Center For Transportation Research

Hancock County



A background literature search on the topic of Appalachian burial practices was conducted in the John Hodges Library of The University of Tennessee. Secondary sources and primary (archival) documents specifically relative to Hancock County history and settlement were examined in the Hancock County courthouse, Hancock Public Library, McClung Collection of the Knox County Public Library, and the Special Collections Library of The University of Tennessee Library. Oral history and/or ethnographic interviews about the Cool Branch Cemetery site (40HK9) were conducted in Hancock County with Brown Johns, Phoebe Seals Greene, and Scott Collins.


Phase II testing procedures at sites 40HK5 and 40HK6 included clearing, plowing, and disking the areas to be investigated on each site. The two sites were gridded at 10 m intervals. A contour map was generated for each site utilizing an electronic total station, graphic expandable calculator with survey card, and Tripod Data Systems, Inc. Easy Survey Plus computer software. A controlled surface collection was then conducted on each site. Two 1 m x 1 m test units were hand excavated on each site to determine the subsurface stratigraphy and to recover a sample of cultural material. Machine excavated 2 m wide transects were then placed at 10 m intervals across each site. A sample of exposed subsurface pit features and postholes was excavated.

Additional Phase II testing consisted of machine excavation of block areas on sites 40HK5 and 40HK6, excavation of exposed subsurface pit features, bisecting of all potential postholes, and excavation of 1 m x 1 m units in the subplowzone soil horizon containing cultural material. All pit features located were photographed and mapped in plan view. The pits were excavated in halves and by natural strata. One-half of each pit feature was excavated to define the size, shape, and any variation (zones) within the fill. The exposed profile wall was mapped in the 1:10 scale and photographed. The remaining half of each pit was excavated and all relevant data were recorded. During the excavation the fill was measured with buckets calibrated in liters to determine the total volume of each pit. Samples of fill for flotation (10 liters) were collected from arbitrary 10 cm levels established in one-half of each pit and the remaining fill was waterscreened. Samples of pit fill were retained for flotation to maximize the recovery of floral and faunal material.

A subplowzone soil horizon containing cultural material was identified on site 40HK6. Six 1 m x 1 m units were hand excavated in the deposit in 10 cm levels. A 10 liter flotation sample was collected from every level of half of the units and the remaining fill was waterscreened. The unexcavated portion of the soil horizon in the right-of-way was removed with a backhoe to locate features intruding into the subsoil beneath the deposit.

The postholes were cross-sectioned with a shovel and trowel. The profile was viewed with approximately 10 cm of subsoil bordering the posthole fill. Dimensions of the postholes along with a sketch of each profile were recorded. Posthole fill was not waterscreened or retained for flotation.

Fill to be waterscreened was processed through a system consisting of paired upper and lower screen boxes. The upper box was lined with 6.4 mm (¼ inch) mesh hardware cloth and the lower box was lined with 1.6 mm (1/16 inch) mesh hardware cloth. The fills from selected features on site 40HK5 were only waterscreened through 6.4 mm mesh hardware cloth. Water was pumped to the apparatus to separate the soil from the residue. The 6.4 mm and 1.6 mm residue fractions were recovered, dried, and stored for later analysis.

The 6.4 mm waterscreened fractions were completely sorted. The cultural material was divided into several categories and noncultural gravel was discarded. The sorted cultural material from the 6.4 mm waterscreened fractions was analyzed with the trowel sorted material. The 1.6 mm waterscreened fractions were sampled for analysis.

The archaeological investigation of site 40HK9 included the clearing of brush and small trees on the entire site. A contour map was then generated for the site. All potential graves markers and grave depressions were photographed and mapped in plan view in the 1:20 scale. A block area was machine excavated on the site in the proposed right-of-way to expose unmarked graves. The unmarked graves located were photographed and mapped in plan view.

The fill in the shaft of one of the graves was dryscreened through 6.4 mm mesh hardware cloth and the fills in the remaining grave shafts were trowel sorted. The subfeatures of subadult individuals were floated while the subfeatures of adult individuals were waterscreened through 6.4 mm and 1.6 mm mesh hardware cloth. Flotation samples were collected from the pelvic areas of the adult individuals. The graves were profiled in the 1:10 scale and photographed after excavation. Artifacts noted during the excavation of the subfeatures were piece plotted and included in the 1:20 scale map(s) of each grave.



Debitage from feature context on sites 40HK5 and 40HK6 was subjected to a limited mass analysis. Attributes recorded were raw material type, cortex cover, flake portion, thermal alteration, size grade, and weight.

