Stemmed Points and the Ice-Free Corridor

ABSTRACT

Fluting or basal thinning of lanceolate points can be readily recognized, so that much reasoning about the initial occupation of the Ice-Free Corridor has centered on the fluted point phenomenon. Fewer comparisons have been made with stemmed points from the Plateau and Great Basin, which have considerable morphological variability and a time range that extends from 14,000 or more years ago to as recently as ∼8000 years ago. A broad time range is also true for Alaskan stemmed points. Yet, a variety of stemmed point morphologies became quite common in Alberta, as the entire Ice-Free Corridor deglaciated and became biotically habitable no later than 13,200 years ago. Like the northern Great Plains, both regions also had access to the expanding Corridor. The relative abundance of both fluted and stemmed points in the Corridor has significant implications for our understanding of rapidly evolving genetic inferences about early PaleoIndigenous populations as deglaciation proceeded.

KEYWORDS:

• Ice-Free Corridor
• Haskett
• Parman
• Cody Complex
• Sluiceway

1. Introduction

Much of the attention regarding the initial First Nations presence in the Ice-Free Corridor (hereafter, the “Corridor”) has involved fluted points. Fluted point inventories for the region began in an era when that technology still appeared to reflect the initial Indigenous presence in the western hemisphere; basal thinning or fluting attributes are readily recognized (e.g., Gryba 1988; Ives 2006). Geography too played a role in that the deglaciating southern funnel of the Corridor was contiguous with the Great Plains world in which Clovis technology was well known. As archaeological research developed over recent decades, however, the idea of “Clovis first” ebbed, with the recognition that some stemmed points were coeval with Clovis, followed by subsequent determinations that stemmed points in some cases were pre-Clovis by as much as a millennium or more (e.g., Davis et al. 2022; Waters 2019; Waters, Stafford, and Carlson 2020).

One corollary of these changing perceptions would be that populations using stemmed point technologies had opportunities to enter the deglaciating Corridor region from both the north and the south. Recent studies have shown that Idaho, Oregon and Alaskan obsidian is present in fluted point and other early period artifacts in Alberta (Kristensen et al. 2023). There are good reasons for thinking that the Corridor region had the capacity at times to isolate populations present there, while at other times serving as a “confluence” zone for both human populations and ideas. While it cannot be as comprehensive as decades of work with fluted point distributions, it is timely to make a renewed consideration of stemmed point technology in the Corridor, revisiting the scope of Bryan’s (1980) thoughtful treatment, in which he reviewed stemmed point forms from the Great Basin, southern Plateau, Great Plains and Subarctic.

2. A temporal framework

Securing a radiometrically determined timeline for the Corridor region has at times been elusive (see Figure 1 for various site locations). Nevertheless, the last few years have seen an unprecedented level of high-calibre earth science research clarifying the process of deglaciation and the onset of biological habitability. Dalton et al. (2023) synthesized what is becoming a large literature in describing North American ice systems from the onset of the Late Wisconsinan through to the retreat of ice into the Canadian high Arctic. Table 1 flags several temporal sign posts, coordinating radiocarbon, ¹⁰Be, and OSL methods. The southern Cordilleran and Laurentide Ice Sheets (CIS and LIS) had decoupled by ∼14,900 years ago.¹ By 14,000 years ago large parts of southern and central Alberta had deglaciated. There followed a 1200 km rapid retreat of the LIS to a position on the Canadian Shield in northwestern Saskatchewan by 13,000 years ago (Froese et al. 2019; Margold et al. 2019; Norris et al. 2021; 2022; 2024; Reyes et al. 2022). By then, the “Corridor” was vast, although it would take several more millennia before the Keewatin ice dome vanished from the Barren Grounds.

Figure 1. Locations and sites mentioned in the text. Background map after Dalton et al. (2020), but with ice margin approximations following Norris et al. (2022; 2024; Dalton et al. 2023).

Table 1. Paleoenvironmental conditions.

Paleoenvironment	Location	Age (cal yr BP)	Sources
Decoupling of LIS-CIS	Central and southern Alberta, NWT	14,900	Dalton et al. (2023) and sources therein
Deglaciation of entire Corridor	Large areas of central and southern Alberta ice free	14,000	Dalton et al. (2023)
Rapid, 1200 km retreat of LIS	To Cree Lake Moraine, northwestern Saskatchewan	13,000	Dalton et al. (2023), but see especially Norris et al. (2021; 2022; 2024)
Entire length of Corridor biotically habitable	Mouth of Mackenzie to Montana	13,200	Dalton et al. (2023), Froese et al. (2019), Heintzman et al. (2016)
Final Dissipation of Keewatin Ice Dome	Barren Grounds	∼7000–8000	Dalton et al. (2023)

Numerous dates on Pleistocene megafauna from the Wally’s Beach site in the St. Mary Reservoir, as well as other locations, show that ecesis had occurred in southern Alberta by 13,500 years ago, when bison, horse, camel, helmeted musk-ox and scimitar cat were present (Devièse et al. 2018; Waters et al. 2015). Pleistocene megafauna have produced 32 additional radiocarbon dates more generally from Alberta (29 of which are AMS), ranging from 11,620 ± 150 to 11,010 ± 25 radiocarbon years ago (¹⁴C yr BP) (calibrating to ∼13,600 to 13,050 calendar years ago (cal yr BP)), involving Bison (18 instances) and Equus (eight instances), as well as caribou (Rangifer tarandus), camel (Camelops) and lion (Panthera leo atrox) (one instance each) (Bellissimo 2013; Heintzman et al. 2016; Ives et al. 2013; Jass and Allan 2016; Jass, Burns, and Milot 2011; Schwartz-Narbonne et al. 2019; Supplementary Material [SM] Table S1). One of the Edmonton area bison in this time range came from an eastern Beringian clade, and it, along with horse, bison and Populus dates from northwestern Alberta and northeastern British Columbia indicate that the entire Corridor region had large mammal communities by 13,200 years ago (Heintzman et al. 2016; see Ives 2024b). Generally speaking, early deglacial environments were cold and relatively open, eventually giving way to white spruce gallery forests, and increasingly to the south, grasslands (Beaudoin and Oetelaar 2003; Dyke 2005). Extensive faunal exchange between eastern Beringia and unglaciated North America through western Canada ended as initial versions of a more closed boreal forest appeared in the early Holocene (MacDonald and McLeod 1996; Pedersen et al. 2016).

From a human perspective, the Wally’s Beach horses and camel had been butchered, with the very best available AMS dating methods providing an age of 13,450 years ago (Table 2) (Devièse et al. 2018). Those same dating methods have recently been applied in connection with the Anzick Clovis child showing that his burial with accompanying wapiti antler foreshafts and Clovis artifacts took place ∼12,900 years ago (Becerra-Valdivia et al. 2018). Both the Vermilion Lakes site in the Banff area of Alberta and Tse’K’wa (Charlie Lake Cave) in northeastern British Columbia have early occupations in the 12,600 year range, with the lowest Tse’K’wa component producing the only dated instance of a small basally thinned point in western Canada (Driver et al. 1996; Fedje et al. 1995; Heintzman et al. 2016). The component 2 occupation of the Vermilion Lakes site took place ∼11,500 years ago, with a leaf-shaped point and microblade technology being present there; microblade technology also appears at Tse’K’wa in a similar time range (Driver 1999; Fedje et al. 1995). Shouldered, stemmed points of the Eclipse Phase are dated in the Banff area to ∼11,800 years (Fedje et al. 1995; 1996; Landals 2013, 102–106, Plate 14). It is likely that the Lindoe stemmed point that Bryan (1980, 98) recovered in southern Alberta eroding from a bison bone bed is of similar age.² Stemmed point instances found in a wider array of private as well as local, regional, and Royal Alberta Museum collections can be evaluated against this temporal backdrop.

