Sherman Mound and Village

Sherman Mound and VillageAddress Restricted - Mississippi
Address Restricted
Listed in National Register of Historic Places on 02/01/2018

View PDF

The Sherman Mound site (3MS16) possesses statewide significance under Criterion A (Events and Broad Patterns of Events), as it typifies a fortified Middle Mississippian mound center that is part of a broader pattern associated with the establishment of chiefdom-level societies and population nucleation during ca. a.d. 1200–1600; and it possesses statewide significance under Criterion D (Information Potential), in that it has yielded information important to the Prehistory of Arkansas, and it retains the potential to produce additional knowledge relative to a number of broad research topics including: regional cultural chronology; settlement patterns; chiefdom level societies; mound-and-plaza architecture and ceremonialism; mortuary practices; exchange relations; and subsistence patterns.


Narrative Statement of Significance

Site Chronology

As a result of the PTC investigations, the dating and sequence of the Prehistoric occupation of the Sherman Mound site (3MS16) has been greatly refined. Comparative analysis of the recovered material culture, in particular ceramics, reveals the site was most likely colonized during the Late Woodland period, and re-occupied during the Middle Mississippian period during which it developed into a 44-ac. fortified town. The AMS dates derived from the two features on the southwestern margin of the Sherman Mound place the site’s Middle Mississippian occupation at cal. a.d. 1280–1390, and, equally important, reveal the presence of a previously unrecognized cal. a.d. 1650 component. During the Historic Tenant period (a.d. 1875–1950) the site was reoccupied and several residences were distributed along roads within the former Native American town.

Late Woodland Occupation

The initial occupation of the site took place during the Woodland period and is characterized by the presence of clay-tempered ceramics. Previously Lantham (2001a) had reported a few sand-tempered sherds from the site, these would presumably be Late Woodland Barnes series. However, as the Sherman Mound is outside the typical distribution of sand-tempered Barnes ceramics (Morse and Morse 1983:Figure 9.1) and since no sand-tempered sherd was recovered during this study, a Barnes component at the site is considered doubtful.


Clay-tempered sherds represent 27.4 percent of the Sherman pottery assemblage and, given this frequency, the Woodland component at Sherman is considered to be considerably less intensive than the subsequent Middle Mississippian occupation. It is difficult to precisely date the Woodland occupation because the majority of the clay-tempered sherds (61/63, or 96.8 percent) are plain (Baytown Plain; see Table 2), and we have no lithic diagnostic, nor any dated Woodland feature. The two decorated clay-tempered sherds were not typed, but included an incised and notched rim and a red-filmed specimen. Overall, the impression of the Sherman clay-tempered assemblage is that it is derived from a Late Woodland Baytown phase ca. a.d. 400–700 occupation, although the red-filmed clay-tempered sherd is suggestive of the transition from Baytown to Mississippian, which Morse and Morse (1983:190) date to “around a.d. 700–800.”


Lafferty’s (2008) analysis of over 150 radiocarbon samples from dated contexts suggests that in the Baytown South area, which includes the Sherman Mound locality, shell-tempered ceramics first appear in ceremonial contexts ca. a.d. 800. Between a.d. 800 and a.d. 1100 there was a period of experimentation with mixed tempering, and after a.d. 1100 straight shell tempering was the dominant temper.

Middle Mississippian Occupation

Shell-tempered ceramics form the majority of the Sherman ceramic assemblage (72.6 percent) and attest to an intensive Mississippian occupation of the site. The bulk of the shell-tempered sherds are a coarse plainware typed as Mississippi Plain, var. unspecified (143/167, or
85.6 percent). This is a fairly typical pattern for Middle Mississippian assemblages in the central Mississippi Valley; for example the large (>20,000) sherd assemblages at the Middle Mississippian Moon and Kochtitzky Ditch sites were each over 99 percent Mississippi Plan (Benn 1998:249; Buchner et al. 2003:73). As a result, as Morse and Morse (1983:247) lament, this “frustrates archaeologists’ attempts to seriate pottery collections” from this period.


Fortunately, the relatively small Sherman Mound shell-tempered assemblage contains a dozen decorated sherds (7.2 percent) and 12 Bell Plain sherds, which allow for some refinement of the site’s chronology. The key ceramic diagnostics for a Middle Mississippian occupation at Sherman were recovered in low to trace frequencies, and include: O’Byam Incised, var. O’Byam (n=3, or 1.8 percent of shell-tempered); Matthews Incised, var. Manly (n=1, or 0.6 percent of shell-tempered); and Mound Place Incised, var. Mound Place (n=1, or 0.6 percent of shell-tempered).


Phillips (1970:144) characterized O’Byam Incised, var. O’Byam as a Late Mississippian type; however, most modern researchers consider it a solid Middle Mississippian diagnostic. For example, Morse and Morse (1983:245-247) deem it an “earlier” Middle Mississippian type, principally due to its recovery from the Middle Mississippian deposit at the base of the Hazel site, a stratum that is proposed to date to a.d. 1150–1250. An impressive assemblage—15 plates minimum—of O’Byam Incised, var. unspecified was recovered from the Middle Mississippian Moon Village (Benn 1992:101). At the Kochtitzky Ditch site, Buchner et al. (2003:83) reported two O’Byam Incised plate rims from well-dated contexts, and suggest an “early to mid-thirteenth century use” there. Moon Village and Kochtitzky Ditch exhibit complementary suites of radiocarbon dates that suggest use of this variety continued after a.d. 1200. Earlier, Morse and Morse (1983) suggested that the use of var. O’Byam declined after a.d. 1200.


Elsewhere in Mississippi County, Middle Mississippian deposits at the Knappenberger site (3MS53) produced an O’Byam Incised sherd (Klinger 1974:12j), and Morse (1998:5) briefly mentions another probable O’Byam Incised plate rim with a mace motif that was dug by an amateur there. At Cherry Valley, Perino (1967:Figure 41 bottom) recovered an obvious O’Byam Incised “deep plate” that was mistyped as Matthews Incised. To the north, in the Cairo Lowland, Perttula (1998:190) and others also consider it a Middle Mississippian type. More generally, McNutt (1996:239; 3CG1) noted that during the “Mature Mississippian period (a.d. 1250–1350),” ceramics become “increasingly complex, with the O’Byam Incised plates giving way to the various varieties of Matthews incised jars.”


The CSC produced one Matthews Incised, var. Manly sherd from a habitation area north of the Sherman Mound. Phillips (1970:127-128) distinguished three varieties of Matthews Incised in collections from the Mississippi Valley: var. Matthews; var. Beckwith; and var. Manly. Morse and Morse (1990:158) suggest that vars. Matthews and Manly represent a “Matthews Horizon” and are “marker ceramic types for the a.d. 1200–1400 late middle period Mississippian for most if not all of the Central Valley.” The radiocarbon dates substantiating this chronology (see Morse and Morse 1990:Table 14) are associated with Powers phase towns in southeastern Missouri (Price and Griffin 1979) and the Lawhorn site (Moselage 1962) in northeastern Arkansas. More locally, the Kochtitzky Ditch site (3MS599) produced 54 Matthews Incised var. Matthews sherds, and associated dated contexts suggest a use range of cal. a.d. 1170–1282 (Buchner et al. 2003:77-79). Matthews Incised is thought to have been replaced by the types Barton Incised and Parkin Punctated around a.d. 1400 in the central Mississippi Valley.


Jars decorated with arcaded drapes—similar to the Sherman Matthews Incised specimen—are part of a pan-southeastern Middle Mississippian horizon that originated in the Cairo Lowlands and radiated outward. For example in middle Tennessee, Smith (1992:79) cites the Morses (1990) and then states that “...the prevalence of Matthews Incised as the primary decorative mode for the Thruston phase should place it within the a.d. 1200–1400 range, particularly when taken in conjunction with the complete absence of Barton Incised and Parkin Punctated in the region.” In Thruston phase assemblages Matthews Incised vessels “often occur in conjunction with highly stylized zoomorphic and/or anthropomorphic vessels...” (Smith 1992:79). To the southeast, at Moundville, two varieties of Matthews Incised are reported, Beckwith and Manly, and these are dated by grave lot serration to the late Moundville II or early Moundville III period (ca. a.d. 1350–1550; Steponaitis 1983:333). Interestingly, Steponaitis (1983:333) notes that one var. Manly vessel “ very similar to examples illustrated from northeast Arkansas (e.g., Perino 1966:Figure 38 [bottom]).” A related type, Moundville Incised, var. Snow’s Bend, is described as a decorative treatment found only on jars (Steponaitis 1983:57-58, 325-326). The chronological position of var. Snow’s Bend is Moundville I and II, with the variety reaching its greatest popularity in the early portion of period II (a.d. 1250–1400).


A single Mound Place Incised, var. Mound Place sherd was recovered from a residential area north of the Sherman Mound. Phillips et al. (1951:147-148) originally defined Mound Place Incised, and subsequently var. Mound Place was established by Phillips (1970:135). This variety has a wide distribution, minimally from Vicksburg (Phillips 1970) up to the Cairo Lowlands (Morse and Morse 1983:263) and the western lowlands of Missouri (Martin et al. 1993), and in northwestern Tennessee (Garland 1992:68; Mainfort 1996:91-92). Morse and Morse (1983:267) consider the Mound Place type to have first appeared during the Middle Mississippian period. This was demonstrated at the Kochtitzky Ditch site, as Mound Place Incised, var. Mound Place was recovered from the same large, dated trash pit (F-154; cal.
a.d. 1170–1282) that also yielded O’Byam Incised, var. O’Byam and Matthews Incised var. Matthews specimens (Buchner et al. 2003:81).


Importantly, the Middle Mississippian occupation of Sherman is nailed down by the two AMS dates from F-1 on the southwestern flank of the main mound. The obtained cal. intercept results are tightly clustered, a.d. 1285 and a.d. 1295, and allow for the acceptance of the results with a high degree of confidence. Overall, the F-1 dates indicate the Middle Mississippian occupation at Sherman spans a date range of cal. a.d. 1280–1390.


