Midcontinental Journal of Archaeology 18(1):18-34

Institutional Database of Staff Publications Tennessee Division of Archaeology

Title: Excavations at 40LK1, A Mississippian Substructural Mound in the Reelfoot Basin, Lake County, Tennessee Year: 1993 Name(s): William L. Lawrence and Robert C. Mainfort, Jr. Source: Midcontinental Journal of Archaeology 18(1):18-34

Publisher Link: http://www.midwestarchaeology.org/mcja/journal-info

Division of Archaeology • 1216 Foster Ave. • Cole Bldg #3 • Nashville, TN 37243 Tel: 615-741-1588 • Fax: 615-741-7329 • www.tennessee.gov/environment/section/arch-archaeology EXCAVATIONS AT 40LK1, A MISSISSIPPIAN SUBSTRUCTURAL MOUND IN THE REELFOOT BASIN, LAKE COUNTY, TENNESSEE

William L. Lawrence Robert C. Mainfort, Jr.

ABSTRACT

40LKI consists of the remnant of a large Mississippian substructural mound and an associated habitation area. The uppermost intact summit of the mound was investigated, revealing the remains of a large wall-trench building. Several restorable ceramic vessels were associated with the structure. Radiocarbon assays suggest a construction date of approximately A.D. 1280 for the building.

Introduction

The Reelfoot Lake Basin is a large expanse of alluvial floodplain within the Mississippi River meander belt in northwest Tennessee and southwest Kentucky. Drainage is provided by Bayou du Chien, formerly a major tributary of the Obion River, which flows south across the Reelfoot Basin from its headwaters in southwest Kentucky. In recent times the Mississippi River captured Bayou du Chien at Hickman, Kentucky, creating an underfit, beheaded stream (Glenn 1933:6). Much of lower Bayou du Chien is now submerged beneath Reelfoot Lake. The lake itself occupies several meander scars (Fisk 1944; Russ 1979; Shelford 1963) and covers over 6,000 ha; Indian and Reelfoot creeks are major feeder streams. Flanking the eastern edge of the Basin are the steep loess bluffs that are characteristic of the Mississippi River valley. Prior to logging and clearing for agricultural pruposes, the lake margins supported a cypress-ash forest, with walnut also present. An oak-hickory forest occupied the slightly higher elevations surrounding the lake, while the upland bluffs were characterized by a tulip-oak climax forest (Shelford 1963). Today, the floodplain is planted primarily in row crops, but prior to the construction of levees in the early 1900s lower lying portions of the

Reelfoot Lake Basin were seasonally inundated almost every year and subject to silt deposition and scouring. Reelfoot Lake is located on the Mississippi Flyway and serves as a stopover for several hundred thousand waterfowl annually. The lake is also noted as a winter breeding ground for bald eagles. Both sport and commercial fishing have a long and productive history. White-tailed deer, raccoons, turkeys, and several species of aquatic mammals are common throughout the Basin. Although the archaeological importance of the Reelfoot Basin has long been recognized (Donaldson 1946; Funkhouser and Webb 1932; Thomas 1894), little systematic research was conducted in the region until fairly recently, with the exception of several limited contract projects (Dickson and Campbell 1979; McNerney and Nixon 1980; Schock 1986; Smith 1979). This point is underscored by the conspicuous absence of data from the Reelfoot Lake area in the syntheses of Morse and Morse (1983), Phillips et al. (1951), and Phillips (1970). In the mid-1980s, the Tennessee Division of Archaeology (TDOA) initiated a long-term program of surveys and limited excavations in the region (Mainfort 1989; Mainfort et al. 1986; Mainfort and Kreisa 1988; Moore and Mainfort 1988), and the University of Illinois conducted research in the Kentucky portion of the Reelfoot Basin (Kreisa 1988, 1990; Lewis 1986; Stout 1986). To date, over 140 archaeological sites have been recorded throughout the Basin (Mainfort 1989).