Table 2. Dated archaeological assemblages.

Archaeological finding	Location	Age (cal yr BP)	Sources
Swan Point	Tanana River, Alaska	14,200	Gómez Coutouly and Holmes (2018)
Cooper’s Ferry, Paisley Cave	Washington, Idaho	14,000–16,000	Davis et al. 2022
Pre-Clovis Butchered camel and horses	Wally’s Beach, St. Mary Reservoir	13,450	Devièse et al. (2018)
Anzick child burial	Montana, southern “funnel” of Corridor	12,905–12,695	Becerra-Valdivia et al. (2018)
Basally thinned, fluted points	Earliest Tse’K’wa deposits; Raven Bluff, Serpentine Hot Springs, Alaska	∼12,400–12,700	Buvit et al. (2019), Driver et al. (1996), Heintzman et al. (2016), Smith and Goebel (2018)
Upward Sun River perinate interments	Alaska	∼11,500	Tackney et al. (2015)
Tse’K’wa ravens and microblade technology	Northeastern British Columbia	∼11,400	Driver (1999), Driver et al. (1996), Heintzman et al. (2016)
Vermilion Lakes Component 2 leaf-shaped point and microblade technology	Southwestern Alberta	11,500	Fedje et al. (1995)
Eclipse Phase stemmed points	Southwestern Alberta	∼11,500	Fedje (1988; 1996)
Saskatoon Mountain hearth	Northwestern Alberta	∼10,500	Beaudoin et al. (1996)
Norquay and Fletcher Site Alberta points	Southern Alberta	∼10,500	Fedje (1988), Vickers and Beaudoin (1989)
Agate Basin-like and Kamut points	Barren Grounds	∼8000	Gordon (1996), Noble (1981), Wright (1972; 1981)

3. Selected stemmed points from the Ice-Free Corridor

Apart from some of the stratified and dated sites noted above, I will rely on examples from avocational collections as well as local, regional, and Royal Alberta Museum collections. Some prefatory remarks are in order. While the loss of provenance information that is the case for surface collections is limiting, extensive agriculture and avocational collecting result in a “brute force” kind of sampling that reveals the larger nature of an archaeological record. For issues such as exotic raw material use or the frequency of certain morphologies, the value of censusing collections involving many thousands of artifacts cannot readily be replicated by other archaeological research. It is worth noting that many archaeologists working in Alberta, myself included, received basic training in a northern Great Plains context. Not surprisingly then, it is common to turn to familiar, northern Plains taxonomies in evaluating projectile points encountered in a variety of circumstances. Consequently, various stemmed points from Corridor contexts are often “slotted” into existing Great Plains categories, especially Agate Basin and Hell Gap, with varying degrees of success. In other instances, stemmed points dangle as taxonomically “anomalous.”

Deepening the complexity of this endeavor, there is a genuine focus on functionality in the Western Stemmed Tradition (WST), creating a substantial range of morphological variability – a range that is further exacerbated by extensive maintenance and refurbishing of these artifacts (e.g., Knell, Hill, and Sutton 2021; Rosencrance 2019). Smith et al. (2019a) indicated that there was not then an objective means of classifying WST points and that the chronological and technological relationships among these forms remained somewhat ambiguous (see also Amick 2013). Rosencrance et al. (2024) more recently used a well-controlled sample of radiocarbon dates to suggest a temporal ordering of WST morphologies using Bolling-Allerød, Younger Dryas, and Early Holocene time frames. Cooper’s Ferry, Paisley Caves, and then Lind Coulee forms as well as Haskett precursors (possibly as old as 13,000 years ago) are the earliest, followed by securely dated Younger Dryas Haskett forms. Cougar Mountain, Parman, Windust, Windust/square, and Lake Mohave types follow in the first half of the Early Holocene continuing until about 9500 cal BP. Silver Lake, Stubby, and Bonneville types proliferate later in the Early Holocene and continue after ∼9500 cal BP.

With those caveats, Figure 2 shows a small sample of Corridor points³ illustrating that many are imperfect fits for Great Plains categories but that otherwise fall comfortably within the range of variability typical of WST forms. The most surprising of these would be Figure 2(a) that, with its slightly barbed shoulders and rounded tang for a base, is indistinguishable from several of the Cooper’s Ferry points recently dated to the 15,000–16,000 year time range (Davis et al. 2022, figure 4; see also the Lind Coulee attribution in Smith et al. 2019a, figure 2d). The paleoenvironmental data reviewed above show that such an early date would not be possible: this specimen comes from the Grande Prairie area of northwestern Alberta, where biotic habitability only existed by 13,200 years ago, or perhaps slightly earlier. It does leave open the possibility, however, that there was an Indigenous presence at an early date at this point of constriction in the Corridor. This point is also similar to oversize Parman examples Smith et al. (2024, 205, figure 6) illustrated, so that an early Holocene age is also possible. There is no local analog with which this point might be confused.

Figure 2. (a) and (b), stemmed points from the Grande Prairie Museum collection, northwestern Alberta; (c) shouldered stemmed point from the Iverson collections, west-central Saskatchewan; (d) and (e), large shoulderless stemmed points from the Craig collection, Grande Prairie area, northwestern Alberta; (f) large stemmed bifacial preform or knife from the Bull collection, central Alberta; (g) stemmed point from the Grande Prairie Museum collection; (h) stemmed point from the Glacier Pass quarry area, Willmore Wilderness, northwestern Alberta; (i) to (k), stemmed points from the Grande Prairie Museum collection; (l) stemmed point from the Tse’k’wa area, northeastern British Columbia, Simon Fraser University collections; (m) to (o), Eclipse Phase stemmed points from the Banff area, southwestern Alberta, Parks Canada collections; (p) stemmed point from Mount Norquay, Kananaskis Country, southwestern Alberta. Photographs of (a), (b), (g), (i), (j), and (k) courtesy Todd Kristensen (Archaeological Survey of Alberta) and the Grande Prairie Museum; photograph of (c), courtesy Dale Fisher; photographs of (d) and (e) courtesy Darryl Bereziuk (Archaeological Survey of Alberta); photograph of (l) courtesy of Todd Kristensen and Simon Fraser University; (m), (n), (o) and (p) photographed courtesy of Gwynn Langeman and Parks Canada.

Figure 2(b) resembles Channel Island and other early barbed points from Pacific Coast settings (e.g., Smith, Kuzminsky, and Linderholm 2020). Some caution is warranted. While this might be an appropriate comparison, later Holocene points with similar characteristics are known from southern Vancouver Island, the Gulf Islands, the Puget Sound and the lower Fraser River regions (see various papers in Carlson and Magne (2008)). A number of British Columbia-sourced nephrite celts are present in Alberta, and it is possible that such a point might have arrived through much later exchange processes contemporary with those celts (Kristensen et al. 2016).