The recognition of a Middle Mississippian component at Sherman is not surprising, as it confirms earlier speculations regarding the chronological placement of the site. For example the LMS initially considered the Sherman Mound a possible Middle Mississippian component (10-Q-2 site card), as did Morse and Morse (1983). Latham (2001a:3-12) interpreted the site as a Middle Mississippian mound-and-plaza complex. Additionally, Prentice (2000:302), who conducted no research at the Sherman Mound, classified it as a single component Middle Mississippian site, with a temporal confidence of “moderate” in a regional study of Native American architecture.


A few comments on the duration of the Middle Mississippian occupation at Sherman can be offered. Two factors appear to suggest a relatively brief occupation: (1) most of the structures identified via geophysical imagery show little overlap; and (2) the low or sparse surface density of artifacts. In contrast, the large size of the main mound can be interpreted as resulting from a long duration of use (see Mound Volume section below).

Protohistoric Evidence

The only evidence for a Protohistoric component at the site is the F-2 AMS result with a cal. intercept of a.d. 1650. This date places F-2 in the mid-seventeenth century, around the end of the Protohistoric period, during the French Colonial era. F-2 is a shallow, basin-shaped pit that is interpreted as an intrusive burial on the southwestern flank of the Sherman Mound. It lacks any substantial diagnostic material culture, as it yielded a small and unimpressive assemblage (three residual ceramics, a small retouched piece, a small biface fragment, and two pieces of debitage, plus some material enumerated by mass only).


Sites such as Campbell and Nodena, which were occupied after initial European contact
(ca. a.d. 1541), are considered Armorel phase components (S. Williams 1980). However, if Sherman was occupied ca. 1650, then its occupation is probably more akin to that reported at the Grigsby site (3RA262) near Pocahontas, or Wallace Bottom site (3AR179) near Arkansas Post (House 2013). Morse and Morse (1988, 1992) consider Grigsby to date to ca. 1700 ± 30 and interpret it as the Michigamea village indicated on Marquette’s 1673 map. European artifacts associated with the seventeenth-century component at the Grigsby site include glass necklace beads, brass/copper tinkling cones and points, and a native-made glass pendant. Native diagnostics associated with this component at Grigsby include the Kaskaskia variant arrowpoints and Mississippi Plain ceramics with notched-filleted rims similar to Danner series pottery in the Illinois Country (Buchner et al. 2016).


More generally, the three AMS dates suggest that either two components are present at Sherman, or the site was in continuous use from the a.d. 1200s to the a.d. 1600s. However, given the general absence of Late Mississippian Nodena phase diagnostics at Sherman, particularly the lack of Nodena points and ceramics such as Barton Incised, Parkin Punctated, and Walls Engraved, it is considered most likely that the town was abandoned ca. 1400, and then reused
ca. 1650 by a yet to be determined Native American group.

Tenant Period Occupation

In 1899, the Deckerville, Osceola, & Northern Railroad was built across the western portion of Sans Souci Plantation (Edrington 1962:38). The railroad passed along the western flank of the Sherman Mound, and construction of it impacted the mound to an undetermined degree. Importantly, the railroad opened up the western portion of Sans Souci Plantation for development.


During the Tenant period a number of dwellings existed along the roads within the former Native American town, which was otherwise placed under cultivation. A 1914 map and a 1949 air photograph illustrate the locations of various Tenant period dwellings in and near the former town. During the geophysical survey strong anomalies were detected at these locations, indicating the soil contains abundant metal. The 1949 map shows a row of dwellings extending southwesterly along the railroad, beyond the palisade wall, all the way to the community of Grider. During the CSC a heavy Historic period surface scatter was documented in this area. This Historic scatter is so dense that it is likely masking a Middle Mississippian residential area on a rise immediately southwest of the highway and railroad crossing.


The Historic artifact sample (n=135) recovered during the CSC conforms to the well-documented Tenant period pattern at rural domestic sites in eastern Arkansas. Kitchen Group artifacts dominate the recovery (85.2 percent) and the other functional groups (Personal, Architecture, Activity, and Clothing) are represented in low frequencies (2–5 percent). Diagnostics confirm an intensive ca. 1890–1950 occupation.

Site Plan

Woodland Village

During the Late Woodland Baytown period the site plan for 3MS16 is interpreted as a village with multiple structures arranged in an organic (semi-random) pattern. Excluding two outliers, clay-tempered ceramics are distributed across a roughly 250-x-180-m oval portion (8.9 ac.) of Site 3MS16, centered on the Sherman Mound. There are two apparent concentrations of Late Woodland Baytown period ceramics within the overall distribution, one on the residential ridge south of the mound, and one within a residential area to the north of the mound. Both of these concentrations were likely the locations of structures and/or habitation areas, and each covers a roughly 50-m diameter area (0.48 ac.). We speculate that the Late Woodland population also utilized the Sherman Mound location, because it lies between the two Late Woodland artifact concentrations, thus a Late Woodland Baytown period deposit and/or features may exist under the mound.


Other important Baytown phase sites in the region include Hyneman (3PO53) and DeRossitt (3SF49). The Baytown component at Hyneman is characterized by abundant Baytown Plain, with vessel forms including small bowls and jars with flat bases, and few decorated sherds (Morse and Morse 1983:193). Two radiocarbon dates are reported at Hyneman: a.d. 642 ± 138 and 761 ± 151. Highway salvage excavations at the DeRossitt site revealed four separate Baytown phase loci within the site, and exposed 164 pits and >500 postholes (Spears 1978). Most of the pits were small (between 40 and 400 liters), and interpreted as the by-product of “seasonal nomadic” occupations. The >5,300 sherds from DeRossitt were 64 percent Baytown Plain and nearly
36 percent Mulberry Creek Cord Marked, with just a trace of other decorated types.


While the Morses (1983:184) suggest that most Baytown period sites “were mainly single structure sites,” it is clear that the Late Woodland occupation at Sherman represents more than a single habitation locus. The pattern at Sherman is more akin that that observed at the multi-component Brougham Lake site (3CT98), located along a minor oxbow lake in the lower
St. Francis basin (Klinger et al. 1983). The Baytown component at Brougham Lake extended across 1.21 ac. and was characterized by four concentrations of Baytown ceramics (Klinger et al. 1983:418). The site produced 1,087 Baytown sherds, principally Baytown Plain with minor frequencies of decorated types (Klinger et al. 1983:410). The Baytown component at Brougham Lake was suggested to date a.d. 500–900, and one radiocarbon date of a.d. 780 ± 80 (uncorrected) was obtained on a Baytown pit (Klinger et al. 1983:410).


Importantly, stripping revealed at least 148 Baytown features, including evidence for seven structures at Brougham Lake. The structures were circular to oval, single-poled, and ranged in size from 12.6 m2 to 26.9 m2 with a mean floor area of 19.96 m2 (Klinger et al. 1983:250). Klinger et al. (1983:252) argue that no more than three or four of the Baytown structures at Brougham Lake were contemporary, and interpreted the site as a farmstead. The identification of the circular house patterns at Brougham Lake is significant because it is indicative of a greater investment of time and energy in facilities, and a greater duration of occupation, than had been previously assumed for Baytown populations.


Site 3CS137 is another tested site with a strong Baytown period component that bares comparison to Sherman. It is located on a low terrace of the St. Francis River near Wittsburg, and is a large (>22 ha) multi-component site that contained three conical mounds (Childress et al. 1995). Site 3CS137 is suggested to date “predominantly to the Baytown period (ca. a.d. 300–700)” based on the heavy recovery of Baytown Plain ceramics and low frequencies of decorated clay-tempered types (Childress et al. 1995:100). The lack of exotic raw materials and ceramics at Site 3CS137 (and at other Baytown period sites), is considered evidence of a low ebb in exchange relations and suggestive of an isolated localized population.


During testing at Site 3CS137, a CSC was recovered, and four 1-x-1-m units were excavated, including one in Mound 2, which may be the first conical mortuary mound tested in the lower
St. Francis area. Phillips (1970:904) had previously noted a consistent co-association of Baytown components and small conical mounds (Childress et al. 1995:Table 4.11) for such sites in the Site 3CS137 vicinity. We mention this because the Sherman Mound could have had its origins as a conical Baytown mortuary mound. The Delpro Mound (3MS635), a low conical mound located southwest of Wilson, may also represent another such feature. Phillips (1970:904) advocates that conical mound construction continued from the Marksville through the Baytown period, although he warns against the “hazards of associating unexcavated mounds with surface assemblages.” Childress et al. (1995:112) argue that for a millennia during the Woodland there was “relative stability in broad patterns of settlement and subsistence” that began to be replaced by Mississippian traits ca. 1000.

Mississippian Town

During the Middle Mississippian period the abandoned Late Woodland village was reoccupied, and developed into a large (17.8 ha) fortified Middle Mississippian town. As noted above, Sherman Town contains all the basic architectural design elements of a Mississippian town (Lewis et al. 1998).


The most prominent remaining element of the Mississippian town is, of course, the Sherman Mound, which is the primary mound at the site. The other two mounds at the site, Mounds B and C, located to the east of the Sherman Mound, are plowed down and not readily apparent to the naked eye today, although in 1945 Dr. Hampson apparently observed pothunters at a “small mound” east of the Sherman Mound.


The basal horizontal dimensions of the Sherman Mound, or Mound A, are 101.66-x-40.70 m, and its maximum height is 6.2 m. Excluding the ramp or projection on the southwestern side, the basal dimensions of the Sherman Mound are 72.40-x-40.70 m.


The 1881 woodcut drawing of the Sherman Mound had a strong influence on later archaeologist’s perception of the mound’s shape as three-terraced. Several twentieth-century sketch maps replicate the three-terrace shape in some form. Additionally, in a prehistoric architecture volume, this view is exaggerated even further, as the Sherman Mound is re-drawn, and described as having a “notable architectural quality…its combination of multiple terraces and curves within a disciplined axial symmetry…invite movement to the highest level, a sense of progression that is an arrangement often found in prehistoric American architecture” (Morgan 1980:69).