Archaeologoclal Investigations

The Haynes site (40LK1) consists of a Mississippian substructural mound and a surrounding habitation area located near the west bank of the northern portion of Reelfoot Lake (Fig. 1). The present channel of the Mississippi River lies approximately 4 km to the west. Situated on agriculturally productive Tiptonville silt loam (Brown et al. 1969), the relatively low elevation of the site (285' amsl) made it subject to seasonal flooding prior to the construction of levees along the Mississippi River. In 1946, R. C. Donaldson, a noted local historian and relic collector during the 1930s and 1940s, described the single extant mound: "a square mound stands out as a landmark, on which a barn has recently been built by the owner. This measures 150 feet on each side and is 11 or 12 feet tall. Portions of the top were once covered with burnt clay" (1946:78). Local residents generally confirm Donaldson's description, noting that the earthwork was formerly much larger, with steep sides and a flat top, and ear lier inspections by TDOA staff recorded the presence of large quantities of daub and charred wood on the surface. Presumably, one or more associated ramps were once present, but neither local traditions nor our excavations pro- 20 Lawrence and Mainfort vide evidence of such a feature. Based on the structural remains discussed below, the mound was oriented toward magnetic north. Unfortunately, the earthwork has been severely damaged by agricultural utilization. Mound fill has been borrowed on several occasions to fill nearby low-lying areas, which, coupled with continued plowing, has reduced the mound to a low, somewhat oval remnant measuring approximately 51 m north-south by 44 m east-west, with a height of approximately 1.25 m (Fig. 2). Notes recorded by Donaldson on catalog cards for his collection, now housed at Reelfoot Lake State Park, indicate that "bones of several skeletons (were) plowed up" some "25 feet north" of the large earthwork. The existence of this apparent cemetery cannot be confirmed at this time. Surface collections obtained from 40LK1 during a period of over 20 years are all of modest size and provide little evidence of an associated habitation area. This suggested that redeposited sediments from flooding had largely masked the occupation area surrounding the mound, so a 2 by 2 m test pit was placed west of the mound to determine if habitation deposits were present. Below a 25 to 30 cm layer of sand and silt alluvium (presumably deposited by floodwaters from the Mississippi River), an intact Mississippian midden approximately 30 cm thick was disclosed. No architectural features were encountered, but the artifact assemblage suggests contemporaneity with the nearby earthwork. Additional testing would probably demonstrate the presence of a fairly substantial habitation area. Since continued agricultural use of the site will eventually result in the total destruction of the mound, TDOA conducted limited salvage excavations in March 1990 for the purpose of recording the structure presumed to be present on the uppermost intact summit. Initial testing with a split-spoon sampler revealed a structural surface beneath a thick layer of daub at a depth of approximately 60 cm below surface near the center of the mound and immediately below the plowzone along the northern margin of the earthwork. A total of 41, 2 by 2 m units were completely or partially excavated to this summit, revealing the remains of a wall-trench building with interior dimensions of approximately 13.8 m by 12.5 m (Fig. 3). Complete exposure of the building was not possible due to time and financial constraints. Although the northwest corner of the building had been destroyed by plowing, the remaining three corners were identified by right angle turns in the wall trench; the corners were not open. Based on the locations of the three intact corners, the structure interior encompassed roughly 172 m2• Wall trench fill consisted of a 30 cm deep deposit of yellowish brown clay that sharply contrasted with the surrounding dark grayish brown clay matrix. The clay used to fill the wall trench is quite plastic when moist, but dries to a state of considerable rigidity, providing a firm footing for the upright MISSISSIPPIAN SUBSTRUCTURAL MOUND 21 posts. These properties were almost certainly recognized by the builders, and the clay apparently was brought to the site from some distance spe cifically for use in construction. The northern wall was defined by a wall trench 30 to 50 cm wide, in which a series of 28 carbonized, upright posts averaging 14.5 cm in diameter were located; some posts on the western end of this wall trench were de stroyed by plowing. Preservation of the extant posts was generally excellent, allowing the recognition and easy separation of annual growth rings in cross-section. Wood samples from these posts have been identified as ash (Fraxinus sp.) and elm (Ulmus sp.) (David Stahle, personal communication 1990). The outermost growth rings of one post in the northern wall trench yielded an age of 690 ± 60 B.P., which calibrates to A.D.1264 (1281) 1385 (TX-6858). Providing additional support for the upright posts, a number of logs averaging IO cm in diameter were placed horizontally in the wall trench. Similar Mississippian construction techniques have been recorded at the roughly contemporary Lilbourn and Snodgrass sites in Missouri (Cottier 1977:129; Price and Griffin 1979:31-34). A portion of the east wall, including a continuation of the wall trench and the in situ bases of eight posts, was also disclosed, but subsequent construction and/or destruction has destroyed most evidence, as discussed below. Limited excavations in the vicinity of the south wall suggest at least one rebuilding episode during the lifetime of the structure. The two exposed wall-trench corners are not properly aligned for a symmetrical building and a line of posts in N102/E98 align with the corner to the west, but not the one to the east. Several posts, daub, and roofing material in the southernmost units may represent the remains of an outer structure attached to the large building, but additional excavation would be necessary to adequately interpret these remains. Additional post holes (N = 17) were recorded within the excavated portion of the interior, but it was often difficult to determine the actual level of origin. Some probably represent interior support posts, while others may be intrusive from subsequent building episodes. Interior post holes were generally represented by circular holes in the compact floor, filled with unconsolidated daub rubble. The presence of daub in these holes, as well as the paucity of posts in some sections of the wall trench, suggests that many posts were removed for reuse prior to the destruction of the building. The structure floor was partially covered with daub from the collapsed walls and well-preserved, carbonized organic remains. Charred logs and poles lying horizontally on the floor probably represent collapsed roofing timbers. Quantities of grass thatch and cane matting were also recovered from various localities on the floor. All of the carbonized organic material appears to have been burned rapidly, but was extinguished prior total 22 Lawrence and Mainfort combustion. This suggests that the structure was allowed to burn until it collapsed, at which time the fire was extinguished by the addition of mound fill. Also supporting this interpretation is the observation that some mound fill immediately above the structure exhibits discoloration due to heat oxidation.