Figure 2(c) from west central Saskatchewan would typically be regarded as a Hell Gap specimen, but can readily compared with various Parman or Cougar Mountain examples (e.g., Bryan 1980, 85, figures 7 and 9; Layton 1979, 52–53, figures 4 and 5; Rosencrance 2019). Figure 2(d) is an unusual point for Alberta and could be accommodated in the Haskett type 1 category (Galm and Gough 2008; Rosencrance 2019; Rosencrance et al. 2019). That would also be true of Figure 2(e), although that artifact also conforms well to classic Agate Basin points (Frison and Stanford 1982). Both come from the Grande Prairie area of northwestern Alberta. Figure 2(g) would fall within Lind Coulee or Windust ranges of variability (e.g., Pratt et al. 2019, figure 2g; Rosencrance 2019). It is, however, indistinguishable from a Silver Lake point Knell, Hill, and Sutton (2021, figure 3c) illustrated from China Lake. Figure 2(f) could easily be a preform for a point or knife that would resemble Figure 2(g). The contracting stem of Figure 2(h) made of Glacier Pass concretion in the Canadian Rockies, certainly resembles illustrations of Cougar Mountain points (Kristensen et al. 2019; for comparisons, see Bryan 1980, 85, figure 8; Layton 1979, figure 4; Rosencrance et al. 2019; Smith et al. 2019a, figure 2b). The foreshortened stem of Figure 2(i) could be accommodated among Bonneville points, while the square, elongated stem of Figure 2(j) has no parallels in Alberta, but harkens to the elongated stems Bryan (1980, 87–88) mentioned for similar points from the Dietz site (e.g., Willig 1988, figures 19a,d and 37a). Figure 2(k,l) could be construed as Parman or Cougar Mountain in form but are equally like On Your Knees stemmed points from southeast Alaska illustrated by (Fedje et al. 2021; Dixon 2008, figure 4f, g; Pratt et al. 2019, figure 4e).

For Figure 2(m–o) we do have a greater degree of temporal control. These are Eclipse Phase stemmed points from the Banff area of southwestern Alberta, with dates ranging from ∼11,800 to 11,200 years of age (Fedje 1988; 1996; Fedje et al. 1995; Landals 2013). Fedje (1996) referenced Hell Gap comparators in these instances but was equally aware that there were strong resemblances to WST points for the Eclipse Phase stemmed points. The broad-stemmed Norquay point in Figure 2(p) resembles Alberta points and is ∼10,500 years of age, like the Cody Complex occupation of the Fletcher site (Fedje 1988; Vickers and Beaudoin 1989). While this is a necessarily cursory review, it is sufficient to show that there are just too many rather strong resemblances among the Corridor examples to simply be attributed to coincidence or reference to “generalized technological forms.”

4. Stemmed points in specific site contexts

Another way to proceed with an initial foray into stemmed point variability in the Corridor would be to seek sites with rich early period archaeological records. This would allow us to consider how common stemmed point morphologies might be in the Corridor region, relative to other early period forms. The Wally’s Beach site (DhPg-8) in the Saint Mary Reservoir may be best known for its evidence of early camel and horse butchering, but it is a rich locale for all time periods, with more than one thousand projectile points surface collected from the reservoir perimeter. Figure 3 illustrates the majority of the stemmed points in Royal Alberta Museum collections from this site. It is arranged roughly by morphology from smaller, primarily square-stemmed points in the top row, to more lightly shouldered specimens in the second and third rows, to longer, constricting stem points in the fourth row, and shoulderless stemmed points in the bottom row.

Figure 3. Stemmed points from the Wally’s Beach collection, southern Alberta, Royal Alberta Museum collection: Top row, left to right: H99.22-4276, H99.22-885, H99.22-4283, H99.22-5650, H99.22-881, H99.22-8501; Second row, left to right: H99.22.4506, H99.22.3676, H99.22.609, H99.22.4260, H99.22.4262, H99.22-5833; Third row, left to right: H99.22.730, H99.22.697, H99.22.5687, H99.22.123, H99.22.705; Fourth row, left to right: H99.22.4289, H99.22.4642, H99.22.965, DhPg-8.2856.1, H99.22.1005; Fifth row, left to right: H99.22.4261, and uncatalogued specimen illustrated in Yanicki et al. (2022), H99.22.4263, DhPg-8.75, DhPg-82. 2864.1. All photographed courtesy of the Royal Alberta Museum, save for the uncatalogued specimen, with those photographs provided by Gabriel Yanicki, Canadian Museum of History.

Rather than making many individual comparisons, a few generalizations can suffice. First, it is in no way difficult to find examples that, were they to be found in Great Basin or southern Plateau contexts, would almost certainly be termed Silver Lake, Lake Mohave, Parman, Cougar Mountain, Windust and Windust/square, Lind Coulee, or Bonneville in form (e.g., Bryan 1980; Rosencrance 2019, table 2.11; Rosencrance et al. 2024). Second, in the majority of cases, it would also be true to say that there is little prospect for these points being somewhat atypical examples of what would otherwise be more commonly recognized regional forms for the northern Great Plains or Subarctic. The bottom-most row would be the chief exception to this statement, where the points can more readily be regarded as Hell Gap or Agate Basin instances. That is particularly true of the elegant green argillite point in the lower left-hand corner, or the stem fragment to its immediate right (with its overshot flaking). Finally, if we compare this roster of stemmed points with the frequency of Clovis, Folsom, Goshen, Plainview, Agate Basin, Hell Gap and Cody Complex points Tolman (2001) described in his reporting of the Wally’s Beach projectile point assemblage, it is accurate to say that various stemmed points are about as common as early types recognized as northern Plains-related.

Corridor stemmed points can also be evaluated by considering their frequency at excavated sites with high artifact densities and long-term occupations, but poorly developed stratigraphy, such as Ahai Mneh (FiPp-33) west of Edmonton. While there were no opportunities for radiometric dating of early deposits, excavations and surface collecting at this site produced a poorly fashioned fluted point, an Agate Basin or Hell Gap-like stemmed point (Figure 4, lower right; see also Ives 2024b, figure 10), an Alberta point fragment, and a Scottsbluff point, along with a number of stemmed points. Despite the thin stratigraphic development at Ahai Mneh (and similar sites), there is a general relationship between the depth of artifact recovery and age of the artifact (see Ives 2023). Figure 4 shows the high density of artifacts from all time periods at Ahai Mneh, but also the propensity for stemmed points to occur at greater depth than middle and late period artifacts. These shoulderless stemmed point morphologies extend from the Agate Basin-Hell Gap spectrum to small, Early Holocene lanceolates known to precede the advent of notched points. Basal stemmed point fragments like the pyrometamorphic example in the upper left-hand corner of Figure 4 are quite common. Thus, at an excavated site with an early period component, stemmed points are actually the dominant early period form.

Figure 4. Three-dimensional plotting of the artifact distribution in Area B of the Ahai Mneh site (FiPp-33), south of Lake Wabamun and west of Edmonton, provided by Jen Hallson. Despite the thin stratigraphy, the stemmed points at this site are more deeply buried than succeeding middle and late period artifacts, where subsequent Oxbow, Besant, Avonlea, and Old Women’s Phase artifacts occur nearer the surface (see Ives 2023).