In contrast to the three-tiered vision, the topographic maps, DEM, and slope models prepared during this study suggest an alternative view of the Sherman Mound’s configuration: it is a rectangular platform mound with some type of ramp or projection on the southwestern side. This may be due in part to Historic disturbances to the mound since 1881, including the late 1890s railroad construction, road construction, and the use of a portion of the mound as yard fill. However, overall the topographic maps, DEM, and slope model maps reveal a robust, rectangular platform mound with steep slopes on three sides: the northwest; northeast; and southeast.


Earlier, Phillips (LMS 10-Q-2) had noted the eastern “end looks as if it may have been rectangular,” and today this end of the mound remains the most impressive portion of the mound. In contrast, the ramp or projection on the southwestern side of the mound is only about half as long as the platform mound section of the earthwork, and is much less steeply sloped. Both Evans’ 1881 drawing and Morgan’s (1980) rendering appear to greatly exaggerate the horizontal dimensions of the lower two terraces that compose the ramp or projection. More generally, Lewis and Stout (1998:158) characterize ramps, aprons, and terraced mound sections as “elaborations” that tend to be found at larger and more significant mounds, with Cahokia Monks Mound, Wickliffe Mound A, and Adams Mound A cited as examples.


The summit of the Sherman Mound is thought to have served as the location of an elite residence (i.e., chief’s house) and/or temple. The size of the relatively level area on the mound summit (43.6-x-15.4 m) would accommodate a relatively large structure, or possibly multiple structures. That a Mississippian structure stood here is confirmed by Evans’ excavations that produced daub, which he referred to as “fluted pieces of burned clay” (i.e., cane impressions) that were “supposed to be brick,” in a buried stratum over the entire upper terrace (Thomas 1985[1894]:223). Additionally, Dr. Hampson described of a “layer of burned clay, probably the floor of a temple” atop the mound.


Mound B is located 240 m east of the Sherman Mound and associated with a rectangular geophysical anomaly (see Figure 2), as well as a slight topographic rise (40–45 cm; see Figure 8). It is interpreted as a low, plowed down rectangular mound, with basal measurements of 24.6-x-17.7 m, and a long axis aligned to 60°–240°. Anomalies that are interpreted as Historic disturbances are found within Mound B, and it is possible that either of these disturbances represent the pothunters’ burial excavations within a “small mound” reported by Dr. Hampson, although the distance east of the Sherman Mound does not match his (100 yards). As such, Mound B is tentatively interpreted as a burial mound, or alternatively it could be the location of some other type mortuary facility, such as the charnel house identified at the Early Mississippian Priestly site (3PO490; Benn 1990:452-453). Interestingly, one red-painted, shell-tempered sherd was recovered from Mound B during the CSC. This is the only decorated Mississippian sherd recovered outside of the high-density residential areas north and south of the Sherman Mound, thus this location is unique.


Mound C is located roughly 65 m southwest of Mound B, and is a similar rectangular geophysical anomaly. The 5-cm contour map shows a slight rise here (15–20 cm), which strengthens the interpretation of this anomaly as a plowed down mound. Mound C measures 29.0-x-22.2 m, and its long axis is aligned to 66°–246°.


Finally, regarding potential celestial alignments of the mounds at Site 3MS16, the following is offered. Note that the Sherman Mound and the Mounds B-C pair both exhibit a general northeasterly to southwesterly alignment. The Sherman Mound is aligned on an axis of 32.3°–212.3°. On a mid-December afternoon, it was observed that the view southwest from atop the Sherman Mound is near a Winter Solstice sunset alignment of 242° (Sherrod and Rolingson 1987:28). Thus the Sherman Mound’s long axis may be aligned to the Winter Solstice Sunset—Summer Solstice Sunrise line. The same may be true for the Mounds B-C pair, thus the entire site may have been planned on this alignment. Additionally from Mound C, the Sherman Mound may be in a Summer Solstice Sunset position (298°; Sherrod and Rolingson 1987:28). As the Middle Mississippians were an agricultural people, these celestial alignments would have been an important way to measure time and plan seasonal farming activities.


The identification of an essentially complete palisade wall around the town via gradiometer survey is one of the most significant project results. As previously noted, it encloses 44 ac. with the Sherman Mound roughly near its center, and it is 1,641 m in length. Features interpreted as bastions are present along the palisade; they are most distinctive in the northwestern portion of the palisade where they are regularly spaced at roughly 40-m intervals. Overall, along the entire palisade, the bastion spacing ranges 31.6–41.7 m, and the average spacing is 37.5 m. Given the length of the palisade perimeter and the average spacing, about 43 or 44 bastions were likely present. Ground truthing a section of the palisade should be given attention during any future research at the site.


Lockhart et al. (2011) and Morrow et al. (2013) describe a similar large-scale, site-wide gradiometer survey at Old Town Ridge (3CG41), a 7-ha, fortified, Middle Mississippian town located 45 km northwest of Sherman in the St. Francis basin. This resulted in the identification of multiple geophysical anomalies similar to those identified at Sherman, including evidence for a partly burned palisade on the inside of a fortification ditch. The palisade was interpreted as constructed of large posts spaced at 20-m intervals, and smaller posts at 3-m intervals (Lockhart 2011:58-59). A possible entrance was also identified on the eastern wall of the palisade. It is unclear if bastions are present on the Old Town Ridge palisade, but they should be expected. Lockhart et al.’s (2011:Figures 6–8 and 11) and Morrow et al.’s (2013:14 and 15) gradiometer images—at least as published in journals—are much grainer and more “streaky” than Walker’s images. This likely resulted from the use of 1-m transect intervals, as opposed to the tighter 50-cm interval employed at Sherman, and possibly from walking the geophysical array too fast.


It should be noted that geophysical surveys are not always successful at locating palisade features. For example, at the Warren Wilson site (31BN29), a fortified Mississippian Pisgah phase (a.d. 1250–1450) village in the Appalachian summit that is contemporary with Sherman, no evidence for a palisade was found during a recent geophysical survey (Buchner et al. 2016). This site was extensively excavated during 1966–1974, and a complex feature pattern was exposed, including a lengthy section of a palisade that extended into the later geophysical survey area (Dickens 1976; Keel 1976). The failure of the geophysical study to identify the expected palisade wall section is related to intensive plowing and farming of this portion of the Warren Wilson site. Given this, the fact that nearly the entire palisade wall line is recognizable at Sherman is remarkable, considering that most of the site has been under cultivation for over a century.


Prior to the application of geophysics to archaeology, palisades were traditionally identified as features within excavations. One of the most impressive excavated examples was identified at the Moon Site (3PO488), a fortified Middle Mississippian village located 49 km east of the Sherman Mound in the St. Francis basin (Benn 1998). Benn (1998:238-241) almost missed discovering the palisade at Moon because the post stains were difficult discern visually in the midden and due to the general complexity of the overall feature pattern within the completely excavated site. At Moon two palisades were found, one enclosed each of the village’s two building stages. The outer palisade was composed of single posts 25–30 cm in diameter, spaced 8–20 cm apart within a trench. The inner palisade enclosed an earlier village, and was composed of single posts 20–30 cm in diameter, spaced more than 30 cm apart. Bastions are not mentioned on the Moon palisades (Benn 1998).


Both Moon palisades exhibited entryways consisting of “narrow passageways placed parallel to and on the outer side of the palisade wall” (Benn 1998:241). Lafferty (1973:157) reports that type of entranceway was “common in the Southeast during the Mississippi period and was used during the early historic period on the Atlantic coast.” Given the construction style of the Moon entranceways, it is possible to reinterpret—as entrances—two small “structures” (Structures 30 and 38) at Sherman that are located adjacent to (but outside of) the eastern side of the palisade.


Palisades also have been archaeologically identified at two Late Mississippian sites in northeastern Arkansas: Parkin (3CS29); and Neely’s Ferry (3CS24; Mitchem 2010). In contrast, the 6.3-ha Late Mississippian Upper Nodena Site (3MS4) apparently lacks a palisade or ditch (Mainfort et al. 2007:120).


The construction of the palisade around Sherman town would have required a considerable amount of labor and raw materials, principally wooden posts. Assuming a similar construction approach as at the Moon village palisades—a post roughly every 20–30 cm—and given the perimeter length (1,641 m), then it is estimated that between 5,470 and 8,205 posts would have been needed at Sherman, plus additional posts would be required for the bastions. For comparison, Demel and Hall (1998:204) indicate that the first palisade around the central precinct at Cahokia, which enclosed 82 ha, required around 20,000 posts.

Residential Areas

Four discrete clusters of structures were identified to the east of the Sherman Mound that are interpreted as residential areas. All are located immediately adjacent to the palisade wall. Other anomalies, interpreted as pits and posts, are found in halos around the structures within these clusters. This suggests intensive domestic occupation of these areas, which can be considered neighborhoods within the town. It is typical of Middle Mississippian mound-and-plaza complexes to be surrounded by habitation areas and midden deposits (Lewis 1996:68).

Two of the habitation areas exhibit houses in rough rows that hint at possible residential planning similar to that found at the Moon village (Benn 1998:254) and in Powers phase villages in southeastern Missouri (O’Brien 2002; Price 1978): the Southern Cluster and the Northern Cluster. Elsewhere within the Sherman Mound town, the distribution of structures is more random.


The Southern Cluster of 11 structures covers a roughly 100-x-30-m area, and is associated with a high-density of surface artifacts and a topographic rise. The structures in the Southern Cluster exhibit the most conformity in terms of floor size (75.26 ± 25.12) and are suggestive of standardized construction. This is the second closest habitation area to the Sherman Mound, and it produced a diverse ceramic assemblage that can be interpreted as evidence of elite or high status households due to the higher frequency of decorated pottery types and the fineware Bell Plain.


