 22108.0 90120.0 16056.0 16520.0 23916.0 <LOD 10396.0 28644.0 17200.0 La Prele B Block FS 4412 423600.0 18452.0 37092.0 7336.0 <LOD 71280.0 63560.0 27780.0 20700.0 32936.0 7120.0 6724.0 22216.0 48800.0 La Prele C Block FS 6615 296720.0 161.2 3081.6 789.6 4608 3488.4 150360.0 26660.0 <LOD 4808.0 <LOD 1230.4 22620.0 1600.0 La Prele C Block FS 6629 743600.0 794.4 5344.0 2832.8 7044 6296.0 132280.0 46200.0 3981.2 5868.0 7276.0 2712.4 48560.0 20400.0 La Prele Creek LC1 776000.0 5608.0 28460.0 1162.0 <LOD 4244.0 74480.0 24908.0 5380.0 2129.2 <LOD 2552.0 72320.0 <LOD La Prele Creek LC2 1478000.0 13008.0 179240.0 4192.0 457.2 14444.0 173240.0 14856.0 6740.0 5268.0 7208.0 6400.0 66920.0 20000.0 La Prele Trench LP Trench 268280.0 3276.4 3687.2 1289.2 1328 3012.0 107360.0 6988.0 <LOD 2844.8 <LOD 4980.0 19188.0 <LOD 2080400.0 2768.0 8516.0 2473.6 3412 3476.8 400800.0 67000.0 406.4 5260.0 2231.2 2236.8 67240.0 6000.0 2 8 0.5 2 1 1 4 0.3 2 5 17 0.02 0.3 3 Sheep Mountain SM Detection limit (ppb) 4.2. Geochemical analysis elements examined, twelve have some discriminatory power with six being most useful (Tables 2 and 3) including titanium, chromium, nickel, zinc, niobium, and lead. Aluminum, boron, barium, cobalt, iron, and potassium also show significant differences among groups. Hematite from the Powars source is unique in that it exhibits relatively low concentrations of aluminum and high concentrations of zinc and nickel. The Rawlins source stands out for high concentrations of chromium, lead, and titanium. Local sources can be differentiated from the other two by low concentrations of cobalt, iron, and niobium. Comparing raw elemental concentrations for the archaeological samples to those of the source areas, Block B samples consistently group with Powars ocher (Fig. 6). With respect to those attributes that are unique to Powars hematite, the Block B samples are identical showing high concentrations of zinc and nickel with low concentrations of aluminum. It is also clear that Block B samples show none of the unique attributes of Rawlins Red or Local hematite. Taken together, this would suggest that Block B samples are derived from the Powars source. Block C samples, however, show much greater affinity for local sources (Fig. 6). Block C archaeological samples exhibit all attributes unique to local hematite, namely low concentrations of cobalt, iron, and niobium. Block C samples, like those from Block B, show no attributes of Rawlins Red ocher. Block C samples also lack two of the attributes unique to the Ocher samples were ground with a mortar and pestle to pass through a 125 μm sieve. 25 mg of powdered ocher was combined with 25 mL of 12.1 M ACS grade hydrochloric acid and heated for 30 min at a temperature of 80 °C while stirring constantly. The samples were then centrifuged at 2200 rpm for 5 min. This leaching process does not fully dissolve the ocher samples, leaving a pellet at the bottom of the centrifuge tube, which we estimate to be roughly 10% of the original mineral. A 2.0 mL aliquot of the leachate was diluted with 8 mL of 18.2 meg-ohm DI water to a final volume of 10.0 mL. Leachate samples were analyzed on a Perkin Elmer Optima 8300 ICP-OES at the Geochemistry Analytical Laboratory at the University of Wyoming. More information on this geochemical analysis is included in Appendix 1. 