The Henry Chapman Site (41SM56)

In the mid-1950s, Sam Whiteside conducted excavations at the Henry Chapman site, his site P-5 (41SM56), on Prairie Creek in eastern Smith County, Texas, some 18 miles east of Tyler, Texas. Mr. Whiteside was an amateur archaeologist who discovered and explored numerous sites up and down Prairie Creek as well as other important sites in Smith and adjoining counties. The Chapman site was one of the first sites to be investigated by Mr. Whiteside, and the major part of the work took place there in 1957 and 1958. In August 1957, the site was visited by Edward B. Jelks and Leroy Johnson, who viewed the excavations and examined artifacts collected from there. Mr. Jelks, in notes on record at the Texas Archeological Research Laboratory (TARL) at The University of Texas in Austin, Texas, on August 21, 1958, indicated after observing the artifacts from the Henry Chapman site that the ceramics indicated an Alto focus occupation. He recognized Hickory Engraved and Holly Fine Engraved wares as well as Dunkin Incised and possible Pennington Punctated-Incised sherds. He mentioned that no Crockett Curvilinear Incised or Weches Fingernail Impressed sherds were present (two common types present at the Alto focus type site, the George C. Davis site [41CE19]). E. Mott Davis also visited the site in late 1957, and in February 1959, Davis, Lathel F. Duffield, and William A. Davis collected skeletal material from the site. In a June 15, 1958, newsletter from the East Texas Archeological Society (ETAS), Mr. Whiteside described excavations at the Henry Chapman site that had revealed refuse pits, post holes, and an abundance of pottery sherds indicative of an Alto focus occupation. In the summer of 1983, Mr. Whiteside visited the University of Texas Field School at the George C. Davis site, where he loaned some of his notes and collections from the Chapman site for recording purposes (these are on file at TARL). After Mr. Whiteside’s death, his family graciously allowed me access to his notes and artifacts. Not all of the artifacts have survived, having been stored in paper bags and subjected to several moves, but thanks to Mr. Whiteside’s writing lot numbers on artifacts and listing them in a journal, a majority of them have survived, and thus are suitable for study. Although 50 years have passed since Mr. Whiteside’s work at the Henry Chapman site, it is my intent to now make that work public.


INTRODUCTION
In the mid-1950s, Sam Whiteside conducted excavations at the Henry Chapman site, his site P-5 (41SM56), on Prairie Creek in eastern Smith County, Texas, some 18 miles east of Tyler, Texas. Mr. Whiteside was an amateur archaeologist who discovered and explored numerous sites up and down Prairie Creek as well as other important sites in Smith and adjoining counties (Walters 2005). The Chapman site was one of the first sites to be investigated by Mr. Whiteside, and the major part of the work took place there in 1957 and 1958. In August 1957, the site was visited by Edward B. Jelks and Leroy Johnson, who viewed the excavations and examined artifacts collected from there. Mr. Jelks, in notes on record at the Texas Archeological Research Laboratory (TARL) at The University of Texas in Austin, Texas, on August 21, 1958, indicated after observing the artifacts from the Henry Chapman site that the ceramics indicated an Alto focus occupation. He recognized Hickory Engraved and Holly Fine Engraved wares as well as Dunkin Incised and possible Pennington Punctated-Incised sherds. He mentioned that no Crockett Curvilinear Incised or Weches Fingernail Impressed sherds were present (two common types present at the Alto focus type site, the George C. Davis site [41CE19]). E. Mott Davis also visited the site in late 1957, andin February 1959, Davis, Lathel F. Duffield, andWilliam A. Davis collected skeletal material from the site. In a June 15, 1958, newsletter from the East Texas Archeological Society (ETAS), Mr. Whiteside described excavations at the Henry Chapman site that had revealed refuse pits, post holes, and an abundance of pottery sherds indicative of an Alto focus occupation.
In the summer of 1983, Mr. Whiteside visited the University of Texas Field School at the George C. Davis site, where he loaned some of his notes and collections from the Chapman site for recording purposes (these are on file at TARL).
After Mr. Whiteside's death, his family graciously allowed me access to his notes and artifacts. Not all of the artifacts have survived, having been stored in paper bags and subjected to several moves, but thanks to Mr. Whiteside's writing lot numbers on artifacts and listing them in a journal, a majority of them have survived, and thus are suitable for study. Although 50 years have passed since Mr. Whiteside's work at the Henry Chapman site, it is my intent to now make that work public.

SETTING
There are several Prairie Creeks in Smith County. This particular Prairie Creek has its headwaters in central Smith County some 10 miles east of Tyler. Prairie Creek then flows past the Chapman site about 10 miles in an easterly direction, joining the Sabine River near present day Gladewater, Texas. At the location of the Chapman site, the Prairie Creek floodplain is approximately 1/4-mile wide. The present day Prairie Creek channel is located on the opposite side of the floodplain from the site, but channel scars and meander are evidence that it has migrated back and forth across the floodplain in the past.
The Chapman site is located on a low sandy terrace some 8 ft. above the Prairie Creek floodplain. The terrace is approximately 100 x 160 ft. in size, covering some 16,000 ft. 2 , or 0.4 acres. A spring branch forms the northern boundary of the site; the west side ends at a steep hillside; and the east and south boundaries are the Prairie Creek floodplain.
The landowner reported that the site was under cultivation for a two year period around 1940. After that it was allowed to return to native woodlands, and it was harvested for timber shortly before Mr. Whiteside began excavations in 1957. For those unfamiliar with East Texas, nature here abhors empty spaces, rushing in at its first opportunity to cover the bare spots with all matter of vegetation.
Only after successive generations of plant growth have evolved do the woods reach enough height to crowd out some of the brush and make the woods fairly open again. 1bis mass of vegetation caused Mr. Whiteside to write in his notes at one point in bold letters "excavation ver y difficult."