The eight structures in the Northern Cluster exhibit the largest average floor size (163.48 m2) and were distributed over a roughly 125-x-50-m area parallel to the palisade. No artifact was recovered during the CSC from this area, which is relatively flat and featureless.


The Southeastern Cluster of 20 structures is associated with a moderate-density of surface artifacts and another topographic rise that overlooks the Sandy Bayou floodplain. The structures here are larger on average (105.46 ± 85.16) than those across a swale to the west in the Southern Cluster. This is largely due to the presence of two large structures (Structures 0 and 1); most of the other structures in Southeastern Cluster are smaller and similar in size to those in the Southern Cluster. The center of the roughly triangular Southeastern Cluster is relatively open (i.e., lacks structures) and may represent an interior courtyard within the neighborhood.


The far Eastern Cluster occupies another low rise and is nestled into the curved section of the palisade wall to the east of Mounds B and C. No artifact was recovered during the CSC from this area. The 12 structures here roughly have the same average floor size (75.96 m2) as those in the Southern Cluster. Structure 23, however, is considerably larger (377.38 m2) than the others in this cluster. Note the small Structures 30 and 38 in this cluster are outside of, but abut, the eastern side of the palisade; thus, above we suggested they may represent entryways, rather than habitations.


In the northwestern quadrant of the town, seven possible structures are suggested, and they likely represent five small, widely scattered habitation sites. The northwestern quadrant is generally flat and featureless. A high-density surface scatter was documented during the CSC at Structures 44 and 45 in the far southeastern corner of the field. The two structures may represent a winter-summer house paring. This is the closest habitation area to the Sherman Mound, and it produced a diverse ceramic assemblage that could possibly be interpreted as evidence of an elite or high status household due to the higher frequency of decorated pottery and Bell Plain. Additionally, a Middle Mississippian burial was identified in this area.


Lewis et al. (1998:15) note that Mississippian plazas exhibit a variety of shapes and sizes. At the Sherman Mound, geophysical and CSC data suggest the presence of two plazas. The Sherman Mound and two of the residential areas (Southern and Southeastern clusters), frame a western plaza that covers a roughly 100-x-130-m rectangular space (1.3 ha). An eastern plaza of roughly the same size can be suggested that is bounded by Mounds B and C, and a shared residential area (Southeastern Cluster) with the first plaza, as well as two other residential areas (Northern and Eastern clusters) along the palisade. Alternatively, if the two plazas are combined, then one “L”-shaped plaza covering about 2.6 ha is represented.


In comparison, the plaza at Chucalissa (40SY1), a small Mississippian mound center on the bluffs overlooking the Mississippi River at Memphis, covers only 0.28 ha (Childress 1992:34). At the Upper Nodena site, Dr. Hampson (1989:9) reported a sand-covered plaza covering 0.14 ha south of the main mound, that Mainfort (2010:4) interprets as a “chunky field.” At Langdon, a large, Middle Mississippian, fortified mound complex in the St. Francis basin, the plaza covers 0.74 ha (Dunnell 1998:204). In contrast, much larger plazas are found at Cahokia, where the Great Plaza covered 12 ha, and four smaller plazas each covered about 4 ha (Demel and Hall 1998:206).


It is suspected that the western plaza, that abuts the Sherman Mound to the west, is the original mound-and-plaza complex at the site (Lewis and Stout 1998:234). The eastern plaza, that abuts Mounds B and C on the east, is likely a secondary plaza that developed later in the occupation of the town.


Plazas are thought to have served as communal spaces that allowed all the town residents to “share in the ceremonies, rituals, and daily life experiences that unite and define a community” (Lewis et al. 1998:11). Public events that took place on the plazas within the Sherman Town likely included dances and feasts, and games, such as chunky. These spaces also probably served as market places.

Borrow Pit

The Sherman Mound required a large volume of soil to construct (see Mound Volume section below), thus there should be one or more borrow pits nearby. At Cahokia at least 20 borrow pits have been identified (Demel and Hall 1998:223). More locally, at Chucalissa (40SY1), a prominent borrow pit can still be observed approximately 200 m southeast of the main Mound A (Childress 1992:34). Per a revised radiocarbon chronology, Chucalissa Mound A was constructed a.d. 1250–1400, and not during the Walls phase (ca. a.d. 1425–1500) as previously thought, thus it is contemporary with the Sherman Mound (Franklin and McCurdy 2005).


Unfortunately, the topographic maps and DEM of the Sherman Mound vicinity fail to display any significant depressions that can be unambiguously interpreted as borrow pits within roughly 400 m of the main mound (see Figures 8–10). Perhaps the soil source was farther afield. However, for some time we have suspected that the area immediately north-northeast of the Sherman Mound, which is associated with the “Triangle” geophysical data collection tract, may represent a borrow pit that was partly infilled. This location has the lowest topography near the mound (see blue areas on DEM in Figure 10), and the railroad runs on a noticeably high embankment over this locus. A borrow pit here would have been highly efficient in terms of reducing labor during the mound construction, and it offered clay soil that would be good for use in earthworks. Shovel testing and surface examination of this area during the Paymaster Siding construction monitoring failed to locate any artifact within this suspected borrow pit location (Buchner 2014), thus the Middle Mississippian occupation and/or use of this area appears minimal.


If the “triangle” area was a borrow pit, then during the Middle Mississippian occupation of the site this locus was likely water-filled and served as a pond feature on the cultural landscape. This follows a similar pattern identified at Cahokia, where borrow pits 5-1 and 5-2 are associated with tri-mound groups (Demel and Hall 1998:222). Such ponds may have functioned as aquatic gardens, fish farms (see Subsistence section below) and/or sources of drinking water.

Burial Distributions

Prior to this study the only information regarding the distribution of human burials at the site came from Dr. Hampson’s early twentieth-century descriptions. He mentions burials in two locations: at the Sherman Mound; and within a “small mound” east of the Sherman Mound that we suggest is Mound B. A ca. 1938 “road cut” at the Sherman Mound exposed graves, and “intrusive” burials were noted in 1945 on the western and southern sides of the mound. Mound B contained an undisturbed burial that produced two bowls and a water bottle that are curated at the Hampson Archeological Museum State Park. This burial is thought to have been a fully articulated, extended-type interment. Presumably pothunters looted other burials at Mound B.


During the PTC study human remains were documented in two locations. On the southwestern flank of the Sherman Mound F-2, a basin-shaped pit, produced a fragmented long bone and 20 m to the south another human bone fragment was recovered that was likely displaced from F-2. AMS dating revealed that the F-2 burial is Protohistoric and dates ca. a.d. 1650. It is considered to be an example of the “intrusive” burials noted by Dr. Hampson.


Within a residential area approximately 100 m north of the Sherman Mound, a highly fragmented set of partial human remains was observed on the field surface adjacent to a mechanically cut field drain. Nearby ceramics that were likely unearthed with the bone fragments include Mississippi Plain (n=6) and an O’Byam Incised, var. O’Byam rim, and suggest that this burial is Middle Mississippian. Additional burials likely exist just beneath the plowzone in this portion of the site, but it is unclear if this area represents a discrete cemetery area or if they are simply associated with nearby structures.

Mound Volume

Platform mounds such as the Sherman Mound play a critical role to archaeologists interpreting Mississippian societies’ sociopolitical organization. Most scholars agree that these mounds served as residences and mortuaries for high status individuals and groups (Swanton 1911; Waring 1968), that they mark political centers within hierarchical settlement systems (Peebles and Kus 1977), and their construction legitimized and/or sanctified leadership and power roles (Anderson 1994; Lindauer and Blitz 1997).


LMS archaeologist Phillips visited the Sherman Mound in 1938 and 1940, and remarked on the 10-Q-2 LMS site card “It is…the largest mound in the locality.” As a result, information regarding the volume of the Sherman Mound can be significant for understanding both local and regional Mississippian sociopolitical organization.


Archaeologists are typically interested in mound volume calculations for two reasons: (1) to estimate the amount of labor needed to raise the earthworks; and/or (2) to make comparisons between mounds and mound sites. There are two opposing camps regarding the meaning of the volume of Mississippian platform mounds: (1) Duration of Use; and (2) Chiefly Power (Blitz and Livingood 2004). Under the Duration of Use model, mound volume is assumed to have accrued throughout a mound’s occupation via the periodic addition of new construction stages. This, in turn, implies that the largest mounds were used over the longest periods of time. In contrast, the Chiefly Power model proponents “consider mound volume to be the direct product of the quantity of labor invested in construction,” and thus mound volume is considered an “indirect measure of the scale of chiefly power” (Blitz and Livingood 2004:293). If the Sherman Mound site was occupied for a short duration, as is suspected, then the large size of the mound implies the authority of a powerful chief.


Blitz and Livingood (2004:292) noted, “archaeologists have been slow to quantify” the wide variation in the “volume and scale” of mound construction in a systematic way, but they utilize Payne’s (1994) main Mound Volume Index (MVI) in their analysis. Payne (1994:107) clearly stresses that her MVI is a simplified measure of volume that “should not be taken as representing the true volume nor used to calculate person-days necessary for construction,” instead MVI is for use only in comparing main mound volumes. The MVI calculation is as follows: MVI = (Basal Length ´ Basal Width ´ Height)/1,000.


To facilitate comparison to Payne’s (1994) and Blitz and Livingood’s (2004) results, the MVI for the Sherman Mound was calculated using 72.4-x-40.7 m as the basal dimensions, this excludes the “ramp,” and a height of 6.28 m. The resulting Sherman Mound MVI is 18.5, a relatively high number. By way of comparison, the main mound at the nearby Upper Nodena Site—with basal dimensions of 36.6-x-34.0 m and a height of 4.7 m per Mainfort (2010:4)—has an MVI of 5.8, considerably less than Sherman.