Features and Artifacts

Four features were associated with the excavated portion of the structure, while a fifth was intrusive from a later construction episode (Fig. 3). Feature l was an irregularly shaped basin approximately 2 m long and 20 to 60 cm wide lying along and within the eastern wall trench; the walls exhibited considerable exposure to heat. The feature was apparently excavated into a section of wall trench since a deposit of yellowish brown clay, identical to wall trench fill, was located on the structure floor immediately northwest of Feature l. Feature fill consisted primarily of daub rubble and ash, but numerous unidentifiable scraps of calcined bone, several pieces of mica, and a small, restorable Mississippi Plain, var. Mississippi jar (Fig. 4) were also recovered. Several sherds of the latter vessel were found on the structure floor, adjacent to Feature 1. Sherds from the feature fill were refired to a dark brown color, while those located outside the feature walls retained their original cream/buff color. It seems likely that this vessel was inten tionally broken and scattered during the use of Feature l. The jar has an orifice diameter of 10 cm and a maximum depth of 10.5 cm. An opposed pair of strap handles is attached to the everted rim at the lip. A small fillet was removed from the center of each strap near the point of attachment to the lip, and paste was modeled along the upper lateral margins, forming a pair of projections. Similar handles from the Chambers site in western Kentucky have been described as double castellations with a medial groove (Pollack and Railey 1987:78). The base of an in situ carbonized post was located at the north end of Feature l, within the wall trench and in contact with the fired wall of the feature. This was the only post located along the central portion of the eastern wall and it appears that prior to the preparation and use of Feature l, most of the eastern wall was removed, with a single post left in place to tenously support the roof. Since a fire could not have been built in the wall trench of the structure while the wall was still present, use of Feature 1 must have occurred just prior to the destruction of the building, and this event may represent this ignition of the structure. A curvilinear clay berm on the structure floor, Feature 4, was roughly 2 m long and 20 cm wide. The function is unclear. The absence of posts, postmolds, and matting suggest that the feature does not represent an MISSISSIPPIAN SUBSTRUCTURAL MOUND 23 interior partition, although it may have served to define an area within the structure. Partially superimposed on Feature 4, Feature 2 was an irregularly shaped basin lined with clay. Large quantities of daub, charred cane fragments, and grass thatch were included in the feature fill ; a fragmentary wooden vessel and most of a Mississippi Plain salt pan were located near the base. The wooden vessel was manufactured from a very large ash (Fraxinus sp.) bole, most likely green ash (Fraxinus pennsylvanica) or pumpkin ash (Fraxinus profunda) (Lee Newsom, personal communication), both of which commonly occur in the central Mississippi Valley bottomlands (Shelford 1963). With an orifice diameter of approximately 40-50 cm and a relatively shallow depth (Ann Cordell, personal communcation 1991), a pan or trough like vessel form is suggested. The orientation of the growth rings indicates that the vessel was carved from a longitudinal section of the bole, rather than a cross-section, i.e., the rim edge was located toward the center of the tree and the base was located near the bark or outer surface. This orientation has the advantage of preventing separation of the vessel along the annual growth rings, which could easily occur if