Houpt site and the Late Archaic of southwestern Ohio, The

Houpt site and the Late Archaic of southwestern Ohio, The

Duerksen, Ken


Investigations at the Houpt site (33BU477) recovered data from a well-preserved example of a briefly occupied Late Archaic extraction camp. The site was situated on a low hummock rising above the floor of a wetland in an extinct pre-Holocene riverbed. Analysis of the lithic assemblage (including Merom/Trimble points) and ethnobotanical study and radiocarbon dating of feature contents are presented as part of a discussion of the site’s placement within the current conception of the terminal Late Archaic period in southwestern Ohio.


This paper describes and discusses a small Late Archaic habitation site in southwestern Ohio and its implications for local and regional settlement and subsistence analyses. Previous studies have defined two related and largely contemporaneous terminal Late Archaic cultural manifestations in the Ohio River valley: the Riverton culture in southern Indiana (Winters 1969) and the Maple Creek phase around the Indiana-Ohio state line and upriver (Vickery 1976). In the years since their definition, the Riverton culture has been examined by Anslinger (1986) and Bergman et al. (1990) and the Maple Creek phase has been discussed by Boisvert (1986) and Ledbetter and O’Steen (1992), among others.

Investigations relating to the Late Archaic period in the Middle Ohio valley region are overwhelmingly biased toward the study of large riverine sites, generally characterized as base camps. Other categories of Late Archaic sites are rarely given attention beyond the survey level of archaeological investigation and are often ranked within settlement models through inference alone. Winters’ (1969:131) definitive Riverton work in the Wabash valley did just this for the majority of sites he characterized as bivouacs, hunting camps, or gathering camps. Investigations at the Houpt site (33BU477), a relatively small Late Archaic encampment in southwestern Ohio (Duerksen and Doershuk 1994), offer insights into these less spectacular but nonetheless integral site types and the Late Archaic cultural environment of this region.

The Houpt Site

The Houpt site was identified, tested, and subsequently excavated in conjunction with a commercial development approximately 1.6 km southeast of the city of Hamilton, Ohio (Figure 1). Field work was conducted between October 1993 and February 1994 (Duerksen and Doershuk 1994).

The site’s environmental setting was greatly influenced by glacial activity and was certainly a central determinant of many aspects of the location’s utilization by prehistoric peoples. The .12-ha site was located on a small hummock rising above the floor of the wide valley left by an extinct tributary of the pre-Holocene Teays River system. The scar left by this extinct waterway averages 2.4 km in width, and offers a difference in relief of 30 to 60 m below the surrounding dissected uplands. Much of this valley was wetland historically, supporting several large natural bodies of water until drained for agriculture in the mid-nineteenth century. Local histories refer to the general area prior to that time with little affection. For example, one source notes “the swampy region which lies south and east of Hamilton was for a long time a great injury to the surrounding country” (Western Biographical 1882:476). Undoubtedly, prehistoric hunter-gatherers of the region had a much greater appreciation for this resource area, which would have supported a considerable diversity and abundance of important plant and animal species routinely utilized for subsistence.

Currently, the valley is occupied by Mill Creek, the headwaters of which originate in the uplands approximately 1.6 km northeast of the site area; these waters now flow to the south along artificial channels. All known prehistoric sites in this area are located on slightly elevated hummocks similar to that occupied by the Houpt site or on old terraces abutting the surrounding uplands.

Field work at the Houpt site consisted of controlled surface collections, hand excavation of 13 1 x 1-m units, and mechanical stripping of approximately 150 m^sup 2^ of the site’s plowzone. Nineteen subplowzone soil stains were identified and investigated, of which 14 were determined to be intact portions of cultural features (Figure 2). All artifacts from the site were found on the surface, within the plowzone, or in the context of truncated cultural features. In all, investigations recovered 314 pieces of lithic debitage, six temporally diagnostic projectile points, two retouched bipolar fragments, one microperforator, four hammer-stones, four greenstone flake cores, one greenstone cobble tool, and approximately 25 kg of fire-cracked rock (Table 1).

Culturally diagnostic stone tools from the Houpt site include four Late Archaic Merom/Trimble points, a single Early Archaic Big Sandy point (Justice 1987:61), and a single Middle Woodland Affinis Snyders point (Justice 1987:204). The latter two points were recovered from surface and plowzone contexts, and do not appear to be associated with any of the features or debitage concentrations. Conspicuously absent are examples of the McWhinney Heavy Stemmed projectile point type (Figure 3) identified by Vickery (1972, 1976) as a major component of the definitive Maple Creek phase assemblage. The absence of this point type from the Houpt site assemblage is unlikely a sample size problem as nearly the entire site was excavated. The McWhinney Heavy Stemmed projectile point type was absent from the Wabash valley Riverton sites investigated by Winters, but was the most frequently encountered diagnostic point type at the Maple Creek site (Vickery 1976:141).