Payne (1994:108-114) was able to calculate MVI for 271 sites throughout the Southeast; however, it is not clear if the Sherman Mound is among these because only summary data are provided (most likely Sherman is not among these). Among Payne’s (1994:108) sites, the MVI results range 0.1–2,291.1, with the maximum being at Cahokia (11MS2). Note, however, the Cahokia result is ridiculously skewed by the presence of Monks Mound, which is >10 times that of the second largest main mound at Etowah (MVI = 215.0). Excluding Cahokia, the mean MVI for the remaining 270 sites falls to 15.2 and the median is 5.5. At 18.5, the Sherman Mound’s MVI is relatively high, but not within the elite group of 28 sites with MVI of >35 (Payne 1994:Table 3-5). In terms of Payne’s (1994:116) “Size Classes for Mississippian Mound Centers” the Sherman Mound and Village–with three mounds and a MVI of 18.5—falls within the “Small,” class which is one above the lowest “Very Small” and below the “Medium Small” category.


Blitz and Livingood (2004:Table 1) calculated MVI for 35 sites with excavated Mississippian platform mounds in an effort to evaluate the two competing interpretative models of mound volume. Their examples are all from east of the Mississippi River, and the nearest example to the Sherman Mound is Winterville (22WS500), a 23-mound complex in the lower Yazoo basin with an occupation) contemporary with Sherman (a.d. 1200–1450). At Winterville, the main mound (G) has an MVI of 7.7, less than half that of the Sherman Mound. Among Blitz and Livingood’s (2004:Table 1) 35 sites, only five mounds have a MVI over 15—Moundville R (1TU500), Dyar (9GE5), Cahokia 31 (11MS2), Angel F (12VG1), and Obion 6 (40HY14)—thus the Sherman Mound’s MVI of 18.5 is impressive.


Biplots for duration and MVI shows a “slight” trend for increasing number of mound stage over time, which Blitz and Livingood (2004:296) indicated “makes logical sense,” as mound construction was periodic and associated with significant events, such as world-renewal ceremonies or the death of a chief. Using another test—Spearman’s rank-order correlation (rs)—the most salient result was that mound building at major centers was “distinctly different” from that at single-mound sites or minor centers (Blitz and Livingood 2004:298). If major centers were excluded from rank-order correlation analysis, then duration accounted for 41 percent of the variation. Blitz and Livingood (2004:299) concluded that this would comfort the Duration of Use camp, while the Chiefly Power camp would counter that the “big sites are the key and that the relationship between duration and mound volume breaks down at these sites, leaving plenty of statistic room to argue for the influence of chiefs.” It is suggested that more sophisticated statistical approach to mound volume is needed.


Regarding the latter, Lacquement (2010) argues for the use of modern contouring software to calculate mound volume. Using the gridding method via the computer application DIDGER 4.0, the volumes of 21 mounds at Moundville were recalculated and compared to earlier volume estimates, including MVI derived volumes (Lacquement 2010:Table 2). The results revealed that the updated figures were much smaller than the original estimates. Lacquement (2010) argues that the overestimation of mound volume, in turn leads to an exaggerated view of the labor required to construct the earthworks. The inflated labor estimates make it problematic to assess the manpower needed to erect the earthworks, and thus complicates our view of the sociopolitical organization necessary to supply it.


Walker utilized the DEM created during his study and computer software to accurately calculate the volume of the Sherman Mound using the gridding method. The result is 5,466 m3 (193,030 ft.3).


Comparison of the Sherman Mound volume to the Lacquement’s (2010:Table 2) updated mound volume statistics reveals that Sherman would rank eighth in volume at Moundville, falling between Mound G (6,730 m3) and Mound C (5,080 m3). The largest earthwork at Moundville is Mound B (49,530 m3), and statistically it falls between Angle Mound A (51,035 m3) and Cahokia Mound 48 (42,230 m3; Lacquement 2010:Table 3).


Regarding construction, the Sherman Mound is most likely principally composed of hard-packed clay. Assuming hard-packed clay weighs 100 lbs. per cubic ft., then the weight of the Sherman Mound can be estimated to be 19,300,000 lbs. Further assuming that each basket load of clay used to construct the Sherman Mound weighed 50 lbs., then it is estimated that >386,000 individual basket loads were required to construct it.


The rate of basket load placement is the determining factor in estimating the amount of time needed to build the Sherman Mound. If one basket load were deposited every minute of a
12-hour workday (i.e., 720 baskets per day), then it would take approximately 536 days
(1.46 years) to build the mound. Alternatively, using a rate of one basket load every 5 minutes within a 12-hour workday (i.e., 144 baskets per day), then it would take approximately 2,680 days (7.34 years) to build the mound.

Settlement Patterns

As a fortified Middle Mississippian town, Sherman falls within a regional pattern of contemporary palisaded towns, such as Old Town Ridge and Langdon, and fortified villages, such as Moon. McNutt (1996:239) has suggested that these sites, which are all east of Crowley’s Ridge, represent an undefined Middle Mississippian phase similar to the Powers phase in Western Lowlands of Missouri. Other contemporary Middle Mississippian components that would presumably fall into this nameless phase would include Knappenburger, Schugtown, Webb (or Bay Mounds), McDuffee, Hazel, Zebree, Kochtitzky Ditch, Perry Dixon, and of course, the Sherman Mound.


More broadly during the Middle Mississippian period, fortified civic/ceremonial centers are a characteristic of most of the Central Mississippi Valley north of Memphis, Tennessee. Roughly halfway to Cahokia from Sherman, at the Mississippi-Ohio rivers confluence, the Cairo Lowland was a major center of Mississippian development (McNutt 1996:231; Phillips 1970:926). Multiple large fortified towns similar to, and contemporary with, Sherman are located there, such as Towosahgy, or Beckwith’s Fort (23MI2), a fortified 8-ha town (Cottier and Southard 1977). Dye (2009:146) suggests these fortifications not only provided physical protection, but also served to protect chiefly authority and the “power often housed in ancestor shrines.”


Fortified civic/ceremonial centers such as Sherman represent part of a larger Mississippian settlement pattern that can be viewed as a “continuum between fortified, nucleated, defensible villages on one end and economically efficient, dispersed households on the other” (Emerson 1997:64). In the Powers phase, Price (1978) delineated a four-tiered settlement system consisting of civic-ceremonial center, villages, hamlets, and limited-activity areas. Powers Fort (23BU10), a 4.4-ha multi-mound center, is cited as an example of a civic-ceremonial center and the twin Turner and Snodgrass sites are examples of planned fortified villages. Hamlets were thought to contain nine to 12 structures, while limited-activity areas represent farmsteads or household-level sites (Emerson 1997:66). Perttula (1998) suggests that the villages and smaller sites in the Powers phase were occupied for short durations (<5 years); this includes even the large fortified villages such as Snodgrass and Turner. In contrast, Powers Fort “maintained a sizeable population for the entire span of the Powers phase (ca. a.d. 1250–1350)” (Perttula 1998:196).


During the Middle Mississippian period a local settlement system can be postulated focused on the Sherman Mound that is similar to that of the Powers phase. At the apex of the hierarchy is the Sherman Mound, a fortified civic-ceremonial complex with three mounds. From here, a chief and elite faction ruled the polity.


Next down in the proposed hierarchy are villages, of which there appears to be two types. Moon is an example of a planned fortified village similar to Snodgrass and Turner (Benn 1998:255). In contrast, the Kochtitzky Ditch Site, located in the hinterlands of the Little River 17 km northwest of Sherman, appears to have been a non-fortified Middle Mississippian village that developed from an Early Mississippian base (Buchner et al. 2003). Importantly, it contained evidence for an elite household segregated and enclosed by a compound wall and a discrete ceremonial/mortuary area nearby. This implies that even within outlying villages, there was social stratification and ranking.


A local example of the third level settlement is the Big River Steel site (3MS780), a probable Middle Mississippian component located only 2.4 km east of the Sherman Mound. This site is interpreted as a 0.5-ha hamlet, although it could possibly be considered a small village
(M. Williams 1995). Test excavations revealed the presence of intact, stratified Mississippian deposits 1 m deep, including well-preserved architectural features suggestive of a series of superimposed burned structures (Buchner et al. 2012). A gradiometer survey revealed that the site likely contains the remains of 14–16 Mississippian structures distributed in four clusters.


Limited-activity areas, or farmsteads, theoretically “constitute the most common type of site recorded when intensive surveys are dome” (Morse and Morse 1983:253). The Pirani site (3CT324), located on a natural levee west of a relic channel of the Mississippi River in West Memphis, Arkansas, is a good example of a small Mississippian farmstead. Phase II investigations revealed well-preserved remains of a partly burned Mississippian structure—consisting of charred beams, wall poles, fused cane with wattle and daub, and split cane matting—immediately below the plowzone (Childress et al. 1995). The recovered Prehistoric ceramics (n=179) were all plain and “conformed very closely with the types and varieties found at the Chucalissa site (40SY1)” (Childress et al. 1995:74). A radiocarbon date was obtained from sections of a burned hickory pole (F-27) within the Site 3CT324 structure; it yielded an uncalibrated date of a.d. 1310 ± 50 (Beta-7624; Childress et al. 1995:86), which correlates with the late Moorehead phase at Cahokia and the Boxtown phase at Chucalissa. The Pirani site also produced a jar with an unusual Cahokia-like rim profile. Other Middle Mississippian farmsteads that have been tested and radiocarbon dated include Ritter Pecan Grove (3PO495; Spears and Taylor 1987) and the Sunday Site (3PO475; Buchner and Childress 1995).


Civic-ceremonial centers can themselves be ranked. As noted above, under Payne’s (1994:116) seven-level classification of Mississippian Mound centers, Sherman falls within the next to lowest class (“Small”) based on the presence of three mounds and an MVI of 18.5. Emerson (1997:72) and others commonly use a three-level civic-ceremonial center hierarchy. Paramount centers, such as Moundville and Cahokia, occupy the first tier. Sherman would either be a second- or, more likely, third-level mound center.