5. Results Samples were evaluated for the presence of twenty-one different elements (Tables 2 and 3). In Table 4 we present 90% confidence intervals for elemental concentrations of the geologic sample groups. For each element we performed an analysis of variance to determine which elements can be used to distinguish between groups (Table 4). Of the 21 525 Journal of Archaeological Science: Reports 25 (2019) 519–529 S.E. Zarzycka, et al. Table 4 Elemental concentrations for each geologic source group as 90% confidence intervals (x ± 1.65 s). Element (Conc.) Powars Rawlins Local ANOVA F (p)a Al (ppb) B (ppb) Ba (ppb) Ca (ppb) Co (ppb) Cr (ppb) Cu (ppb) Fe (ppm) K (ppb) Mg (ppm) Mn (ppb) Na (ppm) Nb (ppb) Ni (ppb) Pb (ppb) Si (ppm) Sr (ppb) Ti (ppb) W (ppb) Zn (ppm) 1622.5 ± 1052.72 29.8 ± 11.07 26.4 ± 24.84 4.4 ± 5.59 17.2 ± 14.04 5.5 ± 7.96 41.3 ± 53.71 90.1 ± 62.71 581.6 ± 496.48 0.8 ± 0.92 139.6 ± 144.22 0.2 ± 0.1 29.5 ± 22.68 78.3 ± 64.71 13.1 ± 3.79 0.5 ± 0.13 12.1 ± 8.45 148.6 ± 60.13 6.7 ± 11.49 0.2 ± 0.18 3790.5 ± 4120.89 40.6 ± 29.79 49.4 ± 35.65 10.7 ± 15.14 9 ± 12.34 49.5 ± 35.04 51 ± 83.18 66.1 ± 60.1 1434.5 ± 1729.95 3.7 ± 6.78 151.6 ± 212.74 0.3 ± 0.44 16.8 ± 15.51 14.7 ± 10 27.2 ± 10.14 0.8 ± 0.66 12.2 ± 9.83 286.9 ± 185.05 2.3 ± 6.97 0 ± 0.06 2876.7 ± 3274.58 15.4 ± 19.5 137.4 ± 344.58 10.7 ± 27.38 5.7 ± 5.8 3.2 ± 6.33 15.7 ± 22.51 28.7 ± 45.87 472.4 ± 606.66 6.1 ± 14.04 71.1 ± 110.28 0.4 ± 0.29 6.8 ± 16.55 9.7 ± 6.72 9.7 ± 13.89 0.9 ± 1.13 10.1 ± 8.22 141 ± 102.89 1.5 ± 6.94 0 ± 0.06 3.46 (0.050)⁎ 5.29 (0.013)⁎ 4.52 (0.023)⁎ 1.44 (0.260) 4.52 (0.023)⁎ 28.341 (≪0.001)⁎⁎⁎ 1.13 (0.341) 4.24 (0.028)⁎ 4.340 (0.026)⁎ 2.61 (0.097) 0.87 (0.433) 2.12 (0.145) 6.41 (0.007)⁎⁎ 18.29 (≪0.001)⁎⁎⁎ 23.65 (≪0.001)⁎⁎⁎ 3.22 (0.061) 0.23 (0.79) 8.84 (0.002)⁎⁎ 2.00 (0.161) 8.76 (0.002)⁎⁎ a Analysis of variance test results with F- and p-statistics. Tests indicate which chemical elements best differentiate between geologic source areas. p < .05. ⁎⁎ p < .01. ⁎⁎⁎ p < .001. ⁎ Fig. 6. Concentrations of critical elements for the three geologic source and two excavation areas. Powars source, high concentrations of nickel and zinc, but like Powars show relatively low concentrations of aluminum. We also performed principal components analysis (PCA) using those elements that best distinguish between source areas. We excluded barium and potassium due to higher solubility and the possibility of their occurrence in samples as secondary salts. This analysis includes aluminum, boron, cobalt, chromium, iron, niobium, nickel, lead, titanium, and zinc. To perform PCA, we used the prcomp function in R v.3.53 (R Core Team, 2019). We standardized concentrations for each element were standardized to z-scores prior to PCA. Fig. 7 shows component loadings and scores for the first two principal components. We also calculated 95% confidence ellipses for each geologic source group. The PCA analysis partially confirms our source attributions from visual analysis, but there are important differences. Both Block C 526 Journal of Archaeological Science: Reports 25 (2019) 519–529 S.E. Zarzycka, et al. Fig. 7. Principle components analysis. a. Loadings for the first two principal components. b. Principal component scores for all samples. Ellipses are 95% confidence intervals for each geologic source group. samples fall comfortably within the range of variation for the Local group, although FS 6629 also falls within the Powars ellipse. Three of five Block B samples fall within the Powars group and only within that group. FS 1849 falls within the local group, although it is possible that this is due to the small sample size of that group. FS 3785 is an outlier possibly indicating that it is derived from an unknown source. That said, chemically it looks chemically similar to Powars samples except for unusually high concentrations of cobalt, nickel, and zinc. In sum, there are geochemical differences in hematites from the source areas studied that allow them to be differentiated chemically using ICP-OES. With respect to the archaeological samples, there is no evidence for the presence of ocher from the Rawlins Red source at the La Prele Mammoth site. We are confident that at least some of the ocher nodules from Block B were derived from the Powars source. The Block C ocher samples group well with the Local group, which suggests that they might not have a cultural origin, but could have been deposited by La Prele Creek. 