Trimble Side-Notched and closely associated Merom Expanding Stem points (Figure 3) frequently co-occur within assemblages, a fact which has led to their being treated as a single temporally indicative type (e.g., “Riverton points” in Indiana). These points are universally small, averaging around 23 to 27 mm in length (DeRegnaucourt 1991; Justice 1987), and usually exhibit evidence of having been expediently manufactured through pressure flaking of flake blanks (Bergman et al. 1990). Their use as projectile points has been well established through both experimental studies and direct observation of impact damage on numerous specimens (Anslinger 1986; Odell and Bergman 1990), although Vickery (1976) indicates that they served primarily as knives at the Maple Creek site.

The four Merom/Trimble points recovered, including two from feature contexts dated to around 1150 B.C. and 1520 B.C. (calibrated), constitute the primary rationale for defining a terminal Late Archaic occupation as the major component at the site (Table 2). A third feature produced a comparable date of ca. 1510 B.C., while a fourth date of ca. 800 B.C. was derived from a hearth feature that possibly relates to a minor Early Woodland occupation of the site not represented in the diagnostic artifact assemblage.

The debitage assemblage is dominated by late-stage biface reduction debris. Initial reduction flakes and biface initial reduction flakes occur infrequently (11.8 percent and 5.4 percent, respectively). A single Janus flake, one of the initial flakes removed from the ventral surface of a flake blank (Tixier et al. 1980), was recovered (.3 percent of the Houpt site debitage). This suggests infrequent initial reduction of flake blanks at the site. Biface finishing flakes (21.3 percent), conversely, are represented more often than any other debitage category for which a specific reduction sequence could be determined. Comparison of these data with what is known about the reduction strategies associated with Merom/Trimble points suggests that stone tool maintenance was a far more frequent activity than initial manufacturing at the Houpt site.

The most frequently represented lithic raw material at the Houpt site is Laurel chert, obtainable at primary sources within 50 km west of the site. Chert in the form of locally available glacial cobbles and pebbles also was exploited by the site’s inhabitants. Wyandotte chert, the most common exotic lithic material from the site, occurs almost exclusively in the form of late-stage biface finishing/ resharpening flakes removed by pressure (n=10) or as broken bifacial tools (n=2). The remainder of the Wyandotte chert subassemblage consists of a single biface thinning flake and one flake unidentifiable as to reduction sequence position. The only two pieces of debitage identifiable as Coshocton chert, a high-quality material that outcrops in east-central Ohio, occur as finishing flakes removed by pressure.

The raw material data from the Houpt site suggest on-site resharpening and repair of curated tools manufactured elsewhere from high-quality, nonlocal material, and brought to the site after having been utilized and maintained for an indeterminate period of time. Referring to the model of site use and abandonment described by Stevenson (1985) for highly mobile hunter-gatherers, this type of debitage can be construed as representing activities performed early in a group’s sojourn at a site. Consequently, the earlier-stage reduction debitage, which at the Houpt site is composed of lower-quality, locally available glacial chert, would have been produced during the replacement of broken or exhausted tools, and as preparatory restocking as the group mobilized to relocate. This staged model of debris deposition implies an occupation span of short duration, but one which exceeds a merely transient level.

Other classes of data from the Houpt site help to clarify the magnitude and timing of this location’s Late Archaic occupation. The cultural features are densely concentrated on the highest, most level portion of the hummock. Placement of the features was apparently random as no clear clusters are present (although the distribution is clearly not uniform). Most of the features shared similar characteristics, being small (40 cm or less in diameter, and extending to a depth of approximately 15-20 cm below the plowzone), circular in plan view, and basin shaped or conical in profile. Feature fill from these pits was generally characterized by a ubiquitous flecking of carbonized material, a few flakes or lithic tools, and variable quantities of large pieces of fire-cracked rock. Artifacts recovered from feature fill include projectile points, hammerstones, biface fragments, retouched flakes and shatter, and a microperforator. Two features departed from this norm, both being relatively deep and more than a meter in diameter. One of these large pits (Feature 12) yielded lithic debitage, charred plant material, an impact-fractured Merom/Trimble point, and a battered and bifacially flaked igneous cobble. Feature morphology, in addition to the frequency and wide distribution of fire-cracked rock, indicates that stone boiling or baking, followed by occasional feature cleaning and maintenance, may have been common activities at the site.