It is significant to note the presence of another little-known probable Middle Mississippian civic-ceremonial center located only 7.4 km northwest of the Sherman Mound. The Waller Bayou site (3MS81) reportedly covers 50 ha (125 ac.), and air photo analysis suggests it once contained five mounds (Lafferty and Chapman 1984:C-4). Lafferty and Chapman (1984) conducted limited testing at the site during the Tyronza River Watershed Phase I project (Lafferty et al. 1984), but this impressive site has been largely forgotten in “gray literature.” The ceramics recovered from Waller Bayou reveal it was occupied from the Middle Woodland to the Mississippian periods, but Lafferty and Chapman (1984) were reluctant to chronologically “place the site more precisely,” although it appears Middle Mississippian.


Importantly, the Waller Bayou site exhibits a number of intriguing affinities to the Sherman Mound. Structurally, Mound A at Waller Bayou is comparable to Sherman Mound A (Lafferty and Chapman 1984:C-3), and similar to Sherman, the site is capped by clayey Tunica soils. Additionally, Site 3MS81 is located on Waller Bayou, which provides a natural connection to Sandy Bayou and the Sherman Mound. Waller Bayou should be targeted for a research project similar to the PTC mitigation at the Sherman Mound.

Exchange Relations

One of the hallmarks of the Middle Mississippian period is the regional trade network whereby both elite goods and utilitarian items were exchanged. Presently we have limited evidence for participation in this system at the Sherman Mound; however, this is, in part, due to the limited excavations conducted at the site, coupled with the small assemblage size.


At the Sherman Mound, the presence of low frequencies of Burlington chert—one biface and a few pieces of debitage—is the best evidence for Cahokia contact. The Crescent Quarries are located approximately 270 km to the north of the site. Similar chert is reported from Early Mississippian contexts at Zebree, and it is associated with the Barrett complex (a.d. 1100–1200) in the southern St. Francis basin (House 1996:147). Previously, Buchner et al. (2003) argued that the identification of Barrett Complex-related material at Kochtitzky Ditch was significant, as it was the only good indication of southern influence and/or contact. The Barrett Complex, with its mix of traits (northern chert and southern ceramics), and coupled with its chronological position, might well be viewed as one of the precursors to the fully developed (Middle) Mississippian and the rise of chiefdoms in eastern Arkansas.


Other lithic artifacts suggestive of Cahokian influence in northeastern Arkansas include Cahokia type discoidals, Cahokia Notched points, and Mill Creek chert maces and hoes (Buchner and Albertson 2016); however, none is reported to date from Sherman. A stone effigy reportedly from Sherman is not stylistically similar to other Middle Mississippian stone statuary, such as an example from Schugtown (Morse and Morse 1983:Figure 11.5f), and may be counterfeit (i.e., fake).


Evidence for contact or exchange with the groups in the Ouachita Mountains at Sherman is limited to one novaculite artifact, a retouched piece. It is unclear which Prehistoric component this item is associated with, but if it is Middle Mississippian (as is the majority of the assemblage), then it indicates contact with Mississippian-Caddoan East phase (a.d. 1100–1350/1400) populations, that are the “first fully realized expression of the Caddoan tradition in the Ouachita River basin” (Early 2002:5).


More generally, there is limited archaeological evidence for Cahokian influence or trade in northeastern Arkansas (Buchner and Albertson 2016). Three sites within a roughly 40-km radius of the Sherman Mound have produced examples of Ramey Incised pottery: Banks Mound 3 (3CT16; Perino 1967); Knappenberger (3MS53; Payne 2009); and Perry Dixon (3MS600; Buchner et al. 2003). Arguably, the strongest evidence for Cahokia influence in northeastern Arkansas is the presence of shell artifacts at three sites. In the St. Francis River basin two Eddyville-style gorgets are known from the McDuffee site (3CG21; McGimsey 1964) and the birdman gorget was recovered from Old Town Ridge (3CG41; Brain and Phillips 1996:415; Lockhart et al. 2011:51-52; Morrow et al. 2013:4-5). Farther afield, on the edge of the Mississippi Embayment, the Akron Cup from the Akron Mound (3IN3) exhibits a Classic Braden design that “belongs to a Moorehead phase context that spans the period between 1200 and 1275” (Brown 2007:223).


Prior to the 2016–2017 PTC study there was little information regarding subsistence patterns at the Sherman Mound and Village. The special samples derived from F-1 and F-2, located on the southwestern flank of the Sherman Mound, were processed via flotation and yielded significant data regarding diet and foodways at the site.

Animal Remains

The majority of the faunal remains were recovered from F-1, a large ca. 1280–1390 trash pit.
F-2, a shallow, basin-shaped pit interpreted as a ca. 1650 burial, produced a much smaller faunal assemblage. Dr. Susan L. Scott remarked that overall, the Sherman assemblage reveals a strong reliance on fish, and the vast majority of remains are cycloid and ctenoid fish scales (288/586 fragments). At least eight types of fish were consumed at Sherman: Gar; Sucker; Shad; unidentified Catfish; Bullhead/Mudcat; Finfish; Sunfish; and Freshwater drum. Importantly, these species reveal that the Middle Mississippian fisherman at the Sherman Mound exploited both slackwater and river channel habitats. At the Middle Mississippian Kochtitzky Ditch (3MS599) site a related “backwater” exploitation pattern was documented (Buchner et al. 2003:140). Mainfort et al. (2007:116-117) also report a similar “abundance of quiet-water fish taxa” from the Late Mississippian contexts at the Upper Nodena site (3MS4) Block C.


The slackwater fish species were likely obtained from shallow lakes or sluggish bayous near the site, and the presence of such features may have had an influence on the site selection process. Thomas (1985[1894]:222-223) reported that Sherman Mound was located “at the head of Young’s lake, midway between Osceola and Pittsman’s landing,” and the 1847 General Land Office (GLO) plat map shows Swan and Carson lakes to the southwest of the site. Additionally, Sandy Bayou is located only 0.25-mi. south of the Sherman Mound, and the terrain near it is characterized as overflow land in the 1846 GLO survey notes. More generally, the oxbow lakes in the Mississippi meander belt support high fish biomass levels that were annually recharged via flooding. Limp and Reidhead (1979) argue that Prehistoric fishing techniques were more productive in slack-water settings than within river channels, in part because fish become concentrated when water levels drop during the summer.


The large mammals, small mammals, and birds that typically comprise a much larger portion of Mississippian faunal samples proved mostly unidentifiable in the Sherman samples. Scott suggests that deer is under-represented in the Sherman faunal assemblage. Deer bone is present in the general midden at the site, for example during the 2014 Paymaster Siding monitoring project, a bone fragment from a relatively large deer was recovered from a disturbed context near F-1 and F-2.


At the Middle Mississippian Kochtitzky Ditch village, deer meat accounted for 69 percent of the diet (Buchner et al. 2003:139). A much larger Upper Nodena site (3MS4) Block C faunal sample (n=8,292) also revealed that white-tailed deer supplied “most of the meat,” as well as skin, bones, and sinew for raw materials (Mainfort et al. 2007:115). Assuming that deer is under-represented in the Sherman sample, then the Sherman faunal assemblage is generally consistent with the regional Mississippian pattern, with heavy reliance on white-tailed deer and the abundant aquatic resources of the meander belt.

Plant Remains

Despite the small size and limited number of PTC flotation samples, Dr. Neal H. Lopinot delineated a considerable amount of information regarding plant food use and wood exploitation at the Sherman Mound and Village. The vast majority of the plant remains (858 items weighing 9.38 g) are classifiable as fuel/construction materials. The wood charcoal profile of F-1, a large ca. a.d. 1280–1390 trash pit, indicates exploitation of relatively mature forest stands—both lowland bald cypress stands and higher hardwood forests—near the site. The witness trees listed in the 1840s GLO map notes also reveal the presence of hardwood bottoms and seasonal overflow type forests in the site vicinity.


In contrast, the wood charcoal profile from F-2, which dates to roughly 350 years later than F-1, is suggestive of a more disturbed local environment by ca. a.d. 1650. As a result it was inferred that the climax forests in the vicinity of Sherman were cleared or reduced by ca. 1650; presumably by Middle and/or Late Mississippian peoples. At Cahokia, a similar problem with deforestation has been assessed (Lopinot and Woods 1993), and Emerson (1997:152-154) argues this may have impacted the settlement system.


Interesting, no cane was recovered from F-1 and F-2, although there were several cane breaks in the site vicinity during the 1840s, and cane was an important Prehistoric raw material for the manufacture of baskets and mats, as well as the construction of waddle-and-daub structures.


Analysis of the carbonized seeds from F-1 and F-2 reveals the consumption of both wild and cultivated plant foods. Wild plant foods included pecans, nuts, and persimmons, and possibly ragweed and pokeweed. Similar wild plant foodstuffs (i.e., nuts and persimmons) are reported from Late Mississippian contexts at the Upper Nodena site (3MS4) Block C (Blake and Cutler 1979). Persimmon occurred in 48 percent of the features at the Kochtitzky Ditch site (Lopinot and Thomas 2003:140). Lopinot, in Buchner et al. (2017) indicates that persimmon seeds are “almost ubiquitous at Mississippian sites in the Lower Mississippi River valley,” and that persimmon and hickory/pecan were part of a wild food consumption pattern of considerable antiquity.


Cultivated plants at Sherman include maize and maygrass, and probably chenopod. Evidence for beans, squash, gourds, and sunflowers is absent from the F-1 and F-2 samples, but these species were likely cultivated at Sherman and can be expected in future samples. The dominance of maize, maygrass, and chenopod is consistent with the findings from the Kochtitzky Ditch and Perry Dixon sites, well-excavated Middle Mississippi components located 17 km to the northeast of Sherman (Lopinot and Thomas 2003). Thus, farming occurred at the Sherman Mound and Village well before it was part of Sans Souci Plantation.


