Document Type


Publication Date


Archaeological Report Number



With grant support from the National Park Service, Department of the Interior, administered by the Bureau of History, Michigan Department of State, a team of archaeologists from Western Michigan University has undertaken a program of fieldwork (with appropriate literature search and review of the documents) to identify archaeological sites and ascertain the nature of the activity conducted from them in an attempt to explain the nature of the relationship between the native inhabitants of Kalamazoo County and the former grassland environments that occurred here.

A review of the relevant literature prior to initiating a program of survey on Gourd-Neck Prairie in southern Kalamazoo County during Spring 1985, strongly suggested that contact period village sites and prehistoric works including mounds, earthen enclosures, and garden beds were associated with the former prairies in the county. However, reconnaissance level survey work undertaken in 1979 on Gull and Toland's prairies by teams of experienced surveyors had resulted in confirmation of a single previously recorded site and the discovery of no new sites; albeit surveyors systematically evaluated more than 5.4 km2 of farmland affording excellent conditions of surface visibility.

Before concluding that the documents provided by early American residents were inaccurate or incorrect, a more vigorous test of the hypothesis that Indians intensively occupied the prairies prior to American settlement of the county was required. Gourd-Neck Prairie in Schoolcraft Township {T4S RllW) was selected for several reasons: (l) the prairie is reported to have encompassed slightly more than 10 km2 , making a target of 100% surveyor coverage attainable with a small field party and a brief period in which to accomplish the fieldwork; (2) the personnel participating in the project were already familiar with the area, having established important landowner and collector contacts during the 1982 and 1984 field seasons; and (3) the former prairie was now characterized by extensive commercial farming operations, providing for anticipated surface visibility that would be excellent for a program of research employing surface reconnaissance procedures to record archaeological observations.

During a two week period in Spring 1985, a team of three surveyors evaluated 818 ha (2022 acres) or 81% of the area formerly supporting prairie vegetation. In addition, we surveyed 319 ha (788 acres) in adjacent areas that formerly supported oak savanna and bur oak openings. Fourteen new sites were recorded, and four previously recorded sites were revisited during the course of fieldwork. Of the new sites, six occur on the prairie and eight are located near creeks or standing bodies of water to the north, east, and south of the former native grassland. Simila~ly, all previously recorded sites lie between the prairie and Portage Creek and the north shore of Barton Lake on the southeast margin of Gourd-Neck Prairie.

Our analysis of these data suggests that sites occurring on the former prairie represent very task specific or limited activity loci (e.g. the loss of a projectile during an episode of hunting), with the more intensively occupied settlements being situated in oak savanna and bur oak openings affording greater access to resources (e.g. wood) i i deemed critical to support a camp or village and also closer proximity to nearby, resource rich wetlands and the lakes and streams that they flanked.

Clearly, our research to date strongly suggests that the historical documents must be more critically evaluated before recording locations referenced in them as bona fide archaeological sites providing distributional information useful in better understanding settlement patterns (and subsistence practices) of the native inhabitants of Kalamazoo County, Michigan.