Both assemblages indicate that it would be timely to take seriously the role of stemmed point technology in the deglaciating Corridor.

5. A Knife River Flint subset of Wally’s Beach stemmed points

Prior to making additional specific comparisons, there is a subset of Wally’s Beach stemmed points and point fragments not included in the previous section (see Figure 5). Two of these points (Figure 5(a,b)) are associated with a cluster of Knife River Flint artifacts (including seven end scrapers and six flakes or fragments). From a tightly circumscribed collecting area, they represent a lost toolkit, a spatially restricted activity locus, or a cache. Note that the majority of these Wally’s Beach artifacts are heavily patinated, a reasonable indication of relative antiquity. These broad-stemmed points would normally be regarded as Alberta instances; note the strong similarity between them and the Mount Norquay point in Figure 2(n). The dating for the Norquay point and the Fletcher site Alberta points suggest that the age of these Wally’s Beach specimens would be in the range 10,000–10,600 years (Vickers and Beaudoin 1989).

Figure 5. Knife River Flint Alberta points from the Wally’s Beach site, southern Alberta: (a) H99.22.4897, (b) H99.22.5704, (c) H99.22.4502, (d) H99.22.880, (e) H99.22.4467, (f) H99.22.5247, (g) H99.22.4359. Photographed courtesy of the Royal Alberta Museum.

There must have been a remarkably strong axis of social interaction between Cody Complex groups in Alberta and the Knife River Flint quarries. Dawe (2013) found that, for a sample of 475 artifacts (from over one thousand Cody Complex diagnostics), where the toolstone could be accurately determined, 43.8 per cent of Cody knives, 42.7 per cent of Alberta Points, 25.6 per cent of Scottsbluff points and 21.9 per cent of Eden points were made of Knife River Flint. Cody Complex diagnostics fashioned from Knife River Flint are found into northwestern Alberta and northeastern British Columbia (Kristensen et al. 2018). To this point, the majority of the comparisons made here have been to the southern Plateau and Great Basin; in the Cody Complex case, there were clearly extensive exchange processes commonly moving Knife River Flint diagnostics from North Dakota 800–1000 and even 1600 km into the greater Corridor region (Ives 2015; 2024a; Kristensen et al. 2018).

6. Large lanceolate points in the Corridor

There are several striking comparisons with WST points. The point in Figure 6(a), recovered from the Brazeau Reservoir in west central Alberta – apart from its somewhat larger size – is indistinguishable from the cast of an Old River Bed Delta Haskett point in Figure 6(e) (see also Butler 1965, 20, figure 10f; Duke 2015).⁴ Rouseau (2008, 227, figure 4i) illustrated another point from southern British Columbia that is virtually identical to the examples in Figure 6(a,e), drawing a comparison with Hell Gap forms. The lanceolate points in Figure 6(b,c), from the Edmonton area and the Yukon respectively, also share Haskett resemblances when compared with the Old River Bed Delta and Wishbone points in Figure 6(e,f). In western Canadian contexts, the points in Figure 6(a–c) would commonly be called Agate Basin or Hell Gap forms. Such precise similarities do raise the issue of just how different Great Plains Agate Basin and Hell Gap forms really are from both these western Canadian and Great Basin instances. While the above comparisons look southward, there are also grounds for comparing these Corridor lanceolates (Figure 6(a–c), and for that matter, Figure 2(d,e)) with similar forms from Alaskan sites like Putu, Bedwell and various Sluiceway localities.

Figure 6. (a) FfPv-7:9, a stemmed point from the Brazeau Reservoir, west-central Alberta; (b) H10.25:1068, a stemmed point from the Edmonton area Diederichs collection; (c) NaVl-9:1, a stemmed point from the Old Crow region, Yukon; (d) cast of a Haskett point, Old River Bed Delta, Utah, Natural History Museum of Utah; (e) a Haskett point, 42TO6384 220, Wishbone site, Utah; (f) leaf-shaped point from Component 2, Vermilion Lakes site, Banff, Alberta; (g) microblade core from Component 2, Vermilion Lakes site; (h) to (l) leaf-shaped points or bifaces from the Grande Prairie Museum collection; (m) lanceolate point with well-organized parallel, oblique flaking pattern. (a) and (b) photographed courtesy of the Royal Alberta Museum; photograph (c) courtesy of the Canadian Museum of History; (d) cast photographed courtesy of the Natural History Museum of Utah; photograph of (e) provided by Daron Duke; (f) and (g) photographed courtesy of Gwynn Langeman and Parks Canada; (h) to (m) photographs courtesy of Todd Kristensen and the Grande Prairie Museum.

7. Leaf-shaped points in the Corridor

Component 2 at the Vermilion Lakes site in the Banff area of southwestern Alberta produced a leaf-shaped point and an exhausted microblade core dating to ∼11,500 years ago (Fedje et al. 1995; 1996; Figure 6(f,g)). Far to the north, Potter et al. (2014; Moreno-Mayar et al. 2018a; Halffman et al. 2020; Tackney et al. 2015) described perinate burials from the ∼11,500-year-old Upward Sun River child burials of central Alaska. These children provided the first indication that the founding population for the Americas split at an early date into an Ancient Beringian lineage (reflected in these Upward Sun River children) and a founding lineage for the remainder of the Americas. About a millennium later, the Ancient Beringian lineage apparently disappeared. Interred grave goods included antler foreshafts associated with a leaf-shaped point and bipointed biface. The leaf-shaped point bears a striking resemblance to the Vermilion Lakes instance.

These could simply be generalized technological forms appearing at any number of junctures in archaeological records. They are contemporaneous, however, and the presence of campus-style microblade core technology indicates some manner of northward connection for the Vermilion Lakes instance. This raises the issue of whether or not there might be a temporal horizon in which these technologies occurred over much of interior western North America. A subsidiary question would then be “Are there intervening instances of similar point and microblade technology?”

In fact, there are, as can be seen in Figure 6(h–l), with leaf-shaped points and bipointed bifaces from northwestern Alberta. Eldridge et al. (2010; 2012) reported several more examples of leaf-shaped points and bipointed bifaces in the rich, early period record from the Williston Lake Reservoir in northeastern British Columbia, comparing the points directly with the Vermilion Lakes example. The Williston assemblages also had examples of lanceolate and leaf-shaped points with well-developed oblique parallel flaking, as can also be seen in the northwestern Alberta example in Figure 6(m): such flaking is reminiscent of later PaleoIndigenous northern Great Plains forms. There is a broadly contemporaneous microblade core instance at Tse’K’wa in northeastern British Columbia, several microblade cores in the Williston assemblages (some of Mount Edziza obsidian and one that appears to be Knife River Flint), and a number of microblade cores in northwestern Alberta (Eldridge et al. 2010; 2012; Magne 2023). Magne (2023) concluded that western Canadian microblade technology arose from Denali Complex roots, entering Alberta and northeastern British Columbia as early as 11,500 years ago at Vermilion Lakes and Tse’K’wa, and present in southern Alberta during Cody Complex times ∼9000 years ago (Magne, Hughes, and Kristensen 2019). These factors suggest that the Vermilion Lakes and Upward Sun River artifacts of known age are not temporal and morphological coincidences, but rather, reflect a detectable degree of technological continuity along the greater Corridor region, as the Younger Dryas ended.