Analysis of botanical samples recovered through the flotation of Houpt site feature fill, using standard SMAP-type flotation equipment, revealed small quantities of carbonized hickory (Carya sp.), walnut (Juglans nigra), hazelnut (Cor,lus sp.), and acorn (Quercus sp.) nutshell, as well as charred persimmon (Diospyros virginiana) and hackberry (Celtis sp.) seeds (Table 3). While significant concentrations of any one of these species, with the exception of persimmon, would be more likely to be found in an upland environment, this diverse combination of food-producing plant species is considered typical of a forested wetland system (Porter 1986). Hickory remains have the highest representation of any utilized plant species at the Houpt site, but densities are not particularly high for any of the features sampled compared with other Late Archaic sites (Munson 1986). Although the sample is small for accurately determining seasonality, these food plants most likely reflect late summer or fall usage.

The data reported above suggest multiple terminal Late Archaic occupations at the Houpt site. The calibrated dates from Features 1 and 12 are essentially contemporary (1510 and 1520 B.C.). However, the date of 1160 B.C. from Feature 6, while still firmly within the accepted Riverton and Maple Creek ranges, overlaps the others only at two standard deviations. Additionally, while the small size of the artifact inventory suggests low-intensity occupation, the diversity of artifact classes and raw materials present at the site implies a wide range of activities. The tight clustering of features, rather than representing intensive, contemporaneous utilization of the entire site area, probably reflects the limited size of this small, habitable landform and its persistent appeal to small groups of terminal Late Archaic peoples tapping the resources of the surrounding wetlands.

Discussion and Conclusion

In many areas in the Midcontinent, the size and apparent occupational intensity of some Late Archaic sites, usually in riverine settings, has led to their classification as permanent or semipermanent base camps from which a range of specialized extractive forays would have been made (Janzen 1978). One of the recognized archaeological correlates of such a collecting strategy is the occurrence of large feature clusters containing huge quantities of plant remains, representing the processing or storage of a single, densely occurring floral resource (Munson 1986; Stafford 1994). Boisvert (1986) suggests that the base camp site type may also reflect the accumulated debris of repeated, short-term occupations by small foraging bands at particular, highly desirable locations over long periods of time.

If the terminal Late Archaic occupation at the Houpt site reflects an extractive node of a logistically organized collecting system, then the data would suggest that floral resources were not the primary focus. The small quantities of ethnobotanical material from this site exhibit a low degree of specialization not in keeping with the large-scale processing for transport which should characterize such a system. An alternative possibility is that this occupation reflects the intensive acquisition of faunal resources, rather than plant gathering, by special-purpose hunting parties from a local base camp. Certainly, the wetlands which characterized the area prehistorically would have provided opportunities for the largescale exploitation of various mammal or waterfowl species. However, given the lack of faunal material from this site-an absence due to poor preservation conditions-such a situation is not explicitly demonstrable.

Another scenario perhaps more directly supportable by the Houpt site data is that the site was occupied by small groups which periodically split from larger, collecting-oriented macrobands to pursue a foraging strategy for a portion of the yearly cycle. This would be similar to the system described by Winters (1969) in his model of Late Archaic Riverton culture group fission and fusion. The Houpt site data also support Boisvert’s (1986) contention that some Late Archaic groups operated primarily as foragers, and that the base camp concept may be too readily invoked in some cases through misinterpretation of the debris from multiple occupations at single locations. In either of these models, the activities expected at locations like the Houpt site would be presumed to be identical, with diverse pursuits undertaken by site occupants but with low densities of material accumulating.

In sum, the Houpt site offers a relatively uncluttered view of a briefly utilized camp with an assemblage similar to those of both the Riverton culture and Maple Creek phase as they are currently conceived. It constituted a small but important element within the logistically mobile terminal Late Archaic subsistence and settlement system of the Middle Ohio River valley. Much additional work is needed to obtain a clearer understanding of Riverton, Maple Creek, and related peoples in this region.


Donald M. Houpt, III (namesake of the Houpt site) provided much of the motivation for this project-his influence simply can’t be understated. David Snyder, of the Ohio Historic Preservation Office, was instrumental in orchestrating the consultation surrounding the mitigation of unavoidable impacts to this resource. Both 3D/Environmental and the Office of the State Archaeologist of Iowa have generously supported our research efforts-this paper has benefited during its development from numerous conversations with colleagues at both offices. This is a retooled version of a paper first presented by the authors at the (combined) annual meeting of the Southeastern Archaeological Conference and Midwest Archaeological Conference in Lexington, Kentucky, in 1994 and subsequently at the annual meeting of the Ohio Archaeological Conference in Cleveland, Ohio, in 1995. We appreciate the helpful suggestions of a number of listeners at each of these presentations. David Pollack, David Snyder, Jim Mohow, and Al Tonetti each provided useful ideas as well. We thank the MCJA anonymous reviewers for their remarks and are especially grateful to Bill Green for his encouragement and editorial skills. As usual, we retain all responsibility for any remaining shortcomings, errors, or gaffs.