7. Conclusion Considering the potential biases imposed by chipped stone toolkits in reconstructing Clovis foraging behavior, especially bifaces and projectile points, it is necessary to study other materials such as ocher. Thus, we turn to answer the following question, how might ocher provenance alter or confirm our view of Paleoindian mobility? Ocher from Block B of the La Prele Mammoth site is not from local geologic sources. Instead, it was transported approximately 100 km upriver from the Powars II site to the La Prele Mammoth site by Clovis foragers. The presence of Hartville Uplift chert within Block B further supports the idea of Paleoindian transport of lithic raw materials from near the Powars II locale to the La Prele Mammoth site, and largely supports the traditional view of Paleoindian raw material transport. Our results confirm Paleoindians' inclination to transport materials long-distances, however the materials they are willing to carry are not limited to lithics. While many Paleoindian archaeologists use chipped stone raw materials as evidence for long-distance transport, ocher can also map out Paleoindian networks, and contribute to discussion of whether Paleoindians acquired stone and other materials through embedded procurement, special trips, or trade. Ocher is a mineral that has no nutritional and arguably limited utilitarian value, which begs the question of why this material was moved so far across the landscape. Noneconomic reasons seem likely. The same cannot be said of lithic raw material, where there are unambiguous utilitarian reasons why people might carry it long distances. Given that hematite was transported a long distance, it also seems somewhat odd that so much of it was left at the site. The reason for this apparent contradiction, for now, will have to remain unanswered. However, this discussion speaks to the need for further work exploring the diversity of ways that foraging societies use iron oxides and other minerals and the archaeological correlates of those uses. 6. Discussion Various geochemical analytical techniques have proven useful for differentiating ocher sources (Cavallo et al., 2017a, 2017b; Bernatchez, 2008; Gialanella et al., 2011; Moyo et al., 2016; Erlandson et al., 1999; Tankersley, 1995). One such technique is ICP-OES analysis. Results from ICP-OES analysis suggests that ocher from Block B is geochemically similar to Powars II ocher. Powars II ocher consists of low Al, Cr, Pb, and Ti values and high Fe, Nb, Ni, and Zn values. Rawlins Red chemical signature consists of low Ni and Zn, high Al, Cr, Pb and Ti values, and a moderate amount of Nb. Local sources (Sheep Mountain, La Prele Creek, and La Prele Trench) contain high Al values and low Co, Ti, Pb and Zn values. Principal component analysis revealed that all samples from Block B correlate with Powars II samples (Table 4; Fig. 7). Though there is evidence that some ocher from Block B was transported from Powars II, Block C ocher was not. Ocher from Block C is likely naturally occurring Casper or Amsden formation hematitic sandstone redeposited by overbank flooding from La Prele Creek. Geochemical analytical techniques like ICP-OES have vast implications for archaeological provenance research in the future, as they have successfully proven that ocher can be differentiated by geochemical signatures. Acknowledgements