8. Mesa-like points from Northern Alberta

Kunz (2011) and Admiraal (2013) saw an underlying relationship among Alaskan lanceolate projectiles and those from Mesa and related sites. Figure 7(a–l) illustrates a number of square to concave-based points from northern Alberta and northeastern British Columbia. Figure 7(m–s) shows the range of variability in Mesa points from northern Alaska (Kunz, Bever, and Adkins 2003). The Corridor points might be termed Agate Basin points, but they diverge from the type site Agate Basin forms in their emphasis on a thick cross-section. That, plus basal forms suggestive of damage and repair, and traits including robust, parallel, comedial flaking, do warrant consideration for their resemblance to Mesa forms – bearing in mind also the perception that Mesa assemblages have been thought to represent northward incursions of later PaleoIndigenous technologies (e.g., Dixon 2015; Kunz, Bever, and Adkins 2003). The square-based point in Figure 7(k) comes from the Quarry of the Ancestors site in northeastern Alberta and also resembles several of the Mesa site points (Bever 2006; Clarke, Ronaghan, and Bouchet 2017, 143; Kunz, Bever, and Adkins 2003, figures 19 and 20; Reeves, Blakey, and Lobb 2017). The oblique basal thinning flakes on this point occasionally occur on Alberta points and are relatively common on the northern Plano points described below. While further testing would definitely be merited, it is worth noting that this point produced a positive result for proboscidean antisera (see further discussion below for northern Plano; Reeves, Blakey, and Lobb 2017, 128–129).

Figure 7. Square and concave-based stemmed points: (a) to (f), (h) to (j), and (l) points from the Grande Prairie Museum collection; (g) a point from the Tse’k’wa area of northeastern British Columbia; (k) a point from the Fort MacKay area, lower Athabasca River, northeastern Alberta; (m) to (s) Mesa points: (m) kir-102 81-56-1456, (n) kir-102 98-759, (o) kir-102 81-56-973, (q) kir-102 81-56-24, and (r) kir-102 79-160; (p) and (s): casts of Mesa points: (p) kir-102-92-01 and (s) kir-102 85-1. (a) to (j) and (l), photographs courtesy of Todd Kristensen and the Grande Prairie Museum; (k) photographed by J. Ives; (p) photographed courtesy of Lifeways of Canada; photographs (m) to (o), (q) and (r) courtesy of Mike Kunz.

9. Oblanceolate points or knives in the Corridor

There are other indications of north–south connections through the Corridor. Figure 8(a) illustrates a thick, vitreous gray quartzite point from the Poohkay collection (GlQl-3), a lost tool kit or cache with shoulderless stemmed points from northwestern Alberta. This point and the Poohkay assemblage are described in greater detail elsewhere in Ives (2024b); the tip of this point has been damaged and partly repaired. It is distinctively different from the other shoulderless lanceolate points in the GlQl-3 assemblage (see accompanying article), but in terms of morphology and metrics, it is indistinguishable from Sluiceway points from the Brooks Range in Alaska (Rasic 2000; 2011). Figure 8(b–e) illustrate other points resembling Sluiceway materials from northwestern Alberta. These can be compared with typical Sluiceway materials from the Caribou Crossing and Irwin Sluiceway sites in Figure 8(f–k).

Figure 8. Top row, (a) through (e), oblanceolate or spatulate points from northern Alberta: GlQl-3:5, Poohkay collection, Grande Prairie area, northwest Alberta; (b) Hofer collection, Grande Prairie area; (c) and (d) Craig collection, Grande Prairie area; (e) Grande Prairie Museum. Bottom row, Sluiceway points from the Brooks Range, Alaska: (f), NOAT 18710 Caribou Crossing site in the western Brooks Range, Noatak National Preserve; (g) NOAT 18735, Caribou Crossing; (h) through (j) Irwin Sluiceway (XHP-00496) site in the western Brooks Range, Noatak National Preserve (NOAT 2502, 2503, 3055, and 2499 respectively). GlQl-3:5 photograph courtesy of the Royal Alberta Museum; Grande Prairie area points (b) through (e) courtesy of Todd Kristensen, Archaeological Survey of Alberta. Sluiceway NPS photographs courtesy Jeff Rasic.

Figure 9(a) illustrates two of the spatulate or oblanceolate points from the 10,500 to 11,000 year-old Component 2 at the Dry Creek site in Alaska’s Nenana valley, while Figure 9(b) illustrates stacked tablets for two microblade cores from that component (Graf et al. 2015; Powers and Hoffecker 1989). Figure 9(c) is a spatulate point from Batza Tena site RkIg-52 in Alaska’s Koyukuk River region; Figure 9(d) is an oblanceolate point from Batza Tena site RkIg-28. Clark (1974) initially noted that these northern instances were virtually identical to a number of Lake Mohave points or knives typical of the Great Basin and California (e.g., Bryan 1980, 83; Knell, Hill, and Sutton 2021; Rosencrance 2019). Clark and MacFadyen Clark (1993) later made comparisons with Mesa site points in Alaska (Kunz, Bever, and Adkins 2003).

Figure 9. (a) Oblanceolate or spatulate points from Component 2, Dry Creek, Alaska; (b) microblade cores with refitted tablets from Component 2, Dry Creek, Alaska; (c) RkIg-52 174, Koyukok River, Alaska; (d) RkIg-28 81, Koyukok River, Alaska; (e) NOAT 2499, Irwin Sluiceway (XHP-00496), western Brooks Range, Noatak National Preserve, showing how closely resharpened Sluiceway points resemble these other thick, oblanceolates; (f) HkPa-4 I 1118 (Eaglenest Portage), (g) HjPd-1 667 and (h) HjPd-1 43 (Gardiner Lake Narrows) are oblanceolates from the mountains of northeastern Alberta; (i) and (j) Birch Lake, Innisfree area, central Alberta; three oblanceolate points from the Oyen area of southeastern Alberta, as illustrated by Wormington and Forbis (1965, 72, figure 16); McConachie collection, Tisdale area, northeastern Saskatchewan (note basal facet). Dry Creek artifacts photographed courtesy Roger Powers, University of Alaska, Fairbanks; Koyukok River artifact photographs courtesy of the Canadian Museum of History; NOAT 2499 courtesy of Jeff Rasic, Bureau of Land Management, Alaska; McConachie collection artifact photographed in the Phenix collection, courtesy of David Meyer, University of Saskatchewan.

Note how similar the Alaskan specimens are to three oblanceolate points from the Birch Mountains of northeastern Alberta (Figure 9(f–h)). Figure 9(i,j) are similar points from the Innisfree area of central Alberta (Ives 1993; 2017). The three points in Figure 9(k) are of historical interest. In pioneering 1930s avocational archaeological work in southeastern Alberta, Johnston recognized that these artifacts were distinctively different from the other points in the Little Gem Complex, a Cody Complex expression in this region (Wormington and Forbis 1965). A Campus-style microblade core is associated with this collection, recollecting that microblade technology also appears to be associated with Cody materials in southern Alberta’s High River locality, as well as with Cody materials at Montana’s Mammoth Meadows locality (Lee et al. 2016; Magne, Hughes, and Kristensen 2019). Figure 9(l) comes from the Tisdale area of northeastern Saskatchewan; it possesses an unaltered, oval basal facet without further trimming, true for a notable percentage of this kind of artifact (e.g., Figure 9(c [right],f,g,k [left],l)), presumably because no further flaking was required in socketing a robust point base, or affixing the artifact to a “clothespin” haft as appears to be the case for the Sentinel Gap Haskett points (Galm and Gough 2008). All of these points show signs of distal resharpening, a factor that may well obscure their original form.

