Obsidian in Vermont: Analysis of an Arrowhead in the Gerald Coane Collection

on Jul 27th 2008

by Matthew T. Boulanger, Archaeometry Laboratory, University of Missouri Research Reactor and Thomas R. Jamison, Hartgen Archeological Associates, Inc., Putney, Vermont


Archaeologists are particularly interested in identifying evidence of prehistoric long-distance trade and exchange, and artifacts made from stone are some of the best records of such exchange because they can be traced back to specific geological outcrops. Archaeologists often develop an intuitive knowledge about the types of stone and their potential sources that were used prehistorically. In Vermont for example, most archaeologists recognize quartzite from the Cheshire formation or chert from the Champlain Valley. But, when archaeologists encounter an artifact made from stone not found in their region of inquiry, they use the term “exotic” to describe it.

Occurrences of so-called exotic artifacts are not uncommon in Vermont. Loring (2002) reports a biface found in Franklin County that appears to be made from quartzite from northern Labrador. Other archaeologists have found white chert artifacts believed to be from the Midwest (Cassedy 1991: 19, 30; Loring 1985; Olsen 1936). An obsidian arrowhead in the collection of Gerald B. Coane, curated by the Putney Historical Society (PHS) is another occurrence of a so-called exotic artifact reportedly found in Vermont; however, the distances between Vermont and any known obsidian source are substantially greater than chert sources in the Midwest or Labrador. Could this obsidian artifact have been traded into Vermont prehistorically? We have used multiple analyses to provide information about the geological source and the relative age of this artifact because we believe that these factors can be used to assess the authenticity of this arrowhead. We also hope to direct professional attention to the condition of artifact collections throughout Vermont, collections that have long been ignored and that are in dire need of inventorying and analysis.

Brief History of Gerald Coane and his Collection

Born in 1898 in Brattleboro, Vermont, Gerald B. Coane (Figure 1) assembled a moderately sized artifact collection during the middle twentieth century. Coane worked in a local factory after having served in World War I. Later in his life, he operated a railroad crossing over the Connecticut River. After retirement in the early 1960s, Coane began collecting artifacts from archaeological sites in southern Vermont on a somewhat regular basis. Like many collectors in the State, Coane presented lectures on archaeology and history to local schools, and he used his artifact collection as an educational tool. Coane donated his collection to the PHS in the early 1970s, where it remains today. A catalog of his finds indicates that Coane focused his collecting close to home. Although he did not record find spots (provenience) for all of the collection, most of the artifacts appear were found at the West River site (VT-WD-3) located at the confluence of the West and Connecticut rivers. Like many artifact collections held in private institutions throughout the state, Coane’s collection has received only passing attention from the professional community.

Obsidian Artifacts in the Collection

Coane listed two obsidian artifacts in his catalog. The first piece, numbered Y10, is described as a “Large obsidian (lava glass) arrow head. Made by the Tarahumare [sic] Indians of northern Mexico.” Immediately below this entry, Coane listed artifact number Y11: “Obsidian arrow head found on the shore of the Connecticut River” (Figure 2). The Mexican arrowhead appears to be a tourist-trade item, and, for the purposes of this report, we will only refer to the arrowhead reportedly found in Vermont.
The first mention of the point in a professional context is a brief examination of Coane’s collection made by Stephen Loring and Shelley Hight in 1978 under the auspices of the Vermont Collections Survey. The two researchers visited the museum in the late afternoon after having spent the better part of the day surveying other collections in southeast Vermont. An early snowstorm and the absence of heat in the PHS museum produced somewhat uncomfortable working conditions, yet they photographed and briefly described the collection before leaving for the day. Loring’s description of the collection drew some general conclusions about its potential usefulness in addressing research questions in the Connecticut Valley. Although he mentioned the obsidian arrowhead, he did not speculate as to its origin.

In 1988, the point reported to have been found in Vermont caught the attention of Daniel Cassedy, then working at the Vermont Division for Historic Preservation (VDHP). Cassedy, collaborating with a researcher at the State University of New York at Albany (SUNY–Albany), attempted to conduct nondestructive trace-element analysis on the point to determine its geological source. The researchers could not match the point to any obsidian sources they knew about, and they retained possession of the piece to conduct at least one more analytical technique on the piece to obtain more detailed information on its composition.
Several years after loaning the point, a representative of the PHS realized that it had not yet been returned to their museum. Cassedy had left the VDHP by this time, and his colleague at SUNY–Albany was no longer at the laboratory on a regular basis. A series of letters among the PHS, VDHP, Cassedy, and SUNY–Albany was necessary to locate the arrowhead and ensure its return to the PHS. Since then, the piece has been held by the PHS, and no further attempts at analysis have been made.

Other Finds of Obsidian in Vermont?