the rings were exposed in cross-section. The piece of wood selected for carving most likely represents a section of waste wood, perhaps obtained as the byproduct of canoe manufacture. Severe leaching prevented restoration of the Mississippi Plain salt pan, but partial reconstruction of the rim suggests that this vessel was shallow, with a diameter or roughly 40 cm. No indications of fabric impressions are observable on the exterior. Feature 5 was a roughly circular fired-clay surface on the structure floor, covering an area approximately l.6 min diameter. There were no associated artifacts, but the feature probably served as a large hearth. Intrusive from the subsequent building stage, Feature 3 was a deep, straight-sided cylindrical pit measuring approximately 75 cm in diameter. The pit originated in what is now the plowzone and reached a depth of 55 cm. The walls exhibited considerable discoloration from exposure to heat, and a quantity of charcoal was found at the base. A large Mississippi Plain, var. Mississippi jar, measuring 75 cm in diameter and 31 cm deep rested on the charcoal deposit. The rim is straight, with a pair of opposed strap handles (Fig. 5). Interestingly, the diameter of the pit (Feature 3) was almost precisely large enough to accommodate this vessel. A 425 g sample of carbonized shelled corn was recovered from the interior of the jar and discoloration of the vessel interior indicates that corn occupied approximately one third of the total vessel volume. A hardwood charcoal sample obtained from beneath the large jar within Feature 3 returned a radiocarbon age of 24 Lawrence and Mainfort

720 ± 70 B.P. and a calibrated date of A.D. 1257 (1279) 1377 at one sigma (TX-6857). Several additional ceramic vessels were recovered from non-feature con texts. A shallow, partially restorable O'Byam Incised, var. Stewart bowl, was located within the matrix of the northern wall trench near the west end. Manufactured from a Bell Plain paste, this vessel exhibits a filleted rim and measures 21 cm in diameter and 6.5 cm deep. Decoration consists of line-filled triangles carelessly engraved on the interior (Fig. 6 and 7; Fig. 7 is an artistic reconstruction of the entire vessel as viewed from above). A large, amorphous piece of pumice exhibiting considerable use as an abrading tool was associated with this vessel. Two completely restorable jars, one a large Matthews Incised, var. Mat thews vessel, the other var. Manly, and a single large sherd from a second Manly vessel were recovered as a sherd concentration just above the building floor, where they apparently were intentionally deposited as whole vessels after the destruction of the building, prior to or during subsequent mound construction. Unlike the large jar associated with Feature 3, there was no indication that these vessels were intrusive. Surface treatment on the Matthews vessel consists of three parallel incised lines that form an undulating arcaded motif along the shoulder (Fig. 8). A pair of opposed strap handles is attached at the lip, and the rim is slightly inslanting. The orifice diameter measures 34 cm, while the maximum depth is 38 cm. The shoulder of the Manly jar is decorated with a single incised arcaded line and two parallel sets of punctations (Fig. 9). Appendages consist of a pair of opposed strap handles, each with a medial groove and double castellations, as well as an opposed pair of nodes or nonfunctional lug handles. The everted rim exhibits four raised, opposed rim tabs that cor respond to the appendages attached below the lip. Viewed from above, this arrangement creates a multifaceted appearance of four opposing quadrangles around the vessel orifice. This vessel has an orifice diameter of 22 cm and a maximum depth of 23 cm.