EXCAVATIONS
Mr. Whiteside discovered the Chapman site in the winter of 1957 and obtained landowner permission to conduct excavations there. He noted that there was a lack of surface visibility due to the previously mentioned mass of vegetation and he also noted that there were few artifacts present in the first 0.5 ft. of excavations. It is not known whether this sterile layer was the result of soils being deposited on the site sometime after the prehistoric occupation ended, or if it is the product of the action of earthworms bringing sterile soil to the surface over a number of years.
After establishing a grid system over the site in 5 x 5 ft. squares, Mr. Whiteside excavated a number of trenches across the site, expanding them into a series of contiguous 5 x 5 ft. excavation blocks when cultural materials were encountered. Mr. Whiteside used feet and inches for measurements, that being the accepted practice at the time. Shovels, trowels, and a screen (probably with a 112-inch mesh) were used in the excavations. At times, when working alone, Mr. Whiteside stated in his notes that he did not screen all of the soil but relied on shovel-skimming, taking thin slices with the shovel or trowel to recover artifacts and expose features. Artifacts were collected in 6 inch (0.5 ft.) intervals within each 5 x 5 ft. excavation square. After washing, the recovered artifacts were given an individual catalog number (or lot number as used in the report) corresponding to their depth below the surface and North and East grid coordinates. A plan view was made depicting the excavation areas and the locations of identified features (most of the time  (Figure 1). In his notes, he indicated that excavations revealed three roughly circular 20-30 ft. diameter areas (I, II, and III) marked by increased artifact concentrations and darker carbonstained soils (Figure 1). He surmised that these areas represented circular house locations . Area I had the deepest midden deposits along with a large trash pit (Feature 2). Area II was marked by increased artifact densities; an arc of possible post holes (Feature 11 ); and several pit features. Soils were thinner in Area II and post holes were detected only when they extended into the clay subsoil. Area III was suggested to represent the location of a burned structure because of the presence of a fired mud dauber nest with grass and reed impressions on one side. Such nests are often constructed in protected locations such as inside structures (Walters 2008:66). There were also pieces of daub with grass/cane impressions and portions of a "green" unfired vessel that had been distorted by intense heat (Figure 2). There were no features identified in Area III, however.
Mr. Whiteside described a typical soil profile at the Chapman site as: Zone 1, a dark brown (1 OYR3/3) to very dark brown (10YR2/2) organically enriched loamy fine sand that varies from 1 ft. to 1.5 ft. bs. in depth. Area I is described as having the thickest Zone 1 deposits. Most of the prehistoric artifacts were recovered from this zone. At places, this zone is capped by 0.5 ft. of a sterile overburden attributed to slope-wash; Zone 2, this is a yellowish-brown (10YR5/4) loam y fine sand that varies from 0.75-2 ft. in depth. Few artifacts were recovered from this zone; and, Zone 3, a sticky, yellowish-red (5YR5/6) sterile sandy clay that varies in depth across the site. In Area II, the Zone 3 soils were closest to the surface, making features in this area more distinct because of the distinct soil color changes between Zones 112 and the sandy clay.

FEATURES
Mr. Whiteside made note of several features discovered during excavations at the Chapman site,  some in greater detail than others. Most of these features, excluding the two burials, were associated with the three areas (I-III) that Mr. Whiteside identified as possible house locations with associated pits and hearths (see Figure 1). Large pits are common on early Caddo sites and probably represent food storage facilities reused as trash receptacles after they were emptied of their intended contents. Hearths occur inside structures or in outdoor activity areas. One set of features in Area II was a partial post hole pattern from a Caddo house. Of the two burial features, one was a partial cremation that was associated with the Caddo occupation; the cultural affiliation of the other burial has not been determined.