Unfortunately, no maize kernel from Sherman was sufficiently intact to provide size and shape information. The maize remains at the Upper Nodena site (3MS4) Block C were much better preserved (Blake and Cutler 1979), and concentrated in the C-42 Block where Morse (1990:75) suggests an aboveground granary burned.





Anderson, D.G.

1994 The Savannah River Chiefdoms. University of Alabama press, Tuscaloosa.


Arkansas Department of Parks and Tourism

2017 Hampson Archeological Museum State Park page on the State Parks of Arkansas webpage. Available at


Benn, D.W.

1990 Excavations at the Priestly Site (3PO490) An Emergent Mississippian Community in Northeastern Arkansas. Two volumes. Center for Archaeological Research, Southwest Missouri State University, Springfield. Submitted to the Arkansas Highway and Transportation Department.


1992 Excavations at the Moon Site (3PO488) A Middle Mississippian Village in Northeastern Arkansas. Two vols. Center for Archaeological Research, Southwest Missouri State University, Springfield. Submitted to the Arkansas Highway and Transportation Department.


1998 Moon: A Fortified Mississippian-Period Village in Poinsett County, Arkansas. In Changing Perspectives on the Archaeology of the Central Mississippi Valley, Edited by M.J. O’Brien and R.C. Dunnell, pp. 225-257. The University of Alabama Press.


Blake, L.W., and H.C. Cutler

1979 Plant Remains from the Upper Nodena Site. The Arkansas Archeologist 20:53-58.


Blitz, J.H., and P. Livingood

2004 Sociopolitical Implications of Mississippian Mound Volume. American Antiquity 69(2):291-301.


Brain, J.P., and P. Phillips

1996 Shell Gorgets Styles of the Late Prehistoric and Protohistoric Southeast. Peabody Museum Press.


Brown, J.A.

2007 On the Identity of the Birdman within Mississippian Period Art and Iconography. In Ancient Objects and Sacred Realms: Interpretations of Mississippian Iconography, edited by Kent Reilly III and James F. Garber, pp. 56-106. University of Texas Press.


Buchner, C.A.

1993 Cultural Resources Survey and Initial Site Testing along Kochtitzky Ditch No. 1, in the Tyronza Sunk Lands, Mississippi County, Arkansas. Garrow & Associates, Inc. Final report submitted to the U.S. Army Corps of Engineers, Memphis District.


2014 Phase I Cultural Resources Survey of a Reported Unmarked Cemetery (3MS926) within a Proposed BNSF Railroad Siding, Mississippi County, Arkansas. Panamerican Consultants, Inc. Report No. 34083. Submitted to Hanson Professional Services, Inc.


2015 Archaeological Monitoring During the BNSF Railway Siding Construction at the Sherman Mound (3MS16), Mississippi County, Arkansas. Panamerican Consultants, Inc. Final Report 34119. Prepared for Hanson Professional Services, Inc.


Buchner, C.A., and E. Albertson

2016 Searching for Cahokia Connections in Northeast Arkansas. Paper prepared for the 37th Mid-South Archaeological Conference, July 16-17, 2016, University of Memphis.


Buchner, C.A., E.S. Albertson, N.H. Lopinot, L.A. Thomas, E. Breitburg, and J.V. Ward

2003 Mississippian Transitions at John’s Lake. Arkansas Archeological Survey Research Series 60.


Buchner, C.A., and M.R. Childress

1995 Archaeological Investigations in Northeast Arkansas, 1988–1989. Garrow and Associates, Inc. Final report submitted to the U.S. Army Corps of Engineers, Memphis District.


Buchner, C.A., K. Oesch, and J. Rawls

2016 Archaeological Mitigation at the Grigsby Site (3RA262/3RA379/3RA380), Randolph County, Arkansas. Panamerican Consultants, Inc. Prepared for MidSouth Utility Contractors, LLC.


Buchner, C.A., K. Oesch, C.P. Walker, S.L. Scott, N.H. Lopinot, and E.S. Albertson

2017 Archaeological Mitigation at the Sherman Mound and Village (3MS16), Mississippi County, Arkansas. Panamerican Consultants, Inc. Report 36085. Prepared for BNSF Railway.


Buchner, C.A., K. Oesch, W. Wilson, and C.P. Walker

2016 Archaeological Survey at Warren Wilson College for the Swannanoa River Valley Flood Risk Management Study, Buncombe County, North Carolina. Panamerican Consultants, Inc. Report No. 35162. Submitted to the U.S. Army Corps of Engineers, Nashville District.


Buchner, C.A., and C.P. Walker

2014 Geophysical Survey of the Proposed BNSF Rail Siding at the Sherman Mound (3MS16), Mississippi County, Arkansas. Panamerican Consultants, Inc. Report No. 34074. Submitted to Hanson Professional Services, Inc.


Buchner, C.A., C.P. Walker, and K. Oesch

2012 National Register of Historic Places Eligibility Testing of Three Sites at the Great River Super Site, Mississippi County, Arkansas. Panamerican Consultants, Inc. Report No. 31182. Report submitted to Mississippi County & Great River Economic Development.


Childress, M.R.

1992 Mortuary Vessels and Comparative Ceramic Analysis: An Example from the Chucalissa Site. Southeastern Archaeology 11(1):31-50.


Childress, M.R., J.R. Oliver, C.A. Buchner, L.W. Weaver, and G.G. Weaver

1995 Archaeological Investigations in Eastern Arkansas, 1989-1991. Garrow & Associates, Inc. Submitted to the U.S. Army Corps of Engineers, Memphis District.


Childress, M.R., G.G. Weaver, J.L. Hopkins, and M. Oats

1995 Cultural Resources Investigations at Mound City Plantation, Crittenden County, Arkansas. Garrow & Associates, Inc. Submitted to Promus Companies, Inc., Memphis.


Cottier, J.W, and M.D. Southard

1977 An Introduction to the Archaeology of Towosahgy State Archaeological Site. The Missouri Archaeologist 38:230-271.


Davis, H. (editor)

1982 A State Plan for the Conservation of Archeological Resources in Arkansas. AAS Research Series No. 21. Arkansas Archeological Survey, Fayetteville. Revised 2010.


Demel, S.J., and R.L. Hall

1998 The Mississippian Town Plan and Cultural landscape of Cahokia, Illinois. In Sacred Spaces: Searching for an Architectural Grammar, edited by R.B. Lewis and C. Stout, pp. 200-226. University of Alabama Press.


Dickens, R.S., Jr.

1976 Cherokee Prehistory: The Pisgah Phase in the Appalachian Summit Region. The University of Tennessee Press, Knoxville.


Dunnell, R.C.

1998 The Langdon Site, Dunklin County, Missouri. In Changing Perspectives on the Archaeology of the Central Mississippi Valley, Edited by M.J. O’Brien and R.C. Dunnell, pp. 200-224. The University of Alabama Press.


Dye, D.H.

2009 War Path, Peace Paths. Altamira press.


Early, A.M.

2002 The East Phase. Field Notes 304:4-8. January/February 2002. Arkansas Archeological Society, Fayetteville.


Edrington, M.F.

1962 History of Mississippi County, Arkansas, edited and compiled by Mabel F. Edrington.


Emerson, T.E.

1997 Cahokia and the Archaeology of Power. University of Alabama Press.



Fisk, H.N.

1944 Geological Investigation of the Alluvial Valley of the Lower Mississippi River. U.S. Army Corps of Engineers, Vicksburg District.


Franklin, J.D., and T.D. McCurdy

2005 A Radiocarbon Chronology for Mound A [Unit 5] at Chucalissa in Memphis, Tennessee. Tennessee Archaeology 2(1):32-45.


Garland, E.B.

1992 The Obion Site: An Early Mississippian Center in Western Tennessee. Cobb Institute of Archaeology, Report of Investigations 7. Mississippi State University.


Hampson, J.K.

1989 The Nodena Site. In Nodena, edited by Dan F. Morse, pp. 9-22. Arkansas Archeological Survey Research Series No. 30.



House, J.H.

1996 East-Central Arkansas. In Prehistory of the Central Mississippi Valley, edited by C.H. McNutt, pp. 137-154. University of Alabama Press, Tuscaloosa.


2013 Native American Material Culture for the Wallace Bottom Site, Southeastern Arkansas. Southeastern Archaeology 32(1):54-69.


Hunt, Steven

2016 Construction Monitoring for the Burlington Northern Santa Fe Railroad Positive Train Control River Subdivision Mile Post 257.85, Mississippi County, Arkansas. Quality Services, Inc. Manuscript on file at the Arkansas State Historic Preservation Office.


Keel, B.C.

1976 Cherokee Archaeology: A Study of the Appalachian Summit. The University of Tennessee Press, Knoxville.


Klinger, Timothy C.

1974 Report on the 1974 Test Excavations at the Knappenberger Site, Mississippi County, Arkansas. The Arkansas Archeologist 45-72.


Klinger, T.C., S.M. Imhoff, R.J. Cochran, Jr.

1983 Brougham Lake. Historic Preservation Associates Reports 83-7. Final report submitted to the U.S. Army Corps of Engineers, Memphis District.


Lacquement, C.H.

2010 Recalculating mound Volume at Moundville. Southeastern Archaeology 29(2):341-354).


Lafferty, R.H., III

1973 An Analysis of Prehistoric Southeastern Fortifications. M.S. thesis, Southern Illinois University, Department of Anthropology.


2008 The Diffusion of Shell-Tempered Pottery into the Baytown Area of the Northern Lower Mississippi Valley. Southeastern Archaeology 27(2):188.


Lafferty, R.H., III, and L.M. Chapman

1984 Appendix C: Intensively Tested Sites. In A Cultural Resources Survey and Evaluation in the Tyronza River Watershed Phase I Area Mississippi County, Arkansas. Mid-Continental Research Associates, Inc., Report 84-2. Final report submitted to the Soil Conservation Service.


Lafferty, Robert H., III, Lawrence G. Santeford, Phyllis A. Morse, and L. Michael Chapman

1984 A Cultural Resources Survey and Evaluation in the Tyronza River Watershed Phase I Area Mississippi County, Arkansas. Mid-Continental Research Associates, Inc., Report 84-2. Final report submitted to the Soil Conservation Service.