Anslinger, C. Michael

1986 The Riverton Culture: Lithic Systems and Settlement Parameters. Master’s thesis, Department of Anthropology, Washington University, St. Louis, Missouri.

Bergman, Christopher A., David J. Rue, and John F. Doershuk 1990 Riverton Lithics and Woodland Ceramics: Archaeological Data Recovery at 12B815, A Multicomponent Prehistoric Site in Bartholomew County, Indiana. Report prepared by 3D/Environmental, Cincinnati, Ohio, for Texas Gas Transmission Corporation, Owensboro, Kentucky. Boisvert, Richard A.

1986 Late Archaic Settlement Models in the Middle Ohio Valley:A Perspective From Big Bone Lick, Kentucky. Ph.D. dissertation, Department of Anthropology, University of Kentucky, Lexington.

DeRegnaucourt, Tony

1991 A Field Guide to the Prehistoric Point Types of Indiana and Ohio. Occasional Monographs No. 1. Upper Miami Valley Archaeological Research Museum, Ansonia, Ohio.

Duerksen, Ken, and John F. Doershuk

1994 Report on the Archaeological Data Recovery at 33BU477, A Late Maple Creek Phase Occupation in Butler County, Ohio. Report prepared by 3D/Environmental, Cincinnati, Ohio, for Kiesland Business Park, Butler County, Ohio.

Janzen, Donald E.

1978 An Examination of Late Archaic Development in the Falls of the Ohio Area. In For the Director: Research Essays in Honor of James B. Griffin, edited by Charles E. Cleland, pp. 123-143. Anthropological Paper No. 61. Museum of Anthropology, University of Michigan, Ann Arbor. Justice, Noel

1987 Stone Age Spear and Arrow Points of the Midcontinental and Eastern United States. Indiana University Press, Bloomington.

Ledbetter, R. Jerald, and Lisa D. O’Steen

1992 The Grayson Site: Late Archaic and Late Woodland Occupations in the Little Sandy Drainage. In Current Archaeological Research in Kentucky: Volume Two, edited by David Pollack and A. Gwynn Henderson, pp. 121-142. Kentucky Heritage Council, Frankfort. Munson, Patrick A.

1986 What Happened in the Archaic in the Midwestern United States? Reviews in Anthropology 13:276-282.

Odell, George H., and Christopher A. Bergman

1990 Microwear Analysis of a Sample of Artifacts from the Atterbury Site. In Riverton Lithics and Woodland Ceramics: Archaeological Data Recovery at 12B815, A Multicomponent Prehistoric Site in Bartholomew County, Indiana, by Christopher A. Bergman, David J. Rue, and John F. Doershuk, pp. 113-127. Report prepared by 3D/Environmental, Cincinnati, Ohio, for Texas Gas Transmission Corporation, Owensboro, Kentucky. Stuiver, Minze, and Paula J. Reimer

1993 Extended 14C Data Base and Revised CALIB 3.0 14C Age Calibration Program. Radiocarbon 35:215-230. Porter, Reed B., Jr.

1986 Wetland Plants of the State of Ohio. National Wetlands Inventory, U.S. Fish & Wildlife Service, St. Petersburg, Florida.

Stafford, C. Russell

1994 Structural Changes in Archaic Landscape Use in the Dissected Uplands of Southwestern Indiana. American Antiquity 59:197-218. Stevenson, Marc G.

1985 The Formation Processes of Artifact Assemblages at Workshop/Habitation Sites: Models from Peace Point in Northern Alberta. American Antiquity 50:6381.

Tixier, Jacques, M. Inizan, and H. Roche

1980 Prehistoire de la Pierre Taillee I, Terminologie et Technologie. Valbonne Cedex France.

Vickery, Kent D.

1972 Projectile Point Type Description: McWhinney Heavy Stemmed. Paper Presented at the 29th Annual meeting of the Southeastern Archaeological Conference, Morgantown, West Virginia.

1976 An Approach to Inferring Archaeological Site Variability. Ph.D. dissertation, Department of Anthropology, Indiana University, Bloomington.

Western Biographical Publishing Company

1882 A History of Butler County. (no publication location given). Winters, Howard D.

1969 The Riverton Culture: A Second Millenium Occupation in the Central Wabash Valley. Reports of Investigations No. 13. Illinois State Museum, Springfield.

Ken Duerksen 3D/Environmental 781 Neeb Road

Cincinnati, OH 45233

John F. Doershuk Office of the State Archaeologist The University of Iowa Iowa City, IA 52242

Copyright Office of the State Archaeologist Spring 1998

Provided by ProQuest Information and Learning Company. All rights Reserved