This work was supported by funding from the June Frison Memorial Fund of the George C. Frison Institute of Archaeology and Anthropology. We thank the Amen family, the Baker family, and the Powars II site owners and researchers for allowing access for archaeological and geological sampling. Excavations at the La Prele Mammoth 527 Journal of Archaeological Science: Reports 25 (2019) 519–529 S.E. Zarzycka, et al. site were supported by University of Wyoming Archaeological Field School, the Quest Archaeological Research Program, the Wyoming Cultural Trust Fund, and the National Geographic Society (#9896-19). 171–180. Eiselt, B. Sunday, Popelka-Filcoff, Rachel S., Darling, J. Andrew, Glascock, Michael D., 2011. Hematite sources and archaeological ochres from Hohokam and O'odham sites in central Arizona: an experiment in type identification and characterization. J. Archaeol. Sci. 38, 3019–3028. Ellis, Christopher, 2011. Measuring Paleoindian range mobility and land-use in the Great Lakes/Northeast. J. Anthropol. Archaeol. 30, 385–401. Erlandson, Jon M., Robertson, J.D., Descantes, Christopher, 1999. Geochemical analysis of eight red ochres from Western North America. Am. Antiq. 64, 517–526. Frison, George C., Stanford, Dennis J., 1982. The Agate Basin Site: A Record of the Paleoindian Occupation of the Northwestern High Plains. Academic Press, New York. Frison, George C., Zeimens, George M., Pelton, Spencer R., Walker, Danny N., Stanford, Dennis J., Kornfeld, Marcel, 2018. Further insight into Paleoindian ocher use of the Powars II red ocher quarry (48PL330), Wyoming. Am. Antiq. 83, 485–504. Gauthier, Gilles, Burke, Adrian L., Leclerc, Mathieu, 2012. Assessing XRF for the geochemical characterization of radiolarian chert artifacts from northeastern North America. J. Archaeol. Sci. 39, 2436–2451. Gialanella, Stefano, Belli, R., Dalmeri, G., Lonardelli, I., Mattarelli, M., Montagna, M., Toniutti, L., 2011. Artificial or natural origin of hematite-based red pigments in archaeological contexts: the case of Riparo Dalmeri (Trento, Italy). Archaeometry 53, 950–962. Goodyear, A.C., 1989. A hypothesis for the use of cryptocrystalline raw materials among Paleo-Indian groups of North America. In: Ellis, Christopher J., Lothrop, Johnathan C. (Eds.), Eastern Paleoindian Lithic Resource Use. Westview Press, Boulder, CO, pp. 1–10. Green, Rachel L., Walting, R. John, 2007. Trace element fingerprinting of Australian ocher using laser ablation inductively coupled plasma-mass spectrometry (LA-ICPMS) for the provenance establishment and authentication of indigenous art. J. Forensic Sci. 52, 851–859. Haynes Jr., C.V., 1987. Clovis origin update. The Kiva 52, 83–93. Henshilwood, Christopher S., d'Errico, Francesco, Yates, Royden, Jacobs, Zenobia, Tribolo, Chantal, Duller, Geoffery A.T., Mercier, N., Sealy, Judith C., Valladas, Helene, Watts, Ian, Wintle, A.G., 2002. Emergence of modern human behavior: middle stone age engravings from South Africa. Science 295, 1278–1280. Hoard, Robert J., Holen, Steven R., Glascock, Michael D., Neff, Hector, Elam, J. Michael, 1992. Neutron activation analysis of stone from the Chadron Formation and a Clovis site on the Great Plains. J. Archaeol. Sci. 19, 655–665. Holen, Steven R., 2010. The Eckles Clovis site, 14JW4: a Clovis site in northern Kansas. Plains Anthropol. 