Unlike the types of stone commonly found in Vermont, obsidian is very distinctive and easily identified by its glass-like appearance and sharp edges. Is it possible that other pieces of obsidian have been found in the state? We are aware of only one other obsidian artifact purported to have been found in Vermont. This piece, reportedly found by collector Paul Bilhuber at the Donovan site (VT-AD-2), cannot be located at present, and we believe that if it ever existed, the piece is now lost to history. Importantly, no other finds of obsidian from archaeological sites in the state have been reported. This fact alone leads us to view with some skepticism the reported provenience of Coane’s obsidian point as well as Bilhuber’s claim. However, archaeology is a science, and we have decided to pursue a comprehensive analysis of the point to test the proposition that it is an authentic prehistoric artifact that was traded into Vermont before European contact.

Analysis of the Point

Determining the absolute age of a stone point is not a straightforward process. Radiocarbon dating cannot be conducted on stone, and because the reported provenience of the point is so vague, we cannot identify and excavate the specific location from which the point was taken. Two potential techniques exist for establishing an approximate age for the arrowhead: Culture-historical typology and obsidian hydration dating.

Morphologically, the point (Figure 2) is notched and has a bifurcate base. As such, it could be grouped with the Bifurcate-base Point tradition of the Early Archaic period in New England. Specific point types in this tradition are poorly defined; however, the point resembles somewhat the Swanton Corner-Notched type defined in northern Vermont (Haviland and Power 1994; Thomas and Robinson 1980). Importantly for our discussion, the Early Archaic is poorly documented in Vermont, and long-distance exchange is not considered prevalent during this period.

Designation of the point as being of a specific “type” is not sufficient alone to confirm an age for the point. Perhaps the arrowhead could be a modern creation made to look like points from the Early Archaic. We used obsidian-hydration dating to test this possibility and to determine whether the point is in fact a prehistoric creation.

Like all archaeological and geological dating techniques, obsidian-hydration dating has various benefits and drawbacks. Put simply, when obsidian is fractured, the newly exposed surface begins to capture moisture (hydrate). Measurement of an obsidian hydration rim involves preparing a thin-section of the artifact and determining the thickness of the rim of hydration. Basically, a thicker hydration rim indicates a greater age for the last time the obsidian surface was broken. However, rates of obsidian hydration are not universally constant. Local variations in temperature and humidity, and the chemical composition of the obsidian itself, affect how quickly an exposed surface will hydrate. Despite these factors, obsidian hydration can be a useful technique for establishing a relative age for an artifact provided that all of the examined obsidian artifacts come from the same geological source. As such, we sought first to establish the geological source of the obsidian.

The chemical composition of the arrowhead was determined at two facilities: the University of Missouri Research Reactor (MURR) Archaeometry Laboratory, Columbia, and the Northwest Research Obsidian Studies Laboratory (NWORSL), Corvallis, Oregon. Analyses at MURR were done first using X-ray fluorescence spectrometry (XRF). XRF analysis is essentially nondestructive, and we conducted a series of six assays at varying locations on the arrowhead times. The resulting compositional data were compared against the MURR database consisting of over 250 obsidian sources characterized by XRF. A second assay of the artifact was conducted using neutron activation analysis (NAA). Analysis by NAA requires that a small portion of the artifact be removed and irradiated. Unlike XRF, NAA allows analysis of the entire piece and it provides data for up to 34 elements. MURR has characterized over 15,000 individual pieces of obsidian by NAA including over 500 geological obsidian sources.

Analysis by XRF and NAA both demonstrate that the arrowhead is made from obsidian originating at the Double H source in the northern Great Basin, near the Oregon/Nevada border. Although this source was used prehistorically, obsidian from it was not widely traded outside of the Great Basin (Jones et al., 2003). Further, studies of obsidian artifacts found in verifiable archaeological contexts east of the Mississippi River (e.g., DeBoer 2004; Griffin et al., 1969; Hatch et al., 1990) have never documented obsidian from this particular source.

Following completion of the assays at MURR, the arrowhead was sent to Dr. Craig Skinner at NWROSL for a blind XRF assay and analysis by obsidian hydration. Skinner independently verified that the point was made from the Double H obsidian source. Further, he determined that the point had a hydration rim roughly 7.1 microns thick. This hydration rim was compared to 110 other hydration rims taken from artifacts from Double H, and it fell within the upper five percent of samples. Though it is not reasonable to attempt converting this hydration-rim thickness to a calendrical date, the rim thickness demonstrates that the point was created before European contact, and that it is not a modern fake.


Our results demonstrate that the obsidian arrowhead in Coane’s collection is a prehistoric artifact made of volcanic glass from a relatively obscure source in the northern Great Basin. The remoteness of this obsidian source, and the distance from it to Vermont, suggest that the Coane biface is very likely not the product of prehistoric exchange. The fact that obsidian pieces from the Double H source have not been identified in secure archaeological contexts east of the Mississippi River (e.g., Griffin et al., 1969; Hatch et al., 1990) further supports our conclusion that the piece was not traded to the Northeast before European contact. Further, once the source of the point had been clearly established, we compared its morphology against established typologies of the Great Basin. We found that the point clearly fits the type definition of the Elko Eared type (Justice 2002: 298–310, Fig. 27, Thomas 1981: 20–22, Fig. 8).