~·.······
Feature 1 (N105-110 E20-25) This is a pit feature that is 2 x 2.5 ft. in diameter and 2.5 ft. deep containing 56 sherds; four partially reconstructed vessels (see whole vessels); animal bone, including a deer skull fragment with attached antler; deer teeth and vertebra; mussel shell; flakes; and a smoothing stone. The sherds included two that are engraved, one an engraved rim with opposing diagonallines; 30 incised sherds, including two rims; and 24 plain sherds, including two plain rims.
Feature 2 (N75-80 E0-5, 2ft. bottom depth) The feature is a dark charcoal-rich stain (Figure 3b) that is approximately 3 ft. in diameter and extended to 3 ft. bs. Artifacts included in the fill are flakes, sherds (three incised, one engraved, and three plain body), and animal bones. Charred animal bone collected from 0.5-1.0 ft. bs in the feature fill was submitted to The University of Texas at Austin on February 28, 1958, for a radiocarbon date, but were never actually submitted for radiocarbon dating. Records at TARL show the specimens were discarded on August 21, 1969, without being analyzed. On November 25, 1957, in a trench (N60-70 W 15-20) excavated by Ben Tolbert and Roscoe Ford, a charcoal stain was noted at 2.5 ft. depth. Excavations exposed a 3.5 ft. diameter pit that extended to 3ft. bs with charred logs and human bone underneath the charred logs ( Figure 4). A skull was on the north side of the pit, on its left side facing to the east. The lower 0.67 ft. of the burial pit was ve1y black from charcoal staining. The lower part of the burial pit also had clay mottles that were the result of the pit having been dug into the clay subsoil. A clear pit outline could not be determined above 2.5 ft. bs in the dark brown cultural zone, but below that depth a circular pit approximately 3.5 ft. in diameter was observed. Artifacts in the pit fill indicated that the pit had been dug into and through an existing Caddo occupational deposit.
According to Mr. Whiteside's journal, he turned over skeletal material, including a cranium, from the Hemy Chapman site on February 24, 1959, to E. MottDavis. E. Mott.Davis's personal journal notes on Tuesday, Februa1y 24, 1959, that '\ve (E. M. Davis, L. F. Duffield, and W. A. Davis) drove to Sam Whiteside's at Tyler. .. Sam gave us ... skeletal material from his P-5 site." There were other scattered human bones in the pit, but it is not clear from the notes if all of the bone was charred. Records from TARL indicate that there are two human remain entries from the Chapman site but it is not conclusive at this time if these remains are actually from the Chapman site.
A 0.5 ft. diameter charred log lay east-west across the head/neck area of the grave, and two samples were collected for radiocarbon dating, although the exact whereabouts of these samples are not known. A second log, not as well preserved, lay in a north-south direction on the west side of the pit. Whiteside collection and was submitted by the author for radiocarbon dating, yielding a 2 sigma calibrated age range of AD 1280-1440. There were no obvious grave goods associated with the burial. Mr. Whiteside speculated in his notes that after the body had been placed in the pit, a fire was built over the body, partially consuming it and charring the bones. The pit was then refilled with soil that was used to extinouish c the fire before the logs were completely oxidized. Caddo ceramics in the burial fill of Feature 3 include four incised body sherds; one incisedpunctated rim with diagonal incised lines separating triangular zones of tool punctates; one rim with broad excised horizontal bands and red pioment· c ' and three plain body sherds. There were also two ferruginous sandstone slabs in the burial fill. They were unmodified, and measured 5 x 4 em in lenoth c and width, and were 1.0 em thick. This is a circular-shaped pit, although the exact dimensions are not clear. Based on the profile, it appears to have been ca. 9 ft. in diameter, with a 4 ft. diameter portion that extended to 3 ft. bs, well into the clay subsoil (see Figure 3a). The deeper portions of the pit feature were first identified at 1 ft. bs, and found in it was one rim sherd with diagonal incised lines, an incised body sherd, and three plain body sherds. The shallow portion of the pit contained most of a Dunkin Incised jar ( Figure 5), with a bottom depth of 1.5 ft. bs.

Walters-The Henry Chapman Site 17
Feature 5 (N150-155 E65-70, Bottom depth is 2ft. bs) This is a possible burial feature in Area III (see Figure 1). In Mr. Whiteside's journal, lot numbers 1056-1082 are listed as B2 with artifacts collected from 0-2 ft. However, there is no description of these artifacts. At TARL, there is a long bone and adultsized skull fragment (about 2 x 3 inches in size), neither of which are burned. A tag in the box reads "P-5 bone, Burial2, Whiteside cat. #1147." This is in agreement with Mr. Whiteside's journal entry for Lot 1147: Bone B2.
Feature 6 (N118 E30, Bottom depth, 2.83 ft. bs) Feature 6 is a ca 2.0 ft. diameter circular pit inside an arc of possible post holes. The pit extended into the clay subsoil (see Figure 3d). The only artifact listed as being found in Feature 6 is one plain sherd.
Feature 11-Area II (N105-135 E10-40) Feature 11 is an arc of 14 post holes of what appears to be a circular house structure roughly 20 ft. in diameter (see Figure 1). The distance to the clay sub-soil was shallower (2 ft. bs) in this area and Mr. Whiteside was able to distinguish the post hole stains as they appeared in the clay subsoil. He indicated the depths of the posts were irregular, the pattern being a large post placed at 4 ft. intervals set to a total depth of 2.5-3 ft. bs in the ground, with smaller posts placed between that were set at depths of 2 ft. (see Figure 3c ).

Feature 12
Feature 12 is a hearth (N63.5 Wl) inside Area I, a possible house location (see Figure 1). The hearth was marked by a 2 ft. diameter ash deposit at 1.5 ft. bs.

Feature 13
Feature 13 is a hearth with ash deposits in Area II (N101 E36). This feature was associated with a pattern of post holes from a circular structure, but it was situated outside of the post hole arc, suggesting it may have been an outdoor hearth (see Figure 1). The feature was exposed at 0.67 ft. bs.

Feature 14
Feature 14 is a concentration of rocks (N80-85 E15-20) uncovered at 1.83 ft. bs by Area I (see Figure 1). No other information is available about the character of the rock feature.

Radiocarbon date
There is one radiocarbon date from the Henry Chapman site on charred wood from Feature 3. The conventional age is 580 ± 60 B.P. (Beta-129978). The calibrated intercept is AD 1400 (cal BP 550).

