Introduction

Archaeological surveys in the Rocky Mountains near Yellowstone National Park have identified seven prehistoric sites associated with melting perennial snow drifts or 'ice patches' at elevations above 3000m asl (Figure 1). In addition to palaeobiological remains, discoveries include organic and chipped stone artefacts and butchered animal remains exposed by the melting ice. Ice patches characteristically exhibit little internal deformation or movement, and may contain ancient ice that, unlike glaciers, is kinetically stable and preservative (Hare et al. 2004; Dixon et al. 2005; Grosjean et al. 2007). The most remarkable and oldest artefact recovered in this context is a complete wooden dart foreshaft described here (Figure 2).

The wooden foreshaft

A small sample of wood taken from a break in the foreshaft (Figure 1C) was AMS ¹⁴C-dated to 9230±25 BP (CURL-9635; wood; δ13C=-23.0‰); calibrated to 10 281–10 497 cal BP (p = 1.0 [2 sigma]) (CALIB 2010). This artefact is contemporary with the late Paleoindian Cody complex (c. 9 200–8 400 ¹⁴C BP or 11 220–9 445 cal BP) in North American archaeology (Reference DixonDixon 1999), and is, to our knowledge, the oldest artefact recovered from melting ice in the world. Stone tools characteristic of the Cody complex are well documented from sites in the Great Plains and Rocky Mountains (Reference PitbladoPitblado 2003). Comparatively few details are known about specific subsistence practices at high altitudes or the organic technology associated with the complex.

Figure 2.

Dart foreshaft. Clockwise from large image: A) the complete foreshaft; B) detail of the hafting element at the tip (the probable ownership marks are visible near the bottom of the image); C) detail of animal damage (probably a trampling fracture that occurred when the artefact was saturated and partially buried in slush); D) detail of the base portion of the foreshaft. Scales in centimetres. Photographs: Tara L. Hornung.

The foreshaft is 1.07m long and made from a birch (Betula spp.) sapling, trimmed of its branches (Figure 1). The upper portion has been deformed, possibly by flowing slush on the surface of the ice patch when the artefact was saturated and partially exposed; the base portion retains its original shape. A hoof print at the major break near its mid-point suggests the artefact was trampled by an ungulate (e.g. bighorn sheep [Ovis canadensis]) (Figure 1C). The base tapers symmetrically in diameter from 12.5mm to 3.6mm (Figure 1D) presumably to fit a corresponding socket. Younger comparative material from Gypsum Cave in the Great Basin suggests foreshafts of this length may have functioned with a c. 300mm long dart (Reference HarringtonHarrington 1933). The complete weapon was probably propelled by an atlatl, or spear thrower, which would have provided mechanical advantage by increasing the leverage of the thrower's arm, resulting in greater projectile velocity.

The tip of the artefact, where a stone projectile point (absent) would have been hafted, is 8.9mm in diameter (Figure 1B). The specimen exhibits a 'split end haft' 2.7mm wide and 12.6mm deep, which is consistent with postulated Paleoindian period hafting techniques (Reference DixonDixon 1999). Overcuts preserved in the soft wood suggest the hafting element was manufactured using a sharp stone tool. The sides of the clothespeg-like haft that secured the projectile point are rounded and smoothed for better penetration.

Two groups of three evenly spaced lines are present on opposing sides of the artefact 40.4mm and 52.6mm from the tip (Figure 1B). They are c. 4mm long and 1.5mm apart and may represent ownership marks. Ethnographic observations indicate ownership marks occur on hunting weapons designed to remain in the bodies of large game. They typically consist of simple lines, and can be specific to either an individual or community (Reference BoasBoas 1899). The lines on the foreshaft appear to be embossed or pressed into the wood with a c. 1mm wide tool on the thinnest portion of the shaft near the projectile point haft. If the shaft broke off inside the animal, this portion could link the hunter with the kill.

Social and environmental significance

The ownership marks imply social protocols necessitating the need to identify individual hunters following successful hunts. Ethnographic analogy drawn from hunter-gatherers elsewhere indicates animal products (meat, hides and horn) were distributed according to prescribed social protocols based on kinship and other criteria (Reference WiessnerWiessner 1983). These protocols most probably existed in Paleoindian societies. Unlike Plains bison, bighorn sheep are rarely found in large numbers and it is unlikely that they were the focus of the large communal hunts important during the Cody period (Reference DixonDixon 1999; Reference PitbladoPitblado 2003). Because ice patches only concentrate animal and human activity in the summer, this site is interpreted as documenting summer exploitation of bighorn sheep (Ovis canadensis) by individuals or small groups of Cody complex hunters more than 10 000 years ago.