Conclusion

The 1990 excavations at 40LKI successfully located and defined a substantial portion of a large structure associated with the uppermost preserved mound summit. With the exception of a minimally investigated building exposed near the surface of Mound 2 at the Lilbourn site (Chapman and Evans 1977:83-84), this represents the only reported data on Mississippian mound summit architecture within the central Mississippi Valley since the advent of modern archaeology. MISSISSIPPIAN SUBSTRUCTURAL MOUND 25

The two remarkably consistent radiocarbon determinations on samples from two distinct proveniences suggest that the excavated structure at 40LK1 was constructed around A.O. 1280. It seems virtually certain that a minimum of two higher summits were totally destroyed prior to the investigations described here, so occupation of the site probably extended well into the fourteenth century A.O. No sites of demonstrated comparable age have been recorded on the west side of Reelfoot Lake (Mainfort 1989), a situation partially attributable to depositional conditions revealed by off-mound testing at 40LKI. Ceramics and a single radiocarbon date from the Sassafras Ridge site (15FU3) to the north suggest that the two sites were partially contem poraneous (cf. Lewis 1986). We believe that the assemblage of whole and partially restorable ceramic vessels from 40LK1 represents the best temporal and contextual control for a mid to late thirteenth century occupation in northwest Tennessee and the Mississippi River counties of Kentucky. The assemblage is clearly associated with two sequent episodes of mound construction and use during a relatively brief time period. In contrast, most radiocarbon determinations (and, hence, inferences about the chronological positions of various ceramic types) from roughly contemporary sites in western Kentucky are derived from mound fill contexts and long-term habitation areas in which associated ceramics often represent incidental inclusions in house basins (e.g., Lewis 1986; Wesler 1991). In addition to the recovery of important information about Mississippian public architecture, the data obtained from our limited investigations at 40LK1 should be of considerable value in resolving some of the nagging problems of Mississippian chronology and ceramics in the central Mississippi Valley alluded to in recent summaries by Lewis (1990) and the Morses (1990).

Acknowledgments

We particularly wish to thank Messrs. Jack Haynes and Jack Haynes, Jr., for permission to conduct excavations at 40LK l and for altering their spring planting schedule to accommodate our work. Paris Stripling produced fig ures l through 3, while Shawn Chapman prepared the ceramic vessel illus trations. Lee Newsom (Florida Museum of Natural History) graciously analyzed the wooden vessel from the site. David Stahle (University of Ar kansas) identified additional wood specimens. Mary Kwas reviewed a draft version of this paper and made a number of useful suggestions. An earlier version of this paper was presented at the 1990 Southeastern Archaeological Conference, Mobile, Alabama. Tennessee Division of Archaeology 26 Lawrence and Mainfort

oj FU19. "'

.FUlll

:/ ru1z•

LK6• L .. 13• LK• e LKI• ·

REELFOOT LAKE BASIN KENTUCKY AND TENNESSEE

• MAJOR ARCHAEOLOGICAL SITES .. .. KILOMETERS '======'===='==~=~- WILES

Fig. I. Major archaeological sites in the Reelfoot Lake Basin . MISSISSIPPIAN SUBSTRUCTURAL MOUND 27

HAYNES StTE C40LK I)

Fig. 2. Substructural mound, 40LK1. 28 Lawrence and Mainfort

HAYNES SITE (40lk I) PLAN VIEW OF EXCAVATION

UHEllCAVATED D FEATURE CIT] POST MOLD • POST • WALL TRENCH THATCH ~ D CANE MATTING

UNEXCAVATEO BURIAL ~• BURNED LOGS ~ SLOT TRENCH D

Fig. 3. Plan view of excavated structure at 40LKI. MISSISSIPPIAN SUBSTRUCTURAL MOUND 29

Fig. 4. Mississippi Plain jar (one-fourth actual size).