Latham, M.A.

2001a Phase I Cultural Resources Investigation for Plum Point Energy Associates, LLC Proposed Power Plant Construction near Sand Souci, Mississippi County, Arkansas. Burns & McDonnell. Prepared for Plum Point Energy Associates, LLC. On file at the Arkansas Archeological Survey as AMASDA Report. No. 6570.


2001b Management Summary of the Cultural Resources in the Plum Point Project Area. Burns & McDonnell. Prepared for Plum Point Energy Associates, LLC. On file at the Arkansas Archeological Survey as AMASDA Report. No. 6572.


Lewis, R.B.

1996 The Western Kentucky Border and the Cairo Lowland. In Prehistory of the Central Mississippi Valley, edited C.H. McNutt, pp. 47-76. University of Alabama Press.


Lewis, R.B., and C. Stout

1998 The Town as Metaphor. In Mississippian Towns and Sacred Spaces: Searching for an Architectural Grammar, edited by R.B. Lewis and C. Stout, pp. 227-243. University of Alabama Press.


Lewis, R.B., C. Stout, and C. Wesson

1998 The Design of Mississippian Towns. In Mississippian Towns and Sacred Spaces: Searching for an Architectural Grammar, edited by R.B. Lewis and C. Stout, pp. 1-21. University of Alabama Press.


Limp, F.W., and V.A. Reidhead

1979 An Economic Evaluation of the Potential of Fish Utilization in Riverine Environments. American Antiquity 44(1):70-78.


Lindauer, O., and J.H. Blitz

1997 Higher ground: The Archaeology of North American Platform Mounds. Journal of Archaeological Research 5:169-207.


Lockhart, J., J.E. Morrow, and S. McGaha

2011 A Town at the Crossroads: Site-Wide Gradiometry Surveying and Mapping at Old Town Ridge (3CG41) in Northeastern Arkansas. Southeastern Archaeology 30(1):51-63.


Lopinot, N.H., and L.A. Thomas

2003 Archaeobotany. In Mississippian Transitions at John’s Lake: Data Recovery Excavations of Three Buried Sites in Northeast Arkansas, pp. 140-149. Arkansas Archeological Survey Research Series 60.


Lopinot, N.H., and W.I. Woods

1993 Wood Overexploitation and the Collapse of Cahokia. In Foraging and Farming in the Eastern Woodlands, edited by C. Scary, pp. 206-231. University Press of Florida.


Mainfort, R.C., Jr.

1996 The Reelfoot Lake Basin, Kentucky and Tennessee. In Prehistory of the Central Mississippi Valley, edited by C.H. McNutt, pp. 77-96. University of Alabama Press.


2010 Archeological Investigations at Upper Nodena: 1973 Field Season. Arkansas Archeological Survey Research Series No. 64.


Mainfort, R.C., Jr., J.M. Compton, and K.C. Cande

2007 1973 Excavations at the Upper Nodena Site. Southeastern Archaeology 26(1): 108-123.


Martin, W.W., L.M. Parks, and L.L. Ayres

1993 The Orr Site, 23SO132: Early Middle Mississippian Occupation of the Western Lowlands of Southeast Missouri. The Missouri Archaeologist 54:21-54.


McGimsey, C.R., III

1964 An Engraved Shell Gorget from the McDuffee Site. Arkansas Archeological Society Bulletin V(7):128-129.


McNutt, C.H. (editor)

1996 Prehistory of the Central Mississippi Valley. University of Alabama Press, Tuscaloosa.


Mitchem, J.M.

2010 Mississippian Fortifications at Parkin and Neely’s Northeast Arkansas. A paper presented at the 67th Southeastern Archaeological Conference, Lexington, Kentucky.


Morgan, W.N.

1980 Prehistoric Architecture in the Eastern United States. MIT Press.


Morrow, J., R.A. Taylor, J. Lockhart, and S. McGaha

2013 Old Town Ridge (3CG41): A Palisaded Middle Mississippi Period Village in the Land of Milk and Honey. The Arkansas Archeologist 51:1-24.


Morse, D.F.

1990 The Nodena Phase. In Towns and Temples Along the Mississippi, edited by David H. Dye and Cheryl Anne Cox, pp. 69-97. University of Alabama Press.


1998 A Shark Tooth Club From Arkansas. Field Notes The Newsletter of the Arkansas Archeological Society. No. 282:5-6.


Morse, D.F., and P.A. Morse

1983 Archaeology of the Central Mississippi Valley. Academic Press, New York.


1988 The Seventeenth Century Michigamea Village Location in Arkansas. Paper presented at the Conference on French Colonial Archaeology, Springfield, Illinois, April 9, 1988.


1990 Emergent Mississippian in the Central Mississippi Valley. In The Mississippian Emergence, pp. 153-174, edited by B.D. Smith. Smithsonian Institution Press, Washington, D.C.


1992 The Seventeenth-Century Michigamea Village Location in Arkansas. In Calumet & Fleur-de-lys: Archaeology of Indian and French Contact in the Midcontinent, edited by J.A. Walthall and T.E. Emerson,pp. 55-76. Smithsonian Institution Press, Washington, D.C.


Moselage, J.H.

1962 The Lawhorn Site. Missouri Archaeologist 24-1-105.


O’Brien, M.J.

2002 Mississippian Community Organization the Powers Phase in Southeastern Missouri. Academic press.


Payne, C.

1994 Mississippian Capitals: An Archaeological Investigation of pre-Columbian Political Structure. Unpublished Ph.D. dissertation, university of Florida, Gainesville, University Microfilms, Ann Arbor.



Peebles, C.S., and S.M. Kus

1977 Some Archaeological Correlates of Ranked Societies. American Antiquity 42:421-448.


Perino, G.

1967 The Cherry Valley Mounds and Banks Mound 3. Central States Archaeological Society, Inc., Memoir 1.


Perttula, T.K.

1998 Power Fort: A Middle Mississippian-Period Fortified Community in the Western Lowlands of Missouri. In Changing Perspectives on the Archaeology of the Central Mississippi Valley, Edited by M.J. O’Brien and R.C. Dunnell, pp. 169-199. The University of Alabama Press.


Phillips, P.

1970 Archaeological Survey in the Lower Yazoo Basin, Mississippi, 1949-1955. Papers of the Peabody Museum of American Archaeology and Ethnology Vol. 60. Harvard University, Cambridge.


Phillips, P., J.A. Ford, and J.B. Griffin

1951 Archaeological Survey in the Lower Mississippi Alluvial Valley, 1940–1947. Papers of the Peabody Museum of American Archaeology and Ethnology Vol. 25. Harvard University, Cambridge.


Prentice, G.

2000 Ancient Indian Architecture of the Lower Mississippi Delta Region: A Study of Earthworks. National Park Service, Southeast Archeological Center. Second Edition.


Price, J.E.

1978 The Settlement Patter of the Powers Phase. In Mississippian Settlement Patterns, edited by B.D. Smith, pp. 201-231. Academic Press.


Price, J.E., and J.B. Griffin

1979 The Snodgrass Site of the Powers Phase of Southeast Missouri. Museum of Anthropology, University of Michigan. Anthropological Papers No. 66. Ann Arbor.


Saatkamp, J.A.

2007 Cultural Resources Survey of Approximately 3,000 Acres South of Osceola, Mississippi County, Arkansas. Panamerican Consultants, Inc. Report No. 27009. Prepared for the Great River Economic Development.


Saucier, R.T.

1994 Geomorphology and Quaternary Geologic History of the Lower Mississippi Valley. Volumes I and II. U.S. Army Corps of Engineers, Mississippi River Commission, Vicksburg.


Sherrod, P.C., and M.A. Rolingson

1987 Surveyors of the Ancient Mississippi Valley. Arkansas Archeological Survey Research Series No. 28.


Smith, K.E.

1992 The Middle Cumberland Region: Mississippian Archaeology in North-Central Tennessee. Unpublished Ph.D. dissertation, Vanderbilt University, Nashville. University Microfilms, Ann Arbor.


Spears, C.S.

1978 The DeRossitt Site (3SF49): Applications of Behavioral Archeology to a Museum Collection. M.A. Thesis, Department of Anthropology, University of Arkansas, Fayetteville.


Spears, C.S., and R.A. Taylor

1987 An Archaeological Survey, Intensive Site Testing and Geomorphology, Pollen Analysis along the Lower Portion of Ditch 1, Poinsett County, Arkansas. SEARS Report No. 87-4. Submitted to the U.S. Army Corps of Engineers, Memphis District.


Steponaitis, V.P.

1983 Ceramics, Chronology, and Community Patterns: An Archaeological Study at Moundville. Academic Press, New York.


Swanton, J.R.

1911 Indian Tribes of the Lower Mississippi Valley and Adjacent Coast of the Gulf of Mexico. Bulletin No. 43. Bureau of American Ethnology, Smithsonian Institution, Washington D.C.


Thomas, C.

1985 Report on the Mound Explorations of the Bureau of Ethnology. Smithsonian Institution Press, Washington, D.C. Classics of Smithsonian Anthropology, originally printed in 1984.


Waring, A.J., Jr.

1968 The Southern Cult and Muskogean Ceremonial. In The Waring Papers, edited by Stephen Williams, pp. 30-69. Papers of the Peabody Museum of Archaeology and Ethnology Vol. 58, Harvard University, Cambridge.


Williams, M.

1995 Chiefly Compounds. In Mississippian Communities and Households, edited by J.D. Rogers and B.D. Smith, pp. 124-133. University of Alabama Press.


Williams, S.

1980 Armorel: A Very Late Phase in the Lower Mississippi Valley. Southeastern Archaeological Conference Bulletin 22:105-110.


1991 Fantastic Archaeology: The Wild Side of North American Prehistory. University

of Pennsylvania Press.

Go Back