55, 299–310. Jones, J. Scott, 1996. The Anzick Site: Analysis of a Clovis Burial Assemblage (Unpublished Master's Thesis). Department of Anthropology, Oregon State University, Corvallis. Kent, Meghan M., 2017. GeoChemical Sourcing of the La Prele Mammoth Kill Site (Honors thesis). University of Wyoming. Lovering, Thomas Seward, 1929. The Rawlins, Shirley, and Seminoe Iron-ore Deposits Carbon County, Wyoming. Government Printing Office, Washington. Mackie, Madeline E., Surovell, Todd A., Kelly, Robert, O'Brien, Matthew, Pelton, Spencer, 2016. The 2016 field season at the La Prele mammoth site. In: Prepared for: Quest Archaeological Research Program, (Unpublished). Morrow, Juliet E., 2016. Evidence of Paleoindian spirituality and ritual behavior: large thin bifaces and other sacred objects from Clovis and other Late Pleistocene-Early Holocene cultural contexts. Arkansas Archaeological Survey, Fayetteville 67, 18–65. Moyo, Stanley, Mphuti, Dikeledi, Curkowska, Ewa, Henshilwood, Christopher S., et al., 2016. Blombos cave: middle stone age ochre differentiation through FTIR, ICP OES, ED XRF, and XRD. Quat. Int. 404, 20–29. Popelka-Filcoff, Rachel S., Robertson, J.David, Glascock, Michael D., Descantes, Christopher, 2007. Trace element characterization of ochre from geological sources. J. Radioanal. Nucl. Chem. 272, 17–27. Popelka-Filcoff, Rachel S., Miska, Elizabeth J., Robertson, J.David, Glascock, Michael D., Wallace, Henry, 2008. Elemental analysis and characterization of ochre sources from southern Arizona. J. Archaeol. Sci. 35, 752–762. Popelka-Filcoff, Rachel S., Lenehan, C.E., Walshe, K., Bennett, J.W., Stopic, A., Jones, P., Pring, A., Quinton, J.S., Durham, A., 2012. Characterisation of ochre sources in south Australia by neutron activation analysis (NAA). J. Anthropol. Soc. S. Aust. 35, 81–90. R Core Team, 2019. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria URL. https://www.R-project. org/. Roebroeks, W., Sier, M.J., Nielsen, T.K., De Loecker, D., Parés, J.M., Arps, C.E.S., Mücher, H.J., 2012. Use of red ochre by early Neandertals. Proc. Natl. Acad. Sci. 109, 1889–1894. Roper, Donna C., 1991. A comparison of contexts of red ochre use in Paleoindian and upper Paleolithic sites. N. Am. Archaeol. 12, 289–301. Ruth, Susan, 2013. Women's Toolkits: Engendering Paleoindian Technological Association (PhD Dissertation). University of New Mexico, Proquest LLC, Michigan. Sando, William J., Gordon Jr., Mackenzie, Thomas Dutro Jr., J., 1975. Stratigraphy and Geologic History of the Amsden Formation (Mississippian and Pennsylvanian) of Wyoming. Geologic Survey and Professional Paper. United States Printing Office, Washington. Southerland, Wayne M., Cola, Elizabeth C., 2015. Iron Resources of Wyoming. Wyoming Appendix 1. Geochemical analysis continued Samples were analyzed by ICP-OES on a Perkin Elmer Optima 8300 for 20 elements that had quantifiable concentrations based on the pilot study. Sample introduction consisted of a peek Mira-mist nebulizer, quartz cyclonic spray chamber, and ceramic torch. All wavelengths were quantified axially. The spectrometer was allowed to warm up and the plasma to settle for a minimum of 1 h prior to analysis. Standards and quality control standards (QCs) were prepared in 2% Optima grade nitric acid and Type 1 (18.2 meg-ohm) deionized water. For each element, five to eight levels of calibration were used depending upon the expected range of sample values. A minimum of nine QCs (three each at the bottom, middle, and top of the curve) were run at the beginning, middle and end of the run, in order to quantify any lack of precision due to drift. Additional QCs were added for large calibration ranges. Samples were completely randomized within the analytical run. Postrun bracketing was permitted especially where the calibration range covered more than two orders of magnitude. Data quality objectives included: calibration curve r2 values ≥ 0.995; QCs used were appropriate for the sample range for each element; mean recovery of QC standards within 90–110% of actual values; and relative standard deviation of replicates