''.C!~rJt'{Gf j;'''T «f1w0 SW~,··. · . .

:. :.' ~... <·· ..

. ·.· ·.· ..

Fig. 5. Mississippi Plain jar from Feature 3 (one-tenth actual size). 30 Lawrence and Mainfort

Fig. 6. O'Byam Incised bowl (one-fourth actual size).

Fig. 7. Artistic reconstruction of O'Byam Incised bowl (one-fourth actual size). MISSISSIPPIAN SUBSTRUCTURAL MOUND 31

Fig. 8. Matthews Incised, var. Matthews jar (one-tenth actual size).

,, ...• ·.:·~?~:;;;:·•· ···

. ·... .-. : ·:,. ··· ·· ·/.:YJ~'

Fig. 9. Matthews Incised, var. Manly jar (one-fourth actual size). 32 Lawrence and Mainfort

References

Brown, William T., W. C. Jackson, G. L. Keathley, and L. C. Moore 1969 Soil Survey of Lake County, Tennessee. U.S. Department of Ag riculture, Soil Conservation Service, Washington, D.C. Chapman, Carl H., and David R. Evans 1977 Investigations at the Lilbourn site 1970-1971. In Investigation and comparison of two fortified Mississippian tradition archaeological sites in southeast Missouri: a preliminary compilation, by Carl H. Chapman et al, pp. 70-104. Missouri Archaeologist 38.

Cottier, John W. 1977 The 1972 excavations at the Lilbourn site. In Investigation and comparison of two fortified Mississippian tradition archaeological sites in southeast Missouri: a preliminary compilation, by Carl H. Chapman et al, pp. 123-1~4. Missouri Archaeologist 38. Dickson, D. Bruce, Jr., and L. Janice Campbell 1979 Reelfoot and Lake Isom National Wildlife Refuges: a cultural resources survey. New World Research Report of Investigations No. 20. Report submitted to Interagency Archaeological Services, Atlanta.

Donaldson, R. C. 1946 Mound Builders of the Reelfoot Lake Region. Tennessee Archae ologist 2(4): 77-81. Fisk, Harold N. 1944 Geological investigation of the alluvial valley of the lower Mississippi River. War Department, U.S. Army Corps of Engineers, Mississippi River Commission Publication 52. Vicksburg. Funkhouser, William D., and William S. Webb 1932 Archaeological Survey of Kentucky. University ofKentucky, Reports in Archaeology and Anthropology no. 2. University of Kentucky, Lexington.

Glenn, L. C. 1933 The geography and geology of Reelfoot Lake. Journal of the Tennessee Academy of Science 8(1): 3-12. Kreisa, Paul P. 1988 Second Order Communities in Western Kentucky. Western Kentucky Project Report No. 7, Department of Anthropology. University of Illinois, Urbana. 1990 Prehistoric Settlement Patterns in the Big Bottoms of Fulton County, Kentucky. Western Kentucky Project Report No. 8, Department of Anthropology. University of Illinois, Urbana. MISSISSIPPIAN SUBSTRUCTURAL MOUND 33

Lewis, R. Barry 1990 The late prehistory of the Ohio-Mississippian Rivers confluence region, Kentucky and Missouri. In Towns and Temples along the Mississippi, edited by D. H. Dye and C. A. Cox, pp. 38-58. University of Alabama, Tuscaloosa.