Precisely what these wide-ranging similarities indicate in terms of directionality is not immediately clear. Minimally, it would appear that there is a distinctive early period tool type distributed widely in interior western North America. I am inclined to favor a northern connection: many specimens attributed to the Lake Mohave form retain a much broader blade, only occasionally resharpened fully into the stem and then resembling the artifacts under discussion. The northern forms begin as lanceolate to oblanceolate in shape, from which re-tipping proceeded. In the absence of better dating, these Corridor artifacts are broadly contemporaneous with northern and southern instances. This chain of occurrences, stretching from eastern Beringia to points south of the continental ice masses, again suggests the former Corridor region remained an avenue of communication well after deglaciation, but likely prior to the establishment of a closed Boreal Forest biome.

10. “Northern Plano” and the waning glacial world

The process of deglaciation continued later into Holocene time in Arctic Canada, with the Keewatin ice dome finally dissipating after 8000 years ago (Dalton et al. 2023). Steppe-tundra conditions likely receded on a temporal gradient toward the northeast. Harp (1958) and Forbis (1961) first reported what came to be called Northern Plano points and artifacts from the Dismal, Kamut, and Acasta Lake regions. Subsequent work by Noble (1971; 1981), Wright (1972; 1975; 1981), and Gordon (1996) continued to reveal Agate Basin-like points or knives, frequently burinated, although generally in much more recent contexts than the Wyoming Agate Basin type site.

Dating of these sites has proved difficult, with a number of assays where the association between artifacts and the material being dated was uncertain. It would appear that rougher forms approximating Agate Basin points began appearing 8000–9000 years ago, persisting later into the Holocene. Noble’s Acasta Lake assemblages revealed a trend in which a generalized Agate Basin-like lanceolate form gave way to “waisted” points with broad basal notching that created a tanged stem. These have been termed Kamut or Acasta Lake points. Bryan (1980) attributed this form of notching to stimulus diffusion as the newer notion of notching was incorporated into an existing northern technological tradition, a reasonable supposition. The situation of a number of Northern Plano sites on esker ridges and at water crossings certainly is suggestive of caribou hunting with various authors favoring caribou intercept strategies, as caribou preferentially course linear features while moving across the landscape (authors as above, and MacKay and Andrews 2016). This hunting would be supplemented by fishing.

Five of the artifacts originally described for Acasta Lake by Forbis (1961) are illustrated in Figure 10(a–e), with Figure 10(c) a less well-formed biface, and Figure 10(d) the base of a fourth point or knife. Forbis described Figure 10(a) (porphyritic basalt) and Figure 10(c) (quartz) as Lerma points, and Figure 10(b) as Agate Basin-like. Note the light shouldering of Figure 10(b) as observed by Forbis on this vein quartz specimen: further elaboration of that could readily lead to the notched Kamut form noted above. Figure 10(e) is a quartzite knife Forbis felt resembled large specimens Borden had recovered from the Pasika complex on the Fraser River of British Columbia.

Figure 10. Top row, Acasta Lake artifacts from the University of Alberta collections: (a) Acasta Lake 1, 972.20.1, (b) Acasta Lake 2, 972.20.2, (c) Acasta Lake 3, 979.20.3, (d) Acasta Lake 5, (e) Acasta Lake 8, 972.20.7. Second row, artifacts from the Fort Chipewyan Airport site: (f), (g) and (h), note basal facet on (f) and burin blow at transverse fracture on (g); (i) dorsal and ventral surfaces of two of several bifacial thinning flakes, Fort Chipewyan Airport site. Third row, (j) through (m), bifaces or bifacial performs, Fort Chipewyan Airport site. Fourth row, broadly notched, Kamut-like points: (n) Guttormson collection, Fort Mackay region, northeastern Alberta; (o) FfPv-:1508, Brazeau Reservoir, west central Alberta; (p) through (t), Fisher collection, west central Alberta; (q) Turner collection, northwestern Alberta; (u) and (v) west central Saskatchewan. Acasta Lake artifacts photographed courtesy of the University of Alberta; Fort Chipewyan Airport artifacts photographed courtesy of Darryl Bereziuk, Archaeological Survey of Alberta; FfPv-:1508 photograph courtesy Madeline Coleman and Royal Alberta Museum; Fisher and Turner collection artifact photographs courtesy of Todd Kristensen, Archaeological Survey of Alberta; west central Saskatchewan artifact photographs courtesy of Muriel Carlson, Saskatchewan Archaeological Society.

The two middle rows in Figure 10 provide a sampling of artifacts from a Lake Athabasca site, near the Fort Chipewyan airport (IeOs-5). This site (or site complex, as it is not far from IeOs-6 and 7), has workshop areas exposed in blown-out sandy deposits (Kjorlein 2007). The three points in Figure 10(f–h) fit well within the Northern Plano conceptualization. Note the transverse tip fracture and burination on Figure 10(g): burins on points occur in the Agate Basin type site and are common in Northern Plano assemblages (e.g., Frison and Stanford 1982; Gordon 1996). Figure 10(i) illustrates dorsal and ventral surface examples of relatively large, facetted, bifacial thinning flakes recovered in this workshop area, while Figure 10(j–m) are examples of various bifaces likely to have been intended as point or knife preforms.

The IeOs-5 assemblage has not been dated, but its geographic situation is suggestive, as it occurs along the edge of the Canadian Shield on the north side of Lake Athabasca. While the LIS lingered at the Cree Lake Moraine ∼13,000 years ago, it is unlikely that the IeOs-5 assemblage could date to that time frame, as the massive outburst flood from the northwestern arm of Glacial Lake Agassiz would take place sometime between ∼12,900 and 12,600 years ago, providing a terminus post quem. It would be plausible for this location to be a jumping-off point for subsequent Northern Plano expansion into the deglaciating Barren Grounds by ∼10,000 years ago. Just when and where a transition to the broad notching that would create the tanged effect for Kamut style points took place is not known. Working with the Johnston collection in the Oyen area of southeastern Alberta, however, Brink (2011) was able to secure a radiocarbon date of 9380 ± 50 ¹⁴C yr BP (Beta-254976, calibrating to 10,500–10,720 cal yr BP) for bone directly associated with a tanged point strongly resembling the one in Figure 8(n), from the Fort MacKay region of northeastern Alberta. Quite a number of points like this are present in an arc extending from northwestern and central Alberta into western Saskatchewan (see Figure 10(n–v)). It is conceivable that the idea for broad notching or corner removal of large lanceolate points emerged in the greatly expanded Corridor region ∼10,500 years ago. An obsidian lanceolate point from Alberta’s Peace River Country, resembling Agate Basin forms and sourced to Batza Tena in Alaska, indicates that there were wide-ranging contacts far to the North (Kristensen et al. 2023, 15, figure 12).