ARTIFACTS FROM THE HENRY CHAPMAN SITE
A total of 543 artifacts remain from the 1950s excavations by Sam Whiteside at the Henry Chapman site (fable 1 ). More than 84% of the recovered artifacts are plain and decorated ceramic sherds and vessels/partial vessels, with small amounts of lithic tools and debris (8.4%), animal bone (6%), and mud dauber nests and pieces of daub (0.9%). Notable by their absence are ceramic pipes, which occur in varying numbers on most East Texas Caddo sites.

Ceramic Sherds
There are 453 sherds included in the Henry Chapman collection, of which 286 are decorated  (see Table 1). The plain to decorated sherd ratio is only 0.6: 1. The decorated sherds (Table 2) from the site have four different decorative methods. The utility wares include sherds with punctated, incised, and punctated-incised designs; these sherds comprise 70.5% of the decorated sherds. Utility ware types that are found on East Texas Early Caddo sites (ca. A.D. 900-1200), such as Dunkin Incised and Crockett Curvilinear Incised, are present at the Chapman site, although other early utility ware types such as Weches Fingernail Impressed and Pennington Punctated-Incised are apparently absent. The remainder of the decorated sherds are from engraved fine ware vessels. The absence of brushed wares at the Chapman site, found in increasing percentages on later (post-AD. 1200) Caddo sites in the area (see Walters 2008;Walters and Haskins 1998), would suggest that the Chapman site was primarily occupied earlier than ca. A.D. 1200 in the Caddo sequence in this part of East Texas.
On average, the utility wares at the Henry Chapman site are thicker (Table 3) than the fine wares, and about the same thickness as the plain wares. The utility wares usually have a coarser paste than the fine wares, and are mostly large jars with smoothed interior surfaces.
As our knowledge of prehistoric Caddo potters and the pottery they made increases, it is evident that they chose from a number of different techniques to achieve a desired effect when making and firing their wares. Like all learned activity, this process had temporal and spatial differences. Were there differences in, for instance, how utility and fine wares were fired at the Chapman site? At the Chapman site, 87% of the fine ware sherds came from vessels that were fired in a reduced atmosphere, and 13% are sherds from vessels fired in an oxidized atmosphere (Table  4). Of the plain wares, 81% of the sherds were fired in a reduced atmosphere and only 19% were from vessels fired in an oxidizing atmosphere. The utility wares were fired in much the same manner, as 78% of the utility ware sherds are from vessels that were fired in a reduced atmosphere and 22% of the sherds were from vessels fired in an oxidizing atmosphere. Of the 10 bases, eight (80%) were fired in a reduced atmosphere and two (20%) were fired in an oxidizing atmosphere. At the Chapman site, then, the Caddo potters employed similar firing techniques for all pottery wares that emphasized the firing of vessels in a low oxygen environment, probably smothered in a bed of coals. The tempers used by the Caddo potters at the Henry Chapman site were determined by visual observation, or aided by a lOX hand lens, of fresh breaks along sherd cross-sections. Grog is the major tempering agent, occurring in almost 99% of the sherds (Table 5), primarily as the sole temper, or in combination with bone and/or hematite. In some of the thick utility wares, a coarse grog is so prevalent that the sherds have a coarse and crumbly texture. Bone in varying amounts is present in over 35% of the sherds, while crushed ferruginous sandstone, or hematite as it is commonly referred to in the literature, occur in 12.4% of the sherds but never in any significant amounts.

Walters-The Henry Chapman Site 19
The fine wares (n=84) had 49% grog temper alone and 41% had bone and grog or bone alone. Plain wares (n=157) had 57% grog as the sole temper, compared to utility wares (n=202) that had 55% grog temper. Plain and utility ware pottery sherds had almost equal amounts of bone temper, 36% and 38% respectively, in combination with grog and hematite or as the sole temper (see Table 5).

Engraved wares (n=84)
The engraved sherds comprise almost 30% of the decorated sherds from the Chapman site; 55 are body sherds and 29 are rim sherds. The average sherd thickness for the engraved wares is 5.8 mm. The rim sherds are slightly thicker (6.2 mm) than the body sherds (5.4 mm) (see Table 3). This is probably a sampling error as most of the rims showed  (Figure 6a-d) consisting of fine engraved lines in sets; six also have excised areas (Figure 6c-d). Six of the Holly Fine Engraved rims are from carinated bowls; the mean orifice diameter of these vessels is 18.5 em. Nine of the rims have design elements similar to Hickory Fine Engraved; three are from bottles. Three of the Hickory Fine Engraved rims have parallel curved lines, one with red pigment (Figure 6h-i). Another from Feature 3 has broad (4.6 mm wide) horizontal excised lines with red pigment ( Figure  6m). One engraved rim, thinned with a rounded lip, has a cross-hatched design ( Figure 6e). Another rim, direct with a flattened lip that is slanted towards the inside of the vessel, has diagonal engraved lines ( Figure 6g). Finally, one engraved rim has a diagonal ladder motif (Figure 6f) Twenty-one (72%) of the 29 engraved rims are direct with rounded lips. Seven (24%) are direct with flat lips. One flat rim has a tear-dropped indentation on top of the lip. There is one everted rim form with a flat lip.
Of the 55 engraved body sherds, 17 are likely from Holly Fine Engraved vessels. These sherds have fine engraved lines in sets (some more finely executed than others), six have excised areas, and two have red pigment in the lines (see Figure 6n).   Nine are from carinated bowls. Eight Hick01y Fine Engraved sherds, six of which are from bottles, represent the other recognizable type in the fine wares. Other design elements are 13 sherds with multiple curvilinear lines and nine sherds with multiple straight engraved lines. There are three body sherds with single straight engraved lines and one with a single curvilinear line. Three sherds had opposing straight lines (see Figure 6o). One engraved body sherd has curvilinear lines with opposing engraved lines (see Figure 6k), and another has parallel engraved lines with opposing lines (see Figure 61). Finally, there is a body sherd with an excised "cross-shaped" element with red pigment (see Figure 6j). Differing from later Caddo ceramic styles in the region, the engraved fine wares from the Henry Chapman site seldom used an upper or lower line on the rim marking the limits of the motif.