Though we admit that it is possible that the arrowhead may have been brought into the Connecticut River valley prehistorically, we believe that the evidence here shows that such a scenario is highly implausible. A more reasonable explanation for the evidence at hand is that the artifact was historically brought to Vermont, and that it somehow came into the possession of Coane. It is impossible to state how Coane obtained this arrowhead, or whether he did in fact find it in the Connecticut River drainage. However, data presented here strongly suggest that the point is not the result of prehistoric exchange into Vermont.

In concluding, we wish to stress that we do not dismiss out of hand the potential for long-distance exchanges of material or ideas by early Native Americans—either into or out of the territory now known as Vermont. Nor do we consider it impossible that such exchanges could have occurred over great distances. However, fantastic claims such as these must be able to withstand scientific scrutiny, and there must be evidence offered to support them.


We take this opportunity to stress to anyone conducting archaeological work the importance of maintaining a detailed and accurate catalog of any artifacts they may find and collect. Although we do not encourage illegal collecting or looting of artifacts, we do recognize that much of the current body of archaeological knowledge in Vermont is a product of avocational archaeologists who have been picking up arrowheads on their family farms for several generations. We encourage all such collectors to be ethical and responsible about the knowledge that they generate, and to consider how best to preserve that knowledge for the benefit of future generations of curious Vermonters.

An expanded and detailed description of the analyses described in this article may be found in the forthcoming edition (Volume 35) of Archaeology of Eastern North America. Questions concerning the Coane collection and the PHS may be directed to Tom Jamison via email at tjamison@hartgen.com. Questions about the analytical techniques used may be directed to Matt Boulanger at boulangerm@missouri.edu. Information about the analytical techniques discussed here may be found at the Web sites of the MURR Archaeometry Laboratory (http://archaeometry.missouri.edu/) and the Northwest Obsidian Studies Research Laboratory (http://www.obsidianlab.com/).

Figure 1. Photograph of Gerald Coane with shovel in hand ca. 1965. Coane is believed to be standing on the bank of the Connecticut River at the site of Fort Dummer. Wantastiquet Mountain and the railroad crossing of the Connecticut are visible in the background.

Figure 2. Dorsal sides of the two obsidian points found in the Coane collection. Left: Catalog number 11, a notched and indented-base obsidian biface reportedly found “On the Shore of the Connecticut River” in Vermont. Right: Large triangular biface reported as “[m]ade by the Tarahumare [sic] Indians of northern Mexico.” Scale is in centimeter


Cassedy, D. F.
1991 A Prehistoric Inventory of the Upper Connecticut River Valley. Garrow & Associates, Inc.

DeBoer, W. R.
2004 Little Bighorn on the Scioto: The Rocky Mountain Connection to Ohio Hopewell. American Antiquity 69(1):85-107.

Griffin, J. B., A. A. Gordus and G. A. Wright
1969 Identification of the Sources of Hopewellian Obsidian in the Middle West. American Antiquity 34(1):1-14.

Hatch, J. W., J. W. Michels, C. M. Stevenson, B. E. Scheetz and R. A. Geidel
1990 Hopewell Obsidian Studies: Behavioral Implications of Recent Sourcing and Dating Research. American Antiquity 55(3):461-479.

Haviland, W. A. and M. W. Power
1994 The Original Vermonters: Native Inhabitants, Past and Present. Revised, Expanded ed. University Press of New England, Hanover.

Jones, G. T., C. Beck, E. E. Jones and R. E. Hughes
2003 Lithic Source Use and Paleoarchaic Foraging Territories in the Great Basin. American Antiquity 68(1):5-38.

Loring, S.
1985 Boundary Maintenance, Mortuary Ceremonialism and Resource Control in the Early Woodland: Three Cemetery Sites in Vermont. Archaeology of Eastern North America 13:93-127.

2002 “And They Took Away the Stones from Ramah”: Lithic Raw Material Sourcing and Eastern Arctic Archaeology. In Honoring Our Elders; A History of Eastern Arctic Archaeology, edited by W. W. Fitzhugh, S. Loring and D. Odess. Contributions to Circumpolar Anthropology. vol. 2. Arctic Studies Center, National Museum of Natural History, Smithsonian Institution, Washington D.C.

McBride, K. A.
1984 Prehistory of the Lower Connecticut River Valley. Ph.D. Dissertation, University of Connecticut.

Olsen, G. J.
1936 Certain Archaeological Investigations in the Lake Champlain Valley, Including the Orwell, (Vermont) Burial Site. In East Creek Collections file, Vermont Division for Historic Preservation, Montpelier, Vermont.

Thomas, D. H.
1981 How to Classify the Projectile Points from Monitor Valley, Nevada. Journal of California and Great Basin Anthropology 3(1):7-43.

Thomas, P. A. and B. S. Robinson
1980 The John’s Bridge Site: VT-FR-69 Area I: Phase III Data Recovery Project BRS 0285 (2). Department of Anthropology, University of Vermont. Report No. 28. Copies available from the Vermont Division for Historic Preservation.

Filed in Articles | No responses yet

Trackback URI | Comments RSS

Leave a Reply

You must be logged in to post a comment.