While caribou would certainly have been a prime resource, recent successes in sedimentary DNA recovery make for another intriguing prospect. While members of the northern Pleistocene megafauna, mammoth, bison and horses in particular, began to diminish during the LGM, sedDNA results raise the possibility that small, relict megafauna populations may have persisted into Holocene times in high latitude North America (e.g., Monteath et al. 2023; Murchie et al. 2021; 2023). Earlier literature cast the lateness of the Northern Plano expression in rather unflattering “slow” development terms (Collins 1963). Another way to see this phenomenon would be that periglacial conditions adjacent to the retreating Keewatin Ice Dome may have harbored the last vestiges of the Pleistocene world – perhaps an attractive thing for some intrepid early populations.

11. Discussion

A comprehensive, new review of early settlement of the Ice-Free Corridor, while merited, lies beyond the scope of the present article. Yet, a preliminary assessment of stemmed point data from the Corridor region indicates that this early period is certain to be far more complex than our present understanding. In current literature, the relevance of the Ice-Free Corridor has largely been relegated to a subsidiary role for the northward expansion of fluted point technology in a Younger Dryas time frame. Too sharp a focus on that phenomenon fails to account for earth science and archaeological data indicating that the CIS and LIS decoupled ∼14,000 years ago, that the entire Corridor had functioning large mammal ecosystems no later than 13,200 years ago, and that established Indigenous populations in eastern Beringia and south of the ice masses had access to biotically habitable northern and southern “funnels” of the Corridor by ∼13,500 years ago. A number of “classic” Clovis forms and related technologies (e.g., macroblades, end-struck bifacial preforms) would also suggest that the Corridor was occupied by fluted point makers in the 12,900–13,100 year time frame.

A few of the stemmed point morphologies presented here raise the possibility that populations using this projectile technology were present in the Corridor in either pre-Clovis or Clovis contemporary periods (i.e., the Bolling-Allerød time frame and forms Rosencrance et al. (2024) identified). Stemmed point morphologies and the few well-dated archaeological sites in or near the Corridor indicate, however, that the majority of stemmed points reviewed here likely come from Younger Dryas and early Holocene time frames. In many cases, these points are poor fits for standard Great Plains typologies but are quite often indistinguishable from typical WST or Alaskan stemmed point variants, whether we speak of Lind Coulee, Haskett, Parman, Cougar Mountain, Lake Mohave, Sluiceway, Mesa or other forms. These interpretations are reinforced by relatively rare, but nonetheless far-ranging early period transport of exotic toolstones involving Oregon, Idaho and Alaskan obsidian sources, as well as Knife River Flint (Kristensen et al. 2018; 2023). They are further warranted by associated technologies reflected in Campus-style microblade cores of ultimate Denali Complex origin.

In an earlier era, assertions that there might be a widespread presence of early period Lerma or generalized lanceolate points could be countered by positing that these are generalized forms with limited diagnostic value, that might appear at a number of different times and places. While this article represents an initial foray into a census of stemmed points in the Corridor, there are too many striking resemblances and either broad or more precise examples of contemporaneity to reflect a series of coincidences. There is a demonstrable degree of geographic connectivity across vast Late Pleistocene and Early Holocene environments that may well have encouraged Indigenous exploration. To what degree that connectivity was the outcome of demic expansion or cultural diffusion – or more likely, some combination of the two – we cannot at present be certain. These distinctions should, however, be a prime research focus for the future. More opportunities for radiometric dating would always be welcome for the Great Basin (but see Rosencrance et al. 2024), and especially the Corridor. Regardless of that constraint, dedicated study of stemmed points ought not to be confined to the Pacific Coast and immediately adjacent regions.

Other authors have raised the possibility that there existed, to varying degrees, a technocomplex in the sense that Clarke (1968) used that term, capable of linking shoulderless stemmed points in Alaska, or even encompassing similar northern Great Plains, Great Basin and southern Plateau forms (e.g., Admiraal 2013; Kunz 2011; Rasic 2011). The Corridor data presented here fills in an extensive distribution of oblanceolate and leaf-shaped points, at times coupled with microblade technology. Although by now well over a millennium into early Indigenous life in the western hemisphere, population densities were presumably still low, but undoubtedly greater than those of the first Indigenous Ancestors. That factor, along with the great geographic scope, makes it unlikely that a single group or culture could be involved at this scale. These occurrences nevertheless confirm that there was significant north–south as well as east–west communication in the Corridor region in post-Clovis times, dramatically accelerated by Cody Complex times.

The precise Haskett and Agate Basin-Hell Gap resemblances also suggest another fruitful area for further thought. As helpful as regionally developed taxonomies can certainly be, it is pertinent to ask if they may not obscure broader continental patterns. It is at times difficult to say, for instance, precisely what distinguishes Haskett points from Hell Gap examples. There is a legitimate possibility, as suggested by Bryan (1980) – and given the Younger Dryas (or slightly earlier) age now demonstrated for Haskett forms – that WST armatures provided the impetus for a Great Plains shift toward shoulderless and shouldered Plains forms in the wake of the fluted point era (see also Amick (2013) regarding Cody occurrences in the Great Basin and southern Plateau).

There is one strong distributional dichotomy in the early Holocene. While it would be helpful to learn a great deal more about the Northern Plano phenomenon, there is a clearcut relationship between it and the Cody Complex, which has a remarkably strong southern presence in the vastly expanded Corridor region. The inset in Figure 11 shows the high density of Cody Complex diagnostics for Alberta in the northern Plains, with its intensive ties to the Knife River Flint source area. That distribution holds for southern and central Alberta, along with extensions into the Peace River Country of northwestern Alberta and northeastern British Columbia, as well as the lower Athabasca River region of northeastern Alberta (Figure 11; Ives 1993; 2017; Reeves, Blakey, and Lobb 2017). This may in part arise from non-analogous post-glacial environments. Large alkali lakes (or, as in the case of the Fletcher site, large lakes that disappeared altogether) in southern and central Alberta were likely freshwater in the more immediate aftermath of glacial retreat. Terrestrial environments involved the time-transgressive appearance of grasslands as well as the presence of white spruce gallery forest environments (Beaudoin and Oetelaar 2003; Ives 2023). Similar conditions also existed in the Peace Country and lower Athabasca regions, all of them potentially attractive to a Cody Complex lifestyle.

Figure 11. Northern Plano and Cody spheres in the greater Corridor region, with a Holroyd collection Alberta point from northwestern Alberta, and a burinated point from Fort Chipewyan, Lake Athabasca. Inset to right: density isopleths in western Canada and the northern tier of states for Cody Complex diagnostics. Base map after Dalton et al. (2020), indicating retreat of Keewatin ice dome.

There must also have been a key social dimension to this, however: the extreme long-distance transport of Knife River Flint appears to reflect some form of craft specialization and over-production of Cody Complex materials in the North Dakota source area surely coupled with other cultural factors (Root 1997; Root, Knell, and Taylor 2013). Those other factors might have encompassed anything from highly exogamous marriage practices, to trading partnerships at major seasonal gatherings, to entrenched rite of passage voyaging to the source area (e.g., Burch 1970; Ives 2015; 2024a; Speth et al. 2013). Whatever fueled this unusual Knife River Flint distribution in Cody Complex times had consequences with an indelible archaeological signature.