within 10%. The exception to this was Lead which has a noisier signal than the other elements. Lead recoveries were overestimated by an average of 25% and had and uncertainty > 10% (hence the high DL). Rawlins PB is higher than other potential source areas, making it useful for the PC analysis eliminating Rawlins as a source. Method detection limits were calculated using the standard deviation of 11 replicates a non-zero standard times the T-statistic for 10 degrees of freedom and a 99% confidence interval. References Bamforth, Douglas B., 2002. High-tech foragers? Folsom and later Paleoindian technology on the Great Plains. J. World Prehist. 16, 55–98. Bamforth, Douglas B., 2009. Projectile points, people, and plains Paleoindian perambulations. J. Anthropol. Archaeol. 28, 142–157. Bement, Leland C., 1999. Bison Hunting at Cooper Site: Where Lightning Bolts Drew Thundering Herds. University of Oklahoma Press, Norman. Bernatchez, Jocelyn A., 2008. Geochemical characterization of archaeological ochre at Nelson Bay Cave (Western Cape Province), South Africa. S. Afr. Archaeol. Bull. 63, 3–11. Binford, Lewis R., 1979. Organization and formation processes: looking at curated technologies. J. Anthropol. Res. 35, 255–273. Boulanger, Matthew T., Buchanan, Briggs, O'Brien, Michael J., Redmond, Brian G., Glascock, Michael D., Eren, Metin I., 2015. Neutron activation analysis of 12,900year-old stone artifacts confirms 450–510+ km Clovis tool-stone acquisition at paleo crossing (33ME274), northeast Ohio, USA. J. Archaeol. Sci. 53, 550–558. Byers, David A., 2002. Taphonomic analysis, associational integrity, and depositional history of the Fetterman Mammoth, eastern Wyoming, USA. Geoarchaeology 17, 417–440. Cavallo, Giovani, Fontana, Federica, Gonzato, Federica, Guerreschi, Antonio, Riccardi, Maria Pia, Sardelli, Giorgia, Zorzin, Roberto, 2017a. Sourcing and processing of ochre during the late upper Palaeolithic at Tagliente rock-shelter (NE Italy) based on conventional X-ray powder diffraction analysis. Archaeol. Anthropol. Sci. 9, 763–775. Cavallo, Giovani, Fontana, Federica, Gonzato, Federica, Peresani, Marco, Riccardi, Maria Pia, Zorzin, Roberto, 2017b. Textural, microstructural, and compositional characteristics of Fe-based geomaterials and upper Paleolithic ocher in the Lessini Mountains, Northeast Italy: implications for provenance studies. Geoarchaeology 32, 437–455. Dana, Edward S., 1932. Textbook of Mineralogy. John Wiley and Sons, Inc, New York. Deviese, Thibaut, Stafford Jr., Thomas W., Waters, Michael R., Wathen, Crista, Comeskey, Daniel, Becerra-Valdivia, Lorena, Higham, Thomas, 2018. Increasing accuracy for the radiocarbon dating of sites occupied by the first Americans. Quat. Sci. Rev. 198, 528 Journal of Archaeological Science: Reports 25 (2019) 519–529 S.E. Zarzycka, et al. State Geological Survey Report of Investigations. 67. pp. 91. Speth, John D., Newlander, Khori, White, Andrew A., Lemke, Ashley K., Anderson, Lars E., 2013. Early Paleoindian big-game hunting in North America: provisioning or politics? Quat. Int. 285, 111–139. Stafford, M.D., 1990. The Powars II Site: A Paleoindian Red Ochre Mine in Eastern Wyoming (Unpublished M.A. Thesis). Department of Anthropology, University of Wyoming. Stafford, Michael D., Frison, George C., Stanford, Dennis, Zeimans, George, 2003. Digging for the color of life: Paleoindian red ochre Mining at the Powars II site, Platte County, Wyoming, U.S.A. Geoarchaeology 18, 71–90. Tankersley, Kenneth B., Tankersley, Kevin O., Shaffer, Nelson R., Hess, Marc D., et al., 1995. They have a rock that bleeds: sunrise red ochre and its early Paleoindian occurrence at the hell gap site, Wyoming. Plains Anthropol. 40, 185–194. 529