Incised rims (n=23)
Twelve of the incised rim sherds have multiple, diagonal, straight incised lines (Figure 7a-b, d), while one sherd has a single diagonal line. One rim with diagonal lines, from N75-SO E0-5, 0.5-1.0 ft. bs, had been re-fired. Three other incised rims have horizontal lines; another has a single horizontal line under a rolled-out lip. These horizontal incised sherds are probably from Davis Incised vessels, a common type found in early Caddo ceramic assemblages.
One incised rim (N70-90 W0-5) from a bowl had a thickened rim tab with three horizontal lines that dipped underneath the tab (see Figure 7c, e). The sherd is grog-tempered and has been fired in a reduced atmosphere and cooled in the open air. The exterior and interior surfaces are burnished. The rim is direct and has a round lip; its orifice diameter is 20 em, and it is 6.4 mm thick.
Thirteen of the incised sherds (57%) have direct rims, mainly with rounded lips. As previously mentioned, one of these direct rims has a lip tab (see Figure 7c); two have exterior folded lips and are smoothed. Ten have direct rims with flat lips. The average thickness of the incised rims is 7.4 mm (see Table 3 ), with a thickness range from 5.4-1 O.S mm.

Incised Body Sherds (n=SO)
Thirty-four incised body sherds have parallel straight lines as decoration. Two of these are from vessels that had been refired (one from N105-110 E30-35, 0-0.5 ft. bs; the other from N70-90 E 0-5, no depth). On one of these sherds, a hollow reed was employed to make the incised lines.
Another of the incised body sherds, with a portion of the base attached, had vertical straight lines. Seven have opposing incised straight lines, one forming a chevron design. Seven incised body sherds have a single straight line. One sherd has a cross-hatched element on a thickened portion of the sherd. None of the incised sherds from the Chapman site had obvious overhanging lines, a characteristic attribute of Coles Creek Incised (Phillips 1970:70-76), a Lower Mississippi Valley type that is occasionally found on early Caddo sites.
The incised body sherds have an average thickness of7 .S4 mm (see Table 3). Forty-six sherds (92%) have smoothed exterior and interior surfaces. Five sherds have exterior burnished surfaces below the incised decoration and interior smoothed surfaces.

Punctated Rims (n=7)
Three of the punctated rims have random tool punctates ( Figure Sa-b), three have randomly placed fingernail punctates, and one rim has horizontal rows of tool punctates. Five rims have direct rim profiles and flat lips. One rim has a direct profile and a round lip, while the other punctated rim is everted in profile with a round lip. Average thickness of the punctated rims is 7.4 mm (see Table 3).

Punctated Body Sherds (n=89)
More than S5% of the punctated body sherds have random fingernail punctates (see Figure Se-d) that appear to have covered the vessel body. Seven others have random tool punctates (7 .9% ), five have rows of tool punctates (6.3%), and one (1.1 %) has rows of fingernail punctates. Average thickness of the punctated vessel body sherds is S.7 mm, but with a range of 4.3-14.7 mm.

Punctated-incised Rims (n=l6)
Eight of the punctated-incised rims from the Henry Chapman site have opposed diagonal lines that form triangular areas that are filled with tool punctates (Figure 9a-b, d). One of these also has a diagonal incised panel filled with tool punctates adjacent to one of the triangular punctated-filled areas (Figure 9a). Five rims have parallel diagonal lines arranged in opposite directions, forming triangles that are instead filled with fingernail punctates. Another punctated-incised rim has opposing straight lines on the rim and random fingernail punctates on the body section; this sherd had been exposed to extreme heat as evidenced by a crazed surface. One rim has vertical parallel lines separating a zone of random fingernail punctates. Finally, there is one rim with curvilinear parallel incised lines and zones of random tool punctates (Figure 9c). The decorative elements on this sherd resemble those documented for Crockett Curvilinear Incised ceramic vessels. This ceramic type is found in association with early Caddo fine ware types such as Holly and Hickory Fine Engraved (cf. Suhm and Jelks 1962).
Twelve punctated-incised rims are from jars. Seventy-five percent of the punctated-incised rims have direct rim profiles with round lips. Three are direct with flat lips and one punctated-incised rim has an everted rim profile with a round lip. The average thickness of the punctated-incised rims is S.O mm (see Table 3).