Beyond that geography, however, Cody Complex materials are absent from Subarctic environments to the north, where Northern Plano forms hold sway (Figure 11). Agate Basin-like points are known throughout Alberta and Saskatchewan, and Northern Plano points do resemble “rough and ready” Agate Basin forms (which may even have “morphed” into broadly notched “Kamut” instances). Northern Plano assemblages therefore seem to have an origin with Agate Basin-like forerunners, but one distinct from Cody Complex developments.

Precisely how these archaeological portions of the equation might fit with current genetic data is even less certain – there are a good many more questions than answers, particularly given the small sample sizes involved. Northern Native American (NNA), Southern Native American (SNA), Ancient Beringian and Big Bar lineages were all extant in the terminal Pleistocene and Early Holocene time periods considered here, and they are regionally relevant (Moreno-Mayar et al. 2018b; Willerslev and Meltzer 2021). The known associations between early technological traditions and genetic identities south of the continental ice masses are nevertheless restricted to the SNA lineage: that is true of WST manifestations at Paisley Caves as well as Spirit Cave, and the Clovis assemblage connected with the Anzick child (Rasmussen et al. 2014; Willerslev and Meltzer 2021). It is the Ancient Beringian lineage that is connected with the Upward Sun River children, and we have little idea of where and to what extent the Big Bar lineage enters into the picture, although that population persisted into the mid-Holocene (Cybulski et al. 2007; Willerslev and Meltzer 2021). Yet, virtually all of Canada and interior Alaska today and in the recent past are homelands for members of the NNA lineage (that is, ultimate Haida, Tlingit, Tsimshian, Salishan, Wakashan, Dene, and Algonquian Ancestors), leaving some profound questions unresolved.

As was well established in the Boasian era, language, culture and human biology vary independently, a reality cogently explored in the identification of distinct Gravettian cultural groups (as revealed by patterns of adornment) and their relative “mismatch” with contemporaneous genetic identities (Baker et al. 2024). The great strength of genetic interpretations lies in the detection of common ancestors and relationships (with some temporal guidance from molecular “clocks”). The strength of the archaeological record lies in more precise determinations of when and where divergences and convergences could conceivably have taken place.

Limiting our discussion to the matter of NNA dominance in northern North America, and given the degree of uncertainty about precisely where the NNA-SNA split occurred, it will remain useful to bear in mind the capacity of an early deglaciated Corridor to have provided isolation of lineages, either at their inception (potentially in eastern Beringia or from an early entrance to the southern funnel of the Corridor), or subsequently. The near total dominance of locally and regionally available toolstones for fluted and stemmed points in the Corridor suggests that a relative degree of isolation is in fact plausible in an earlier time range (Ives 2006; 2015; Ives et al. 2013; 2019). That situation might be important in NNA expansion outward from the Corridor region – the first accessible landscape in Canada after deglaciation. Or is it more likely that NNA dominance of northern North America actually emerged in the Early Holocene, in that Cody Complex-Northern Plano time frame, where NNA populations spread extensively northward with shoulderless stemmed points, but could also have interacted significantly with SNA populations in the southern portions of the Corridor (creating the later, admixed populations described by Reich et al. (2012), Moreno-Mayar et al. (2018b), Scheib et al. (2018), and Willerslev and Meltzer (2021))? Given the presence of some early period microblade technology of Denali origin in the Corridor, as well as the leaf-shaped and oblanceolate points, is it possible that there was an Ancient Beringian presence (or at least, set of influences) within the Corridor?

Our enquiries into these and many other questions will be well-served by seeking a better understanding of both stemmed point and fluted point presence in the Corridor, by continued consideration of extensively exposed reservoir shorelines and their substantial collections (as at the Saint Mary, Brazeau and Williston Reservoirs of Alberta and northeastern British Columbia), and hopes for discovery of additional, well-dated sites as informative as Wally’s Beach, Vermilion Lakes, and Tse’K’wa.

Acknowledgements

I am grateful to the Royal Alberta Museum, the Grande Prairie and Peace River Museums, and the several avocational collectors acknowledged in the figure captions. From the Archaeological Survey of Alberta, Todd Kristensen’s work with Peace Country collections has been particularly valuable, as has been Darryl Bereziuk’s assistance for the same region, along with the Fort Chipewyan Airport collection. Kyle Forsythe, Bob Dawe and Sage Wigger of the Royal Alberta Museum assisted with access to and photos of Wally’s Beach, Poohkay and other stemmed point materials. Various photographs of the Iverson and Graham collections from Dale Fisher were most appreciated. Thanks to Reid Graham (then of Ember Consulting) for Brazeau photographs. Mike Kunz and Jeff Rasic generously provided comparative material from Alaska, as did Daron Duke (Far Western) for Utah. Thanks to Gabriel Yanicki for Canadian Museum of History photos of Batza Téna, Yukon and Wally’s Beach materials as well as to Gwynn Langeman, then of Parks Canada, for opportunities to photograph Vermilion Lakes and Eclipse Phase artifacts. David Meyer, University of Saskatchewan and Muriel Carlson assisted with photographs of Phenix and other collection artifacts. Jen Hallson and Gabriel Yanicki’s work with Ahai Mneh artifact distributions has been most helpful. Thanks to Duane Froese and the A. E. Lalonde AMS lab for the Athabasca area bison (University of Alberta paleontological collections) radiocarbon date reported in the supplementary material. Thanks also to Richard Rosencrance and Nathaniel Nathaniel Kitchel for their invitation to participate in both the 2023 Society for American Archaeology session and this issue, as well as to Geoff Smith and two other reviewers for helpful comments that improved the article.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

John W. Ives

John W. Ives earned his PhD from the University of Michigan. He is an emeritus professor in the Department of Anthropology at the University of Alberta in Edmonton. His research interests include the archaeological records of the Plains, Subarctic, Great Basin, and northeast China, PaleoIndigenous studies, Apachean origins, kinship and socioeconomic organization, archaeological theory, and public archaeology. He is the author of A Theory of Northern Athapaskan Prehistory and co-editor of Holes in Our Moccasins, Holes in Our Stories, Apachean Origins and the Promontory, Franktown, and Dismal River Archaeological Records.

Notes

1 By “years ago,” I am referring to calendar ages before present (AD 1950); radiocarbon ages will be given with standard errors and laboratory numbers.

2 The Lindoe site (EaPo-9) is a bison kill site located on the north bank of the South Saskatchewan River about 16 km downstream from Medicine Hat (Bryan 2000). The Lindoe stemmed point was recovered on a talus slope immediately below a bison bone bed in a peat deposit dating to 9900 ± 120 ¹⁴C yr BP (S-230). Caution is warranted in interpreting this early, conventional date, but it is a plausible age.

3 Rather than always referring to specimens (especially larger ones) as points or knives, I will use the term “point” more broadly. Given the size and nature of the collections discussed here, it has not been possible to conduct use-wear studies for each artifact through which function might be inferred, as valuable as that would be. Smith et al. (2024) noted that over-sized Parman specimens are knives; Rasic (2000) observed that larger Sluiceway points (to be discussed later), often showing impact damage, may in many cases have ended their use life as smaller resharpened knives. It is worth bearing in mind that these distinctions may be important for many of the artifacts illustrated here, whether large in size or heavily retipped, as many are.

4 The Brazeau Reservoir has also produced two fluted points and a 12,700-year-old Equus tooth row near the find spot for this stemmed point (Brink et al. 2017).