Lewis, R. Barry (editor) 1986 Mississippian Towns of the Western Kentucky Border: the Adams, Wickliffe, and Sassafras Ridge Sites. Kentucky Heritage Council, Frankfort.

McNerney, Michael J., and Joseph M. Nixon 1980 An archaeological and historical resources survey of 21 Mississippi River levee berm items in the states of Kentucky and Tennessee: Item 1. Report submitted to the U.S. Army Corps of Engineers, Memphis.

Mainfort, Robert C., Jr. 1989 Cultural history of the Reelfoot Lake Basin, Kentucky and Ten nessee. Paper presented to the Southern Anthropological Society, Memphis, Tennessee.

Mainfort, Robert C., Thomas H. Bianchi, and Richard W. Yarnell 1986 Emergent Mississippian occupation in the Reelfoot Lake Area. Paper presented to the Southeastern Archaeological Conference, Nashville, Tennessee.

Mainfort, Robert C., and Paul P. Kreisa 1988 Late prehistoric research in the Reelfoot Lake Basin, Kentucky and Tennessee. Paper presented to the Southeastern Archaeological Con ference, New Orleans, Louisiana.

Moore, Michael C., and Robert C. Mainfort, Jr. 1988 Test excavations at 40LK3, Reelfoot Lake, Tennessee. Paper pre sented to the Southeastern Archaeological Conference, New Orleans, Louisiana.

Morse, Dan E, and Phyllis A. Morse 1990 Emergent Mississippian in the Central Mississippi Valley. In The Mississippian Emergence, edited by B. D. Smith, pp. 153-173. Smithsonian Institution, Washington, D.C. 1983 Archaeology of the Central Mississippi Valley. Academic Press, New York.

Phillips, Philip 1970 Archaeological Survey in the Lower Yazoo Basin, Mississippi: 1949- 1955. Papers of the Peabody Museum of Archaeology and Eth nology, vol. 60, pts. l and 2. Cambridge. 34 Lawrence and Mainfort

Phillips, Philip, James A. Ford, and James B. Griffin 1951 Archaeological Survey in the Lower Mississippian Valley, 1940-1947. Papers of the Peabody Museum of American Archaeology and Ethnology, vol. 25. Cambridge. Pollack, David, and Jimmy A. Railey 1987 Chambers (15MLJ09): An Updated Mississippian Village in Western Kentucky. Kentucky Heritage Council, Frankfort. Price, James E., and James B. Griffin 1979 The Snodgrass site of the Powers phase of southeast Missouri. Anthropological Papers No. 66, Museum of Anthropology. Uni versity of Michigan, Ann Arbor. Russ, David P. 1979 Late Holocene faulting and earthquake recurrence in the Reelfoot Lake area, northwestern Tennessee. Geological Society of America Bulletin, pt. l, vol. 90, No. 11, pp. 1013-1018. Schock, Jack M. 1986 An Archaeological Excavation of Mounds l, 2, and 4 at 40-0B-6 (The Effigy Rabbit Site) in Obion County, Tunnessee. Draft report submitted to the Tennessee Division of Archaeology, Nashville. Shelford, Victor E. 1963 The Ecology ofNorth America. University of Illinois Press, Urbana. Smith, Gerald P. 1979 Archaeological Surveys in the Obion-Forked Deer and Reeljoot Indian Creek Drainages: 1966 through Early 1975. Anthropological Research Center, Occasional Papers No. 9. Memphis State Univer sity, Memphis. Stout, Charles B. 1986 The Sassafras Ridge Site. In Mississippian Towns of the Western Kentucky Border: The Adams, Wickliffe, and Sassafras Ridge Sites, edited by R. Barry Lewis, pp. 130-151. Kentucky Heritage Council, Frankfort. Thomas, Cyrus 1894 Report on the Mound Explorations of the Bureau of Ethnology. Smithsonian Institution, Bureau of American Ethnology, 12th An nual Report. Washington, D.C. Wesler, Kit W. 1991 Ceramics, chronology, and horizon markers at Wickliffe Mounds. American Antiquity 56(2): 278-290.