Punctated-incised body (n=16)
Five punctated-incised body sherds have opposing parallel straight lines separating zones of random fingernail punctates. Two sherds have parallel straight lines separating zones of fingernail punctates; two sherds have straight lines separating zones of reed punctates; one sherd has diagonal parallel straight lines separating a zone of fingernail punctates; and six sherds have curvilinear lines, four separating zones with fingernail punctates, and two with tool punctates.
One punctated-incised body sherd with reed punctates (N 135-140 E30-35, depth 1.0-1.5 ft. bs) can be differentiated from the majority of the sherds from the Chapman site. The sherd is heavily grog-tempered, is from a vessel fired in a reduced atmosphere, and is 5.5 mm thick. It is decorated with a broad well-defined straight incised line bordering a row of well executed reedlike punctates. The design is similar to Evansville Punctated, var. Rhinehart, a Lower Mississippi Valley ceramic type that dates to the Coles Creek period (Phillips 1970:80-81). c Figure 9. Punctated-lncised rims: a-b, d, triangular incised areas filled with tool punctates; c, curvilinear incised and random tool punctated zone.

Plain rims (n=25)
The average thickness of the plain rims from the Henry Chapman site is 7.6 mm (see Table 3). Eight are direct rims with flat lips and 17 are direct rims with round lips; one of these is a bottle neck rim with an orifice diameter of 4 em. One of the plain rims is from a carinated bowl with an orifice diameter of 24 em. There is also a large plain jar rim from Feature 1.

Plain body sherds (n=132)
There are 132 plain body sherds from the Henry Chapman site. The average thickness of the body sherds is 9.0 mm (see Table 3). Almost 69% have smoothed interior and exterior surfaces, and 30.4% have burnished surfaces; one sherd has a burnished interior and smoothed exterior surface.
Ten plain sherds are from bottle necks. All had smoothed exterior surfaces and rough interior surfaces. The average thickness for these bottle sherds is 6.4 mm. All of the bottle necks are cylindrical to slightly tapered toward the mouth.

Bases (n=IO)
All of the bases are flat, and thicker in the middle than along the edges. The average base sherd thickness i s 11 .1 mm (see Table 3), but with a range of 6.4-15.6 mm. Six of the base s have straight walls leading from the base, and two of these have very abrupt angles. The remaining four bases have body walls that slope at an angle away from the base, such that the base is smaller in diameter than the body of the vessel.

Whole/partial vessels
Seven whole or pattial vessels were found in the excavations at the He my Chapman site. Four of these vessels are from Feature 1, one is from Feature 4, and the other two were not uncovered from feature contexts.
Vessel 1, N75 W4, 1 ft. bs Form: It is a jar that is approximately 25% complete (see Figure 2). The vessel was w~d, possibly from being damaged before final finng and still in a leather stage, then subjected to extreme heat such as would occur in a house fire.  The engraved decoration appears to be a variant of Hickory Fine Engraved (Suhm and Jelks 1962:71-72). Temper: Grog. Color: Exterior color is a grayish-brown with fireclouding; interior color is a light grayish-brown. Firing: undetermined. Thickness: Neck, 5.5 mm, body, 6.7 mm, base, 10.0 mm Figure 11. Large jar with slash punctates from Feature 1.

Walters-The Henry Chapman Site 27
Surface treatment: Exterior surface is burnished.
Vessel 6: Jar from Feature 4 (see Figure 5) Form: Jar; Height: 20.5 em. Orifice diameter: 19.2 em. Base diameter: 10.0 em. Decoration: The rim is decorated with diagonal opposed incised lines, creating a series of incised triangles. The body is covered with four repeating panels of diagonal incised lines separated by vertical incised lines. Dunkin Incised (Suhm and Jelks 1962:37-38). Temper: Grog. Rim and lip profile: Direct rim with a flat lip. Color: Interior color ranges from a dark brown to a light brown with fire clouding. Exterior color is a light reddish-brown to dark brown with fire clouding. Two decorated sherds, one a Holly Fine Engraved sherd, from the Henry Chapman site were analyzed by instrumental neutron activation analysis (INAA) at the University of Missouri Reactor Center. The results show that both sherds are from vessels made from local clay sources, as they were assigned to the Titus chemical reference group (Descantes et al. 2003;Perttula 2002). These findings should be considered preliminary until reanalysis of the entire Caddo INAA database, currently undenvay, is completed.

Mud dauber nests
Two mud dauber nests were recovered in the excavations at the Henry Chapman site. They are from two different parts of the site, suggesting there are burned structures in both locations.
Lot 35, N70-90 E0-5, depth, 1.33 ft. bs. This is a yellowish-brown fired mud dauber nest from a possible burned house. There are grass and cane impressions on one side. Length, 5.5 em, width, 4.0 em, and thickness, 3 em.

Daub
Only three pieces of daub are in the collections from the Henry Chapman site. Perhaps the structures at the site did not have a wattle and daub covering.
Lot 1146, N 135-140 El5-20, depth 1.0-1.5 ft bs. Very hard and light brown fired clay piece with small cane/stick impressions on one side. It was found in the soil zone above Feature 3, the cremation.
Lot 6, N 145-150 E45-50, depth 1.0-1.5 ft. bs. One piece of daub with a sandy paste and hematite inclusions; it is soft and gritty. There are grass impressions on one side, while the other side is not smoothed.