All debitage was size graded by passing the material through a series of nested wire screens. Material was passed through three screens ranging in size from 19.1 mm (¾ inch), 12.7 mm (½ inch), to 6.4 mm (¼ inch). The debitage was weighed using a digital scale. Weight was recorded in grams to the nearest 0.01 g. Cortex was described by the amount present and type. The amount of cortex present on complete flakes and cores was recorded as no cortex, <25% cortex cover, or >25% cortex cover. Cortex cover decreases with the stage of reduction (Magne and Pokotylo 1984; Magne 1985). Cortex types consisted of matrix residual, waterworn, and incipient fracture plane. Matrix residual cortex was identified by a thick chalky exterior surface or rough appearance. Waterworn is the result of tumbling action in a stream or river. It is characterized by a dense, hard, often brown stained appearance with rounded or smooth edges. Incipient fracture planes have flat angular surfaces where the raw material has fractured along a natural cleavage. Debitage raw material was recorded as to geologic rock type (e.g., chert, jasper, chalcedony, quartz), parent geological formation (e.g. Knox chert) and, when possible, a variety of the parent geological formation (e.g. Knox Oolitic chert, Knox Porcelaneous chert). Thermal alteration was recorded as to presence or absence. Thermal alteration has taken place when one or more of the following attributes are observed - color change, increased luster, and heat fractures (pot lids, crenelation, crazing). Pot lids, crenelation, and crazing are interpreted as unintentional products of thermal alteration. Debitage categories consist of complete flakes, flake fragments or broken flakes, and non-flake debris. Complete flakes have a platform, a bulb of percussion, and a distal terminus. Flake fragments or broken flakes are lacking a platform and bulb of percussion or a distal terminus or all of these attributes. Non flake debris consists of blocky debris and cores. Blocky debris are the spent nuclei left from producing flakes. These show no basic flake morphology (platform, ripple marks, force lines, or bulb of percussion) and have no evidence of further utilization or retouch. Cores were used for the production of flakes and have negative flake scars and no evidence of further use or retouch.

PPKs from controlled and general surface collections, power units, test units, piece plots, and feature context were analyzed. Morphological attributes of the blades, stems, and bases of the PPKs were recorded.

PPKs were then compared to previously identified clusters found in the Ridge and Valley physiographic province as well as surrounding regions. Primary sources of comparison were Chapman (1978, 1979, 1981, 1985), Justice (1987), Lafferty (1981) Bentz ed. (1989, 1989), and Cridlebaugh (1977). Early Archaic through Late Woodland PPK clusters were represented among PPKs recovered from sites 40HK5, 40HK6, and 40HK7.

Bifaces and flake tools were noted and inventoried but not analyzed in detail.

Burned or firecracked rock was weighed. Most of it was discarded in the field.


The prehistoric ceramics were initially classified and quantified by the tempering agent. Sherds less than 1.27 cm (½ inch) in size were classified only by temper type. Surface treatments, sherd thicknesses, and temper characteristics were then determined. The surfaces and cores of the sherds were color coded with the Munsell Soil Color Charts (1973). To simplify the color coding system, the various hues, values, and chromas of a color were combined and only the verbal description of the color was noted. Ceramic weights were used in the comparison of various pottery types.

Botanical Remains

Twenty individual flotation samples representing 216 liters of fill from four features at 40HK5 and 14 samples representing 140 liters of fill from two features at 40HK6 were sorted and analyzed. Employing established procedures (e.g., Crites 1978; Kline et al. 1982) for laboratory processing of flotation samples, each was placed in a nested series of standard geologic sieves with mesh sizes of 2 mm, 1 mm, and 500 µm. The screens were underlain with a catch basin. Samples placed in the screens were gently shaken in order to size-grade the material.

Material retained in the >2 mm size class was sorted into constituent categories such as nut remains, seeds, fleshy fruits, and wood charcoal. Where possible, remains were identified to species, genus, or family, then counted and weighed. Material retained in the 1 mm and 500 µm screens was scanned for the presence of any types of fruit remains not observed in the >2 mm size fraction (e.g., grass/forb/drupe seeds and Cucurbita rind remains). Any such remains were identified, counted, and weighed by taxon. The presence of tiny nutshell remains or wood charcoals was noted on the laboratory data sheet but these materials were not quantified. Plant materials remaining in the two smaller mesh screens and catch basin after sorting were weighed as a single sample constituent - sample residue.

Sampling for wood charcoal identification was accomplished by placing wood charcoals from the 2 mm mesh screen onto a 1 cm² grid and selecting fragments from alternate vertical and horizontal grids until 30 fragments (if that many were present and preserved well enough to present necessary anatomical landmarks) were identified to at least the genus level. Some features or zones within features contained more than one excavation level. In such cases, 30 identifications were attempted for each level. Consequently, some feature samples yielded more identifications than did others.

Identification of fragments was confirmed primarily by comparison with extensive comparative collections housed in the ethnobotany laboratory at the Frank H. McClung Museum, The University of Tennessee, Knoxville. Secondary sources for identification purposes included standard commonly employed technical manuals (e.g., Core et al. 1979; Martin and Barkley 1961; Panshin and de Zeeuw 1980).

Hancock County Page Contents:
Continue in Hancock County Page
The Flat Gap Site (40HK5)
The Wilder Site (40HK6)
The Johns Site (40HK7)
Hancock County history
The Cool Branch Cemetery (40HK9)
Southern Appalachian Burial Customs

Click below for general resources.
Prehistoric and Historic Background
References Cited

Return to Projects Page.
Go forward in the Archaeology Group Site.

This page was last updated on 20 Jan 2002.

For comments please send email to Chuck Bentz