A Mississippian Period Cemetery in Northwest Florida
The Hickory Ridge site (8ES1280), located in Escambia County, Florida, is an undisturbed Pensacola cemetery investigated by the University of West Florida (UWF). The site was originally recorded in 1983 during a reconnaissance-level survey of southwestern Escambia County. At that time Hickory Ridge was characterized as a diffuse scatter of Weeden Island artifacts and recent historical material. The site was relocated in late 1988 during a systematic Phase I archaeological survey undertaken for Escambia County by UWF (Phillips 1989a). The horizontal limits of the site were redefined during this survey, and a concentration of shell-tempered sherds was identified. Most of the sherds derived from two partially reconstructible vessels with complex incised motifs. These ceramic types are often associated with Mississippian period burials in this region. Although no skeletal material was encountered, the ceramics suggested the presence of a cemetery (Phillips 1989a:60). Phase II test excavations conducted in 1989 in the area of these unusual ceramics revealed a dense layer of shell-tempered sherds within a 15 m by 18 m area, three Mississippian burials, and several reconstructible vessels. This work produced significant data on Mississippian mortuary practices along the northern Gulf Coast. The grave goods consisted of Bottle Creek and Bear Point phase ceramic material, including Moundville Incised and Moundville Engraved types, and exotic lithic artifacts and whelk shell objects. A radiocarbon date of 500 +/- 60 B. P. (Beta 30702) was obtained from one of the burials. This article describes the methods and results of the test investigations.
Formal archaeological investigations in the north-central Gulf Coast region began with Sternberg's (1876) excavations at the Bear Point site (1BA1), located on the eastern shore of Perdido Bay. This work listed burials and artifact assemblages, and produced a collection of shell-tempered vessels. In the 1880s, Walker (1885) identified shell middens in the Pensacola and Choctawhatchee Bay systems, and provided fairly complete descriptions of the archaeological materials encountered. At the turn of the century, C. B. Moore (1901, 1918) visited the northern Gulf Coast and investigated numerous sites. Among these were Bear Point (1BA1), Santa Rosa Sound (8SR1), Graveyard Point (8SR3), Maester Creek Mound (8SR870), Fort Walton Temple Mound (8OK6), and Hogtown Bayou (8WL9). Primarily interested in the spectacular mound and burial sites, Moore published detailed descriptions of his work in the Journal of the Academy of Natural Sciences of Philadelphia. He described mortuary practices and documented differences in pottery styles between the Mobile-Pensacola and Apalachee Bay regions (Willey 1949:24-25). W. H. Holmes (1903), one of the most significant archaeologists of his day, analyzed Moore's ceramic collections from Bear Point on Perdido Bay, as well as several site collections recovered along Choctawhatchee Bay. His work identified three major ceramic ware groups: the Mobile-Pensacola, the Apalachicola, and the Appalachian (Willey 1949:27). Holmes observed the similarities and differences among these wares, and noted that a decrease in the Mobile-Pensacola ware and an increase in the Apalachicola ware occurred between Choctawhatchee Bay and the Apalachicola River.
The next substantive archaeological work undertaken in the region was conducted by researchers from Columbia University under sponsorship of the National Park Service (Willey 1949). In his monumental Archeology of the Florida Gulf Coast, Willey (1949) developed a prehistoric chronological framework and produced the first ceramic typologies for the Gulf Coast. This work defined the Fort Walton period and described the shell-tempered pottery of the Pensacola Series. Willey associated the Pensacola Series, found near the western extreme of the geographic area of Fort Walton, with this Mississippian culture. Building on Willey's framework, Wimberly (1960) developed a shell-tempered ceramic typology for the southwest Alabama region that remained the standard for many years.
In more recent years, there have been a number of refinements to Willey's original concept of Mississippian culture on the northern Gulf Coast. Brose and Percy (1978) described the "Pensacola-Fort Walton culture" as a coastal adaptation heavily influenced by the Apalachicola-Fort Walton culture. This model views the socio-economic adaptation of Pensacola-Fort Walton as very similar to the preceding Weeden Island period with a veneer of Mississippian attributes. Tesar (1980) defined five Fort Walton subareas; the Pensacola-Fort Walton subarea extends from Choctawhatchee Bay westward to Mobile Bay.
Fuller and Stowe (1982: 48), building on the work of Willey (1949), Wimberly (1960), Phillips (1970), and Jenkins (1981), devised a shell-tempered ceramic typology based on the type-variety system. With this analytical tool they isolated two northern Gulf Coast Mississippian pottery complexes: Bottle Creek and Bear Point. Subsequently, Stowe (1985: 145) identified these manifestations as the "Pensacola variant" and defined the Bottle Creek phase (Stowe 1985:144-149). This late Middle Mississippian phase dates from ca. A.D. 1200-1450 and extends westward from Choctawhatchee Bay into coastal Mississippi (between the Pascagoula and Pearl rivers), and northward near Selma, Alabama. Concurrent work by Fuller (1985:150-155) defined the Late Mississippian and Proto-historic Bear Point phase; this phase encompasses a similar, though somewhat less extensive geographic range. Fuller and Stowe's ceramic typology and the resultant phase designations represent significant advancements in the effort to define the Pensacola complex.
Recent work indicates that there are substantial differences between Fort Walton and Pensacola culture (Milanich 1994: 381). In contrast to the Fort Walton culture with its multiple mound complexes, agricultural economy, and rigidly stratified social system, the Pensacola political and economic systems, as a rule, were probably much less complex (Milanich 1994: 386). Knight (1984: 200-201) suggests that the Pensacola complex represents a coastal adaptation, with a long areal developmental history, that extends from the head of Choctawhatchee Bay westward to the mouth of the Mississippi River, and northward to the Alabama-Tombigbee confluence. This complex has several regional variants that are distinguishable by ceramic assemblages and settlement strategies. Most of these variants are simple chiefdoms that centered around the bay systems in areas with low agricultural potential (Bense 1994:234). Implicit in the models of Knight (1984) and Fuller and Stowe (1982) are significant influences originating with Moundville, an extremely large Mississippian mound complex on the Black Warrior River in central Alabama.
Researchers agree that the Pensacola variant, characterized by shell-tempered pottery with similar design elements, is geographically limited to the coastal areas of extreme northwest Florida and adjacent areas of southwest Alabama and Mississippi. Furthermore, Pensacola subsistence was based on coastal resources with an admixture of limited agriculture, depending on the availability of suitable soils. Researchers also agree that there is a low frequency of mound complexes within the geographic range of the Pensacola variant. Beyond this, Pensacola culture is an archaeological concept that remains poorly understood.
The Hickory Ridge site lies west of Pensacola on a large peninsula that is formed by Perdido and Pensacola Bays (see Figure 1). Perdido Key lies to the south, separating Big Lagoon from the Gulf of Mexico. The site is situated at an elevation of about 20 ft Above Mean Sea Level in a hardwood hammock on the edge of a relict marine terrace. Big Lagoon lies approximately 400 m to the south of the site across an extensive swamp. A first order stream dissects the terrace and flows into the swamp approximately 30 m southwest of the site; a Mississippian village site (8ES1052) lies 50 m west of Hickory Ridge across this drainage. The soils on the Hickory Ridge site, classified as Lakewood sand, level phase, are highly acidic, excessively drained, and contain very little organic material (Carlisle 1960). The preservation of bone and other organics in these sediments is extremely poor.
Hickory Ridge is located in a remote, heavily wooded area of southwestern Escambia County that has escaped intensive development thus far. Although the area appears to have been logged on a number of occasions, producing some shallow gullies and washes, these activities do not appear to have inflicted serious damage to the site.
The original ceramic concentration (Feature 1) that contained the mortuary-related vessels identified in the Phase I survey provided the focal point for the Phase II testing. A 30 m area around this feature was investigated in a systematic manner using shovel testing and excavation of 1 m2 test units. A plan view of the test excavations is shown in Figure 2.
The first step was an intensive shovel testing program that was utilized to identify additional ceramic concentrations and to define the horizontal limits of the cemetery. Shovel testing proceeded in three stages. First, a 30 m square area surrounding Feature 1 was gridded off. Within this 30 m grid, shovel tests were excavated at 5 m intervals. Following this, a 16 m square with a 2 m grid was centered on Feature 1, and shovel tests were excavated at 2 m intervals. Finally, the 30 m grid was extended outward and shovel tests were judgmentally placed along 5 m transects at 1, 2, or 5 m intervals. These judgmentally placed shovel tests were excavated to provide a better definition of the horizontal limits of the cemetery.
The shovel tests measured approximately 30 cm on a side and were excavated to a depth of at least 50 cm below the surface. The fill from these small test units was screened through 1/4 in. hardware cloth. The data were used to plot the horizontal distribution and density patterns of the cultural material recovered within the Hickory Ridge cemetery area.
1 m2 Excavation Units
Ten 1 m2 excavation units were placed within the 30 m square area surrounding Feature 1. The units were placed in those areas where shovel testing produced the highest artifact densities. The strategy for excavating each 1 m2 unit involved first removing the root mat and light gray sand overburden as a unit. This stratum was screened through 1/4 in. hardware cloth. The buried land surface contemporaneous with the cemetery was encountered at the bottom of this gray sand. The strata below the gray sand overburden were shovel shaved and removed in arbitrary 5 cm levels; the fill from these 5 cm levels also was screened through 1/4 in. hardware cloth. The floor of each arbitrary level was troweled to define any features. The 1 m2 units were excavated to at least 10 cm below the last level that yielded artifacts.
Feature and Burial Excavation
Eight ceramic concentrations were encountered on the buried land surface and in the upper levels of the underlying strata. Each concentration was given a feature designation, documented, and then removed as a single unit. Carbon samples for possible radiometric dating were taken whenever in situ carbon was encountered.
Burials were treated in a manner similar to features; however, no human remains were removed. When human skeletal material was encountered, the remains were exposed and pedestalled and the matrix was processed through a 1/16 in. screen. Following the documentation procedures and removal of associated artifacts for further analysis, the burials were very carefully covered with screened matrix.
Following the field work, all artifacts, carbon samples, level, feature and burial forms, notes, drawings, and photographs were returned to UWF for processing and analysis. The artifacts were washed, sorted, catalogued, and classified. The ceramic analysis included an extensive period of vessel reconstruction. The catalogued and classified sherds were placed on a large table. As each vessel was reconstructed, provenience data were recorded, the vertical and horizontal location of each vessel was plotted, and its association with features and burials was identified.
During the Phase II testing, 210 shovel tests were excavated: cultural material was recovered in 57 of these small test units. As Table 1 shows, 671 artifacts were recovered. These shovel tests revealed that artifacts, primarily shell-tempered ceramics, were concentrated within a 15 m by 18 m area (see Figure 2). Inside this small area of deposits, ten 1 m2 units were excavated. These larger excavation units produced additional ceramic material and revealed 8 features and 3 burials. Among the ceramic material recovered by shovel testing, test excavation, feature removal, and burial excavation were 30 partially reconstructible vessels. No midden deposits were encountered during testing. Figure 3 is a topographic map of the cemetery area showing the cemetery boundary, as defined by the cultural deposits recovered during shovel testing, as well as the location of the 1 m2 units, and the features and burials. Cemetery stratigraphy, 1 m2 excavations, features, and burials are discussed below.
Three stratigraphic zones (Zones 1, 2, and 3) were defined (see Figure 4). The expression of each zone varied from unit to unit. Zone 2, for example, varied considerably in thickness: along the western edge of the cemetery this stratum was less than 10 cm thick.
Zone 1 consisted of a gray sand overburden (10 YR 5/1 to 6/1), which blanketed the entire site and varied in thickness from 5-27 cm. This stratum was a deposit of wind-blown sand that originated on the exposed Gulf beaches to the south of the site. No artifacts were recovered in the upper limits of Zone 1. The lower part of the zone contained varying amounts of cultural material that appeared to have been on the surface of the underlying stratum.
one 2 consisted of a dark yellowish brown to brownish yellow sand (10YR 4/6 to 6/6), containing numerous artifacts. The top of this stratum appeared to be a buried A horizon. The contact point between Zone 2 and the overlying Zone 1 produced a significant number of large sherds. Furthermore, several broken vessels, as well as features and burials, were encountered in the lower portion of this stratum. Zone 2 varied in thickness. Along the western boundary of the cemetery, it was less than 10 cm thick. In the area near the burials, this stratum was as much as 60 cm thick.
Zone 3 was culturally sterile. This stratum was a dark yellowish brown sand (10YR 4/6), indistinguishable by color and texture alone from the overlying Zone 2. Zone 3 could be distinguished from Zone 2 by its total lack of cultural material.
1 m2 Excavation units
Ten 1m2 units were excavated within the 15 m by 18 m area of the site with the greatest density of ceramics (see Figures 2 and 3). As Table 2 shows, these test units yielded 2,261 artifacts. Two excavation units, located on the northwest side of the cemetery (194R378 and 195R374), yielded fewer than 100 artifacts each and neither unit produced features or burials. The remaining eight test units produced copious amounts of cultural material, primarily plain shell-tempered ceramics, as well as features and burials.
In most units, a dense cap of shell-tempered sherds was encountered at the contact point between Zones 1 and 2 (the buried A horizon). 196R388, located on the eastern edge of the cemetery, produced a large number of sherds at the interface between Zones 1 and 2 and yielded four discrete ceramic concentrations (Features 6, 7, 8, and 9). 197R379 and 197R380 yielded numerous artifacts, including Moundville Incised var. Snow's Bend and Pensacola Incised var. Bear Point sherds on the buried land surface; Burial One was encountered in these units at ca. 38 cm below datum.
199R379 and 200R379 yielded numerous shell-tempered sherds, including Pensacola Incised var. Bear Point and var. Moore material on the buried land surface. 199R379 also produced Feature 2, a discrete concentration of Pensacola Incised var. Bear Point sherds. Burial Two was encountered in 199R379 and 200R379 at ca. 46 cm below datum.
196R383 and 195R383 yielded a large number of shell tempered sherds at the interface between Zones 1 and 2. Among the ceramics recovered from these units were Pensacola Incised var. Gasque, and Moundville Incised var. Snow's Bend sherds; 195R383 also yielded Feature 5. Burial Three was encountered in 196R383 and 195R383 at ca. 45 cm below datum. 203.5R378.5 produced Features 3 and 4, and a reconstructible Mississippi Plain var. Warrior vessel was encountered on the buried land surface.
Eight ceramic concentrations were given feature designations during testing. In addition, one feature was identified during the Phase I survey. With one exception (Feature 3), the ceramic concentrations appeared to be purposefully broken vessels; one vessel associated with Feature 3 appeared to have been intact prior to testing. All ceramic concentrations contained sherds from more than one vessel. Two concentrations were associated with burials; Feature 2 with Burial Two and Feature 5 with Burial Three. The other ceramic concentrations may also accompany specific burials; however, a definite association with a given interment could not be established during testing. The ceramic material recovered from the features is presented in Table 3.
Feature 1 was a ceramic concentration identified during Phase I (Phillips 1989a). The sherds are from a Pensacola Incised var. Gasque bottle and a Pensacola Incised var. unspecified bottle that resembles Carthage Incised pottery.
Feature 2 was a concentration of Pensacola Incised var. Bear Point sherds primarily from one casuela. Located in 199R379, the concentration measured 26 cm by 40 cm and extended from 16-21 cm below datum. The concentration rested within the upper 5 cm of Zone 2. The bowl evidently represents a grave offering associated with Burial Two and was placed near the surface above the interment.
Feature 3, a ceramic concentration sitting on the original land surface (Zone 2) at approximately 27 cm below datum in 203.5R387.5, was a mostly intact Mississippi Plain var. Warrior jar and part of a Bell Plain var. Hale plate. The feature was approximately 26 cm in diameter and extended from 13-27 cm below datum.
Feature 4, located in the southeast corner of 203.5R378.5, was a concentration of sherds from three vessels: a Moundville Incised var. Moundville jar, a Pensacola Incised var. unspecified jar, and a Mississippi Plain var. Warrior jar. The concentration was located at 20-23 cm below datum and rested within the upper few centimeters of Zone 2, approximately 25 cm southeast of Feature 3. The concentration measured 25 cm by 30 cm.
Feature 5 was a concentration of sherds from a Mississippi Plain var. Warrior jar and a Pensacola Incised var. Gasque vessel. The feature, located in unit 195R383, extended into the north and west walls of the unit. These sherds were concentrated from 13-16 cm below datum at the interface between Zones 1 and 2. These vessels were associated with Burial Three.
Feature 6 was a concentration of Mississippi Plain var. Warrior sherds. No vessels could be reconstructed from the var. Warrior sherds. The feature, located in unit 196R388, rested at the interface between Zones 1 and 2 (16-20 cm below datum) and measured 40 cm by 22 cm.
Feature 7 was located along the south wall of 196R388 and extended into the next unit to the south. It consisted of a concentration of Mississippi Plain var. Warrior, Bell Plain var. Hale, and Pensacola Incised var. Moore sherds.
Feature 8, a concentration of sherds also located in the south wall of 196R388 and the unit to the south, consisted of sherds from a Moundville Incised var. Moundville jar and a Mississippi Plain var. Warrior bottle. Additional var. Warrior sherds and an indeterminate shell-tempered sherd also were found in this concentration. The feature was located on the original land surface (top of Zone 2) at a depth of 17 cm below datum.
Feature 9 consisted of sherds from a Pensacola Incised var. Moore vessel and several Mississippi Plain var. Warrior sherds. This concentration, measuring 20 cm by 25 cm, was located in unit 196R388. The feature extended to a depth of 20 cm below datum and was located at the interface between Zones 1 and 2.
Three burials were located during the test excavations. Due to the extreme acidity of the sediments, bone preservation was extremely poor. Furthermore, burial pit outlines could not be discerned because of the organic leaching caused by this acidity. This made recognition of the burials difficult. Skeletal material could be identified only in situ by the dark brown organic stains produced by decomposition which surrounded the bone.
In addition to the difficulty in identifying the human remains, the lack of observable burial pit outlines made it difficult to associate some mortuary furniture with the burials. Although some grave goods were clearly associated with individual burials, other associations are presumed because of the close horizontal and vertical proximity of artifacts to the human remains. Table 4 presents the artifacts associated with burials; the burial descriptions are presented below.
Burial One is characterized by a dark brown stain (10YR 3/3) that surrounded a partial cranium, mandible, and dentition; no postcranial material was observed (see Figure 5). This interment appeared to be a secondary burial. The dark stain was organic in nature rather than a burial pit, and no burial pit outline could be discerned. The remains measured approximately 20 cm by 30 cm and extended from 38-45 cm below datum.
Several artifacts, including 7 vessels, mica, and 1 red and gray chert point appeared to be associated with Burial One. Among the associated vessels were 2 small Pensacola Incised var. Bear Point subglobular bottles, 1 var. Bear Point beaker, 1 var. Bear Point jar, 1 large var. Bear Point casuela, 1 small Moundville Incised var. Snow's Bend jar, and 1 D'Olive Incised var. Arnica plate (see Figure 6). These vessels evidently were intentionally broken and placed above the burial, on or near the surface. A ceramic bird-head effigy was located approximately 35 cm west of the skeletal material at a depth of 35 cm below datum. A small amount of mica was recovered from excavation unit levels above the remains. The projectile point/knife was recovered at a depth of 29 cm below datum, approximately 61 cm northeast of the burial. A radiocarbon date taken from a carbonized wood sample clearly associated with this burial yielded an uncorrected date of 500 BP +/- 60 years (Beta 30702).
Burial Two appeared as a dark brown organic stain (10YR 3/3) that surrounded cranial fragments, dentition, and other indeterminate skeletal material including possible long bones; bone preservation was very poor (see Figure 7). No burial pit was observed. Burial Two measured approximately 80 cm by 55 cm and extended from 46-56 cm below datum.
Among the grave goods associated with the interment were 3 shell-tempered vessels, 1 whelk columella, 2 greenstone celts, hematite, mica, and 1 projectile point. The three vessels included 1 Pensacola Incised var. Bear Point beaker and 2 var. Bear Point casuelas (see Figure 8). The vessels appeared to have been intentionally broken and scattered on or near the surface above the interment. As Figure 7 shows, 1 greenstone celt fragment was located under the cranium, whereas a second, complete celt was located approximately 20 cm south of the skeletal remains resting at a depth of 44 cm below datum. The whelk columella was located in the cranial area, the hematite was encountered approximately 42 cm southeast of the burial at a depth of 48 cm below datum, and the mica was recovered from excavation units above the burial. The projectile point was situated 20 cm southeast of the interment at a depth of 57 cm below datum.
Burial Three appeared as a dark yellowish brown stain (10YR 4/4) that surrounded very poorly preserved unidentifiable bone (see Figure 9). Grave goods associated with Burial Three included 1 small Moundville Incised var. unspecified jar (Figure 10a), a concentration of sherds from 1 Pensacola Incised var. Gasque vessel (Figure 10b), and a whelk columella. Like the vessels in Burials One and Two, these two vessels appeared to have been intentionally broken; however, the Moundville Incised vessel was placed near the individual rather than on the original land surface.
The test excavations at the Hickory Ridge cemetery produced 3,235 artifacts, including prehistoric ceramic, lithic, and shell artifacts, and historical material. Among these are 3,202 prehistoric pottery sherds (including 30 partially reconstructible vessels), 25 lithic artifacts, 7 shell artifacts, and 1 brick. Descriptions of the Phase II cultural materials are presented below.
Prehistoric Ceramics: Prehistoric ceramics composed 99 percent of the artifact assemblage recovered during Phase II testing. Among these are 3,196 shell-tempered sherds and 6 sand-tempered sherds. In addition to the plain ceramics, surface decorations include incised, engraved, and punctated design elements. The ceramic classification follows several sources including Steponaitis (1978, 1983), Jenkins (1981), and Fuller and Stowe (1982). This classification system was utilized although it is recognized that northern Gulf Coast ceramic typology is the subject of debate and is undergoing revision.
Mississippi Plain var. Warrior
Reference: Steponaitis (1978:15)
Number of sherds: 2,391
Description: Mississippi Plain var. Warrior represents the largest single class of sherds recovered during the investigations. This type is tempered with coarse crushed shell which has leached out leaving a distinctive pock-marked surface; the paste is very sandy. The rims are either straight or flared and appendages include plain loop handles and loop handles with small nodes on top. The vessel forms appear to be primarily globular jars (Figure 11a-b); however, one bottle of this type was recovered. Six reconstructible var. Warrior vessels were recovered. Fuller and Stowe (1982:52-54) place this type within the Bottle Creek phase (A.D. 1200-1450).
Bell Plain var. Hale
Reference: Jenkins (1981:65-66)
Number of sherds: 144
Description: Bell Plain var. Hale is tempered with finely crushed shell and exhibits a burnished surface. Vessel forms recovered from Hickory Ridge include bowls and plates; the rim form is straight. Three animal effigy heads reminiscent of a duck in profile were recovered. One reconstructible vessel of this type was recovered. Bell Plain var. Hale appears to have been used throughout the Mississippian stage, therefore, an association with the Bottle Creek and Bear Point phases is indicated (Fuller and Stowe 1982:50-52).
Moundville Incised var. Moundville
Reference: Jenkins (1981:77)
Number of sherds: 13
Description: Moundville Incised var. Moundville exhibits the same paste as Mississippi Plain var. Warrior. The surface is decorated with incised arches that parallel the rim; a series of short incisions radiate upward from each arch (Figure 12a). This motif occurs on small globular jars with flared rims, often also accompanied by loop handles. Two vessels of this var. were reconstructed. This ceramic type is associated with the Bottle Creek phase (Stowe 1985:149).
Moundville Incised var. Snow's Bend
Reference: Jenkins (1981:78)
Number of sherds: 15
Description: Moundville Incised var. Snow's Bend sherds also exhibit the same paste as Mississippi Plain var. Warrior. Narrow incised arches parallel the rim; a series of small punctations occur above these arches (Figure 12b). The vessel form is a small globular jar with a flared rim (Figure 11c). A variety of adornos occur on these jars, including lugs, nodes, and loop handles. Two var. Snow's Bend reconstructible vessels were located, one of these was associated with Burial One. Moundville Incised var. Snow's Bend is associated with the Bottle Creek phase (Stowe 1985:149).
Moundville Incised var. unspecified
Number of sherds: 10
Description: One partially reconstructible small subglobular jar associated with Burial Three was classified as Moundville Incised var. unspecified (Figure 10a). The vessel has the same paste as Mississippi Plain var. Warrior and the typical Moundville arch. A series of these arches parallel the flared rim. Above these arches are randomly placed circular punctations approximately 7 cm in diameter (Figure 12c). The exact chronological placement of this vessel is uncertain, although its association in Burial Three with a Pensacola Incised var. Gasque vessel, suggests a Bottle Creek phase assignation.
D'Olive Incised var. Arnica
Reference: Fuller and Stowe (1982:57-58)
Number of sherds: 27
Description: One reconstructible D'Olive Incised var. Arnica vessel was recovered from Hickory Ridge in association with Burial One (Figure 6e). The plate is tempered with finely crushed shell and has a burnished surface. The design element, shown in Figure 12d, includes an incised line approximately 2 cm below the notched rim. Beneath this incised line is a series of double incised line arches. Within these arches are two sets of short parallel incisions. Fuller and Stowe (1982:58) associate var. Arnica ceramics with the Bear Point phase (A.D. 1450-1700), though they suggest that the type may have been introduced during the latter part of the Bottle Creek phase.
Pensacola Incised var. Gasque
Reference: Fuller and Stowe (1982:74-75)
Number of sherds: 35
Description: Pensacola Incised var. Gasque is utilized herein to describe all ceramics with the Bell Plain var. Hale paste and realistically rendered incised design elements. Fuller and Stowe (1982:74-77) originally split these realistic design elements into two varieties: Gasque, with an anthropomorphic motif, and Holmes with bird/snake representations. However, they suggested that var. Holmes could be subsumed by var. Gasque and noted that both varieties may be related to Moundville Engraved var. Hemphill. Steponaitis (1983:56) suggested that Moundville Engraved var. Hemphill encompasses a wide array of motifs. Therefore, the Hickory Ridge classification placed the shell-tempered, realistic design elements into a single class, var. Gasque, that parallels Moundville Engraved var. Hemphill.
The var. Gasque motif prevalent at Hickory Ridge is the raptorial bird such as that on a large bottle recovered during Phase I (Phillips 1989a) and shown in Figure 12h. Phase II testing produced part of a second bottle with this motif (Figure 12g). These sherds display the crest and feathers of a raptor rendered with cross-hatching and zone incisions. The ceramics with realistic design elements are associated with the Bottle Creek phase (Fuller and Stowe 1982:75-77).
Pensacola Incised var. Bear Point
Reference: Fuller and Stowe (1982:72-73)
Number of sherds: 188
Description: Pensacola Incised var. Bear Point exhibits the same finely crushed shell tempering and burnished surface as Bell Plain var. Hale. As shown in Figure 13a-c,g, the stylized design elements include abstract curvilinear skull motifs, trilobes, barred ovals, and S-shaped figures. Fuller and Stowe (1982:73) associate the var. Bear Point materials with the Bear Point phase. The Hickory Ridge site produced several var. Bear Point vessel forms. Among these are subglobular bottles, casuelas, and beakers. The rim forms are primarily vertical, flattened, and often notched; however, the small subglobular bottles exhibit a smooth, rounded rim. One of the casuelas has two pointed nodes on the rim. Eight partially reconstructible vessels, including 3 casuelas, 3 bottles, and 2 beakers, were recovered from the cemetery. Pensacola Incised var. Bear Point vessels were associated with two burials: two subglobular jars with Burial One (Figure 6b-c) and 1 beaker and 2 casuelas with Burial Two (Figure 8).
Pensacola Incised var. Moore
Reference: Fuller and Stowe (1982:78-80)
Number of sherds: 21
Description: Pensacola Incised var. Moore is incised with highly stylized design elements (Figure 13d-f). These include nested rectangles and extremely abstract, rectilinear skull motifs. The rim forms are vertical and many be either squared or rounded. No vessel forms could be discerned. This is a Late Mississippian type probably associated with the Bear Point phase (Fuller and Stowe 1982:78-80).
Moundville Engraved var. unspecified
Reference: Steponaitis (1983:54)
Number of sherds: 24
Description: Hickory Ridge produced 1 partially reconstructible Moundville Engraved-like vessel with a Bell Plain paste, and 1 nonassociated rim sherd. The vessel is a small subglobular bottle with a pedestal base and a straight rounded rim (Figures 12f and 14a). The design elements include five narrow engraved lines, a series of small arches below these lines, and a series of cross-hatched triangles above these lines. In addition, a series of small arches parallel the collar of the vessel. These arches are scalloped and exhibit a cross-hatched interior. Stowe (1985:149) places Moundville Engraved ceramics in the Bottle Creek phase.
Pensacola Incised var. unspecified
Number of sherds: 15
Description: This partially reconstructible jar with a four-pointed or undulating rim could not be placed in a particular variety of Pensacola Incised (Figures 12e and 14b). This vessel exhibited a burnished surface and a paste typical of Bell Plain var. Hale. The modified scroll decoration consists of two incised lines that parallel the rim. Radiating from these incisions are five more parallel curvilinear lines that encircle the vessel. The interior of this motif is decorated with a series of small punctations. The chronological position of this vessel is uncertain.
Number of sherds: 313
Description: Several shell-tempered sherds were recovered that were too small to classify. These were decorated with a variety of incised and punctated designs. Among these are narrow and wide incisions, fingernail punctations, and reed punctations. Pastes that are typical of Mississippi Plain var. Warrior and Bell Plain var. Hale pastes are represented.
Number of sherds: 6
Description: Six undecorated sherds tempered with coarse sand were recovered. Sand tempering enjoyed a long usage that spanned the Woodland and Mississippian stages. Lacking surface decoration, no additional chronological statements are offered.
Lithics: A total of 25 lithic artifacts was recovered during Phase II testing. Seven raw material types are represented: chert, novaculite, Tallahatta quartzite, greenstone (chlorite schist), mica, hematite, and sandstone. Two projectile point/knives were recovered from burial contexts. One, associated with Burial Two, is a small triangular-shaped novaculite point with one shallow side notch. The second, associated with Burial One, is a red and gray banded chert projectile point/knife with wide shallow side notches, an excurvate base, and serrated blade edges. Two greenstone (chlorite schist) celts were associated with Burial Two. One measures 27 cm long by 7.4 cm wide by 4.5 cm thick. The second, broken before disposal, measures 6.7 cm long by 4.2 cm wide by 3 cm thick. A small amount of mica (N=11) in the form of very small flakes was recovered in the levels above Burials One and Two. In addition, Burial Two yielded 1 chunk of fine-grained red hematite. Three tertiary Tallahatta quartzite flakes, 1 tertiary chert flake, and 2 sandstone flakes complete the lithic assemblage.
Shell Artifacts: Seven shell artifacts were recovered during testing. Three partial whelk (Busycon contrarium) columellae were recovered; 1 from an excavation unit, 1 from Burial Two, and 1 from Burial Three. Testing also yielded 4 small unidentifiable shell fragments. One of these exhibits cross-hatched striations on one surface.
Hickory Ridge evidently was used exclusively for burial of the dead during the Mississippian period. No midden deposits or other indications of long-term occupation were encountered. Phase II testing revealed three burials and indicated that the cemetery measures about 15 m by 18 m. Cultural material, including numerous partially reconstructible vessels, was densely concentrated within this small area. The investigations indicated that the cemetery has not been significantly disturbed. This was documented primarily by the horizontal distribution of fitting ceramics; sherds from the same vessel were most often recovered in very close proximity to one another. The vertical distribution and large size of the sherds also underscore the integrity of the site. Intact or nearly intact vessels were positioned within a few centimeters of the present land surface. For example, an intact Mississippi Plain var. Warrior jar rested on the buried land surface and extended upward into the overlying wind-blown sand. This vessel was located approximately 15 cm below the present surface.
An array of grave offerings accompanied the three burials (see Table 4). The mortuary furniture included whelk columellae, projectile points, greenstone celts, mica, and hematite. The archaeological context of these materials suggests that they were placed in the burial pits with the human remains. In addition to these grave goods, each of the three burials was interred beneath several shell-tempered vessels. These vessels evidently were ceremonially broken and left on or near the surface after burial. Forming a dense cap over the cemetery, this material was located on or near the contact point between the wind-blown sands of more recent origin and the buried land surface. Some of these vessels appeared archaeologically as discrete concentrations of large sherds directly above the burials, whereas other vessels formed part of the dense cap. Nine concentrations of ceramics that included 15 reconstructible vessels were documented; discrete concentrations of broken vessels were located directly over Burials Two and Three.
Similar Mississippian burial practices have been reported elsewhere in northwest Florida. For example, Moore (1901, 1918) noted the occurrence of secondary burials, ceremonially killed vessels, and dense concentrations of sherds on several Mississippian cemetery sites in northwest Florida. At the Naval Live Oaks cemetery (8SR36) in Gulf Breeze, Florida, 10 miles east of Hickory Ridge, Lazarus et al. (1967:104) reported that "the ceremonial procedure seems to have been the same pattern as evident in other ceremonial sites--human remains deposited in the lower level, some fire, some offering, crowned by a layer of sherds near the perimeter." They also noted that the sherds were concentrated between 3 in. and 18 in. below the surface and that the "offering" was either intact or reconstructible.
The mortuary furniture associated with the Hickory Ridge burials strongly suggests that these were high status individuals, at least in a local sense. A number of the grave offerings are exotic in origin. The raw material source for the celts (chlorite schist), for example, is found in the Carolina Piedmont. Novaculite comes from Arkansas, whereas the red and gray chert point found with Burial One appears to be either Tuscaloosa gravel or Citronelle gravel from the interior Gulf Coastal Plain. The ceremonial nature of some of the grave offerings also indicates high status. The raptorial bird motif, the celts, and the whelk columellae appear at Mississippian ceremonial centers throughout the Southeast (for example, Moundville, Lake Jackson, and Etowah). Milanich (1994:374-375) notes that these symbols were restricted to the elite. In contrast, lower status Mississippian burials often have little or no mortuary furniture. Given the size and isolation of the cemetery, the small number of individuals interred within it, and the exotic and symbolic nature of the grave offerings, Hickory Ridge may have been the burial place for the local elite.
The Hickory Ridge ceramic assemblage exhibits the characteristics of both Bottle Creek and Bear Point. The ceramic decorative elements run the gamut from realistically rendered Bottle Creek designs to abstract Bear Point motifs. Moundville influences also can be seen in this ceramic assemblage. Some of the Pensacola Incised var. Gasque material is very similar to Moundville Engraved var. Hemphill (cf.Steponaitis 1983:56). The small Moundville Incised jars, Moundville Engraved bottle, and a Carthage Incised-like vessel also indicate connections with this very large Mississippian site in central Alabama.
In summary, the investigations revealed that the Hickory Ridge cemetery contains an undisturbed Mississippian cemetery that probably dates from the fifteenth century A.D. A carbonized wood sample associated with Burial One yielded an uncorrected radiocarbon date of 500 +/- 60 years B.P. (Beta 30702) or about A.D. 1450 (1390-1510). The radiocarbon date and the ceramic assemblage indicate a transitional late Bottle Creek-early Bear Point phase component. The site appears to be an isolated cemetery associated with 8ES1052, a Pensacola village site located nearby. Based on the types and quantities of associated artifacts, the three burials located by testing appear to be those of high status individuals. The pristine nature of the deposits, and the richness and diversity of the artifact assemblage, affords Hickory Ridge unique status among Mississippian cemeteries in northwest Florida.
The author gratefully acknowledges the support and assistance of several organizations and individuals. The archaeological investigations were funded by the Escambia County Utilities Authority, and Landfall Development. A note of thanks is extended to Mr. Mike Green of Landfall for supporting and encouraging this work. Thanks also to Dr. Jim Miller, State Archaeologist and Chief, Florida Bureau of Archaeological Research for his advice and guidance. A special note of gratitude is offered to Dr. Judy Bense for editorial advice and support, to Mr. Lee McKenzie for his superb graphics, and to Dr. M. J. Smith for her graphical contributions. Thanks are due to Ms. Jenny Yearous and Ms. Alice Harris for the ceramic motif artwork, and to Mr. Dan McLeod for the artifact photography. Thanks also go to Mr. Bill Baxter, Ms. Amy Carruth, Mr. Warren Carruth, and Mr. John Wright for assistance in the field, and to the legion of student volunteers for ceramic reconstruction.
Bense, Judith A. 1994 Archaeology of the Southeastern United States: Paleoindian to World War I. Academic Press, New York. Brose, David S., and George S. Percy 1978 Fort Walton Settlement Patterns. In Mississippian Settlement Patterns, edited by Bruce D. Smith, pp. 81-114. Academic Press, New York. Carlisle, Victor W. 1960 Soil Survey of Escambia County, Florida. U.S. Department of Agriculture, Soil Conservation Service, Washington, D.C. Fuller, R. S. 1985 The Bear Point Phase and the Pensacola Variant: The Protohistoric Period in Southwest Alabama. The Florida Anthropologist 38: 150-155. Fuller, R. S., and N. R. Stowe 1982 A Proposed Typology for Late Shell Tempered Ceramics in the Mobile Bay/Mobile Tensaw Delta Region. In Archaeology in Southwest Alabama: A Collection of Papers, edited by C. Curren, pp. 45-94. Alabama Tombigbee Regional Commission, Camden, Alabama. Holmes, William H. 1903 Aboriginal Pottery of the Eastern United States. Bureau of American Ethnology, Annual Report 20. Jenkins, Ned J. 1981 Gainesville Lake Area Ceramic Description and Chronology. Archaeological Investigations in the Gainesville Lake area of the Tennessee-Tombigbee Waterway, prepared for the U. S. Army Corps of Engineers, Mobile District. University of Alabama, Office of Archaeological Research, Reports of Investigations No.12. Moundville, Alabama. Knight, Vernon J., Jr. 1984 Late Prehistoric Adaptations in the Mobile Bay Region. In Perspectives on Gulf Coast Prehistory, edited by Dave D. Davis, pp. 198-215. University Press of Florida, Gainesville. Lazarus, Yulee, W. C. Lazarus, and D. W. Sharon 1967 The Navy Live Oaks Reservation Cemetery Site, 8Sa36. The Florida Anthropologist 20:103-117. Milanich, Jerald T. 1994 Archaeology of Precolumbian Florida. University Press of Florida, Gainesville. Moore, C. B. 1901 Certain Aboriginal Remains of the Northwest Florida Coast, Part I. Journal of the Academy of Natural Sciences of Philadelphia Vol. XI. 1918 The Northwest Florida Coast Revisited. Journal of the Academy of Natural Sciences of Philadelphia Vol. XVI. Phillips, John C. 1989a Phase One Cultural Resources Survey of Warrington Effluent Diversion Project Disposal Facility for Escambia County Utilities Authority. University of West Florida, Institute of West Florida Archaeology, Report of Investigations No. 23. Pensacola. 1989b Archaeological Testing of the Hickory Ridge Site (8ES1280). University of West Florida, Institute of West Florida Archaeology, Report of Investigations No. 26. Pensacola. Phillips, Philip 1970 Archaeological Survey in the Lower Yazoo Basin, Mississippi, 1949-1955. Papers of the Peabody Museum of Archaeology and Ethnology 60. Harvard University, Cambridge, Massachusetts. Steponaitis, Vincas P. 1978 Some Preliminary Chronological and Technological Notes on Moundville Pottery. Paper presented at the 35th Southeastern Archaeological Conference, Knoxville, Tennessee. 1983 Ceramics, Chronology and Community Patterns: An Archaeological Study at Moundville. Academic Press, New York. Sternberg, G. M. 1876 Indian Burial Mounds near Pensacola, Florida. Proceedings of the American Association for the Advancement of Science 24(2):282-292. Stowe, Noel R. 1985 The Pensacola Variant and the Bottle Creek Phase. The Florida Anthropologist 38:144-149. Tesar, Louis D. 1980 Leon County Bicentennial Survey Project: An Archaeological Survey of Selected Portions of Leon County, Florida. Florida Department of State, Division of Archives, History and Records Management, Bureau of Historic Sites and Properties, Miscellaneous Project Report Series 49. Walker S. T. 1885 Mounds and Shell Heaps on the West Coast of Florida. Annual Report, Smithsonian Institution for 1883: 854-868. Willey, Gordon R. 1949 Archeology of the Florida Gulf Coast. Smithsonian Miscellaneous Collection Volume 113. Wimberly, Steven B. 1960 Indian Pottery from Clark County and Mobile County, Southern Alabama. Alabama Museum of Natural History Museum Paper 36, University of Alabama, Tuscaloosa. John C. Phillips Archaeology Institute University of West Florida Pensacola, FL 32514 Table 1. Cultural Material Recovered from Shovel Tests Pensacola Incised var. Bear Point - 19 Tallahatta Quartzite Flake - 2 Pensacola Incised var. Gasque - 5 Sandstone Flake - 1 Moundville Incised var. Moundville - 3 Moundville Incised var. Snow's Bend - 2 Fauna D'Olive Incised var. Arnica - 25 Modified shell - 1 Bell Plain var. Hale - 81 Unmodified shell - 1 Mississippi Plain var. Warrior - 432 Shell-tempered indeterminate - 98 Brick - 1 Total Artifacts - 671