Lithic Artifacts
There were not many lithic artifacts recovered from the excavations at the Chapman site. This could be due to the lack of screening of the archaeological deposits and/or the size of the mesh on the screen used in the excavations. It is possible, however, that the Chapman site may be similar to other Caddo sites in the area that do not have an abundance of lithic materials (Walters 2008). This paucity of lithic artifacts may be a product of an increasing reliance on agriculture by the Caddo at the time of the site occupation, and thus less hunting activities requiring stone tools, or simply changing technologies (i.e., increased use of bone and wood tools). Furthermore, not all of the lithic artifacts recovered from the Chapman site can be associated with any certainty with the Caddo occupation. Certainly the dart points, such as the Kent point, are associated with Late Archaic and Woodland period occupations in East Texas. Most favorable landforms in East Texas usually have evidence of repeated occupations through time.
There are retouch pressure flakes on one side of the tool. The cortex color is a strong brown (7.5YR4/6), and the interior color is a yellowish-brown (10YR5/6). Length, 37.7 mm, width, 20.3 mm, thickness, 7.6 mm.

Chipping debris
There are 19 pieces of chipping debris in the collection: 10 with cortex and nine interior or non-cortical flakes. Eight of the flakes fall in the 2.54 em size class, 10 fall in the 1.27 em size class, and one belongs to the 0.64 em size class.
Twelve of the flakes are chert (five gray, two grayish-brown, two light brown, one dark gray, one red, one gray/red inclusions), and there is one white novaculite flake. Three others are quartzite flakes (two gray and one red), and there are also two hematite flakes, and a milky quartz flake.

Ground Stone tools
The ground stone tools at the Henry Chapman site include five celts and three abraders. The celt was complete, and the bit section is polished. The amount of polishing decreased away from the bit to the battered poll end (Figure 13b).
2. Celt. Lot 913, N130-140 E25-30, depth 0.5-1.0 ft. bs. Reddish-black (2.5YR2.5/1) hematite. This is a complete tool showing chip marks on the bit and smoothing striation marks running parallel to the bit on the bit end and lengthwise on the body. The celt showed flake scars on the body from its manufacture. Length, 79.9 mm, width, 50.3 mm, thickness, 29.1 mm (Figure 13c ). Louisiana and occurs in Tertiary age rocks as a whitish tuffaceous sandstone. This abrader from the Chapman site is made from a coarse, white (10YR8/l), Catahoula sandstone and it is broken in two pieces. One side has multiple grooves that are up to 15.0 mm wide and 7.0 mm deep. One groove is V-shaped, 10.4 mm wide, and 5.0 mm deep. The opposite side has two shallow parallel grooves. This abrader was found in association with Celt #l and #3 and a mud dauber nest in a possible burned house.

Animal bone
Mr. Whiteside's journal lists the following entries for animal bone: Animal bone without lot numbers included in the Chapman collection include: 10 fractured large mammal long bones, three of which were burned; one fractured long bone from a large bird; one unburned deer vertebra; nine deer teeth; and five fractured small mammal bone, two that had been burned. c Figure 13. Celts from the Chapman site: a, sandstone celt; b, graywacke sandstone celt; c, hematite celt.

Possible human bone and Reported human bone
There are nine unburned and possible human cranium fragments in the collection, measuring up to 4.3 mm in thickness. None of these have provenience information. Two lot numbers have human remains: Lot 1147 (Bone Burial 2, Feature 5) and Lot 1148 (cremation bone, Feature 3).

COMPARISONS TO EARLY CADDO MOUND CENTERS AND HABITATION SITES IN EAST TEXAS
As time is paramount to the science of archaeology, placing the Chapman site in some chronological order is essential to understanding the archaeological record of the site and the Caddo settlement of this area. Certain ceramic styles associated with the Alto phase are present at the Henry Chapman site. It would be tempting to use that association to identify the Chapman site as a component of the early Caddo [Alto phase] as defined at the George C. Davis site. However, Story reiterates "components of this phase [Alto] are no where common even though some of the diagnostics, such as Weches Fingernail Punctated and Holly fine Engraved, have wide distributions" (Story 2000:20). Again, based on the artifacts recovered from Mr. Whiteside's excavations at the Chapman site, the main occupation at the site would seem to occur during the Early Caddo period (ca. A.D. 1000-1200), but during what part of the Early Caddo era can only be answered by more absolute dates. At the Chapman site, there is no evidence of the stylistic diversity that characterizes the fine ware ceramics characteristic of the Middle Caddo period (ca. A.D. 1200-1400), nor are there increasing percentages of brushing on utility wares. However, alternatively, groups in the Prairie Creek area may have held on to local ceramics traditions while other contemporaneous Caddo groups changed their ceramic styles, or else chose not to decorate their utility wares with brushing.
There are a few sites, either mound centers or habitation locales, known in Early Caddo times that provide some relevant comparisons with the archaeological record from the Henry Chapman site. I begin with mound centers.

Early Caddo mound centers
There are three known Early Caddo mound centers in the general area of the Henry Chapman site. It is not know if the Chapman site was contemporaneous with any of these ceremonial centers but the Chapman site does have ceramics that are similar to some of those found at these locations.
The Boxed Springs site (41UR30) is some 28 km to the north of the Chapman site on the Sabine River. This is an Early Caddo (ca. A.D. 900-1200) multiple mound center that was recorded and partially investigated by Sam Whiteside in the early 1960s (Perttula et al. 2000). Artifacts from excavations and from a cemetery looted in the 1980s include Holly Fine Engraved and Hickory Fine Engraved, Spiro Engraved, Coles Creek Incised, Weches Fingernail Impressed, Kiam Incised, East Incised, Crenshaw Fluted, and Crockett Curvilinear Incised vessels.
Engraved wares at the Hudnall-Pirtle site (41RK4), an early Caddo mound center on the Sabine River some 48 km to the east of the Chapman site, included examples of both Holly and Hickory Fine Engraved but it was noted that not all of the engraved decorations represented "fine engraved" lines but rather the lines were described as being thicker and generally coarser (Bruseth and Perttula 2006:90-91); this was not the case with the Hickory and Holly Fine Engraved sherds from the Henry Chapman site. Other Early Caddo pottery types noted at the Hudnall-Pirtle site were Davis Incised, Dunkin Incised, Weches Fingernail Impressed, Pennington Punctated-Incised, and Crockett Curvilinear Incised. There were also sherds similar to Coles Creek Incised, var. Coles Creek and var. Hardy. There were six pipe fragments from long-stemmed Red River pipes recovered from excavations at the site. A total of 93 projectile points were recovered at the Hudnall-Pirtle site. The main arrow points include: Alba (42%), Catahoula (12.5%), Steiner (10.4%), and Colbert (10.4%) (Bruseth and Perttula 2006).
The George C. Davis site (41CE19), some 85 km to the south of the Chapman site, is one of the better dated/investigated Early Caddo mound centers in East Texas. Dates at the Davis site indicate the Alto phase occupation here dated from the late A.D. 800s to A.D. 1300. Dee Ann Story (2000), in the introduction to the classic The George C. Davis Site, Cherokee County, Texas report, defines the Alto phase based on the following co-occurring ceramic types: Holly Fine Engraved, Weches Fingernail Punctated, Davis Incised, Crockett Curvilinear Incised, Pennington Punctated Incised, Dunkin Incised, and Duren Neck Banded. Less diagnostic, but commonly present on Alto phase sites, are Hickory Fine Engraved ceramics, pinched ridge pottery, long-stemmed Red River pipes, Alba arrow points , Gahagan bifaces, petaloid celts, and expedient tools made on flakes. Story (2000) contends that this phase, found in both mound centers and small habitation sites across a large area, represents the firs t distinctively Caddo remains in the middle Sabine and Neches river basins.

Local Caddo habitation sites
Compared to later Caddo sites, habitation sites with Early Caddo ceramics, such as are found at the Henry Chapman site, are scarce in this area of East Texas. Story (2000) suggests that the local region was not heavily populated at this time.
The Wolf site (41SM1 95), a small Caddo habitation site located on a nearby drainage to the Chapman site, has radiocarbon and OCR dates indicating a mid-fourteenth century occupation (Walters 2003). Significantly, however, the ceramic assemblage from this site lacks brushed pottery, as does the Henry Chapman site ceramic assemblage. Decorative techniques in the ceramics from the Wolf site include: punctated (52.4%), engraved (29.5%), incised (14.3%), and punctated-incised (3.8%). There is one engraved sherd similar to Holly Fine Engraved with excised areas, nested excised triangles, and red pigment in the lines. There were no ceramic pipes recovered from the Wolf site.
The Broadway site (41SM273), located some 24 km to the west of the Chapman site, has evidence of an early Caddo occupation dating ca. A.D. 900-1000 based on radiocarbon dates and the presence of small amounts of sherds from Holly Fine Engraved , Weches Fingernail Punctated, Dunkin Incised, as well as Crockett Curvilinear Incised, and long-stemmed Red River pipe sherds (Perttula and Nelson 2004). Perttula and Nelson (2004) surmised that the early Caddo occupation at the Broadway site was contemporaneous with, and that perhaps there was some level of contact/interaction between, the Caddo groups that occupied the George C. Davis site, but that the Caddo living at the Broadway site were part of a different social group.
Also located in Smith County is the Joe M eyers site (41SM73) in the Neches River basin, where members of the East Texas Archeological Society recovered Alto phase ceramics from habitation areas and burials (Johnson 1961). The Meyers site was the only one of 34 ceramic site in the Lake Palestine project area that contained early Caddo ceramic types (Jelks 1958).

Walters-The Henry Chapman Site 33
The burials at the site included six single burials and one multiple burial that contained the remains of probably four individuals. Fifteen vessels were associated with the burials, including Bowles Creek Plain, Hickory Fine Engraved, Weches Fingernail Impressed, and Canton Incised (Jelks 1958). Surface collections from the site had sherds from the following early Caddo ceramic types: Hickory Fine Engraved, Davis Incised, Dunkin Incised, Canton Incised, and Weches Fingernail Impressed.
The six Weches Fingernail Impressed sherds were all rims with flat lips. A description of the ceramic type Weches Fingernail Impressed describes the lips as being rounded and flat in about equal numbers (Suhm and Krieger 1954:364). Thirteen plain rims from the Myers site were also described as being flat. At the Henry Chapman site 34 (34%) of the 100 rim sherds had flattened lips.
Interestingl y, Jelks (1 958) noted that stone material was not common at the Meyers site, a situation similar to the material culture at the Henry Chapman site.
Grace Creek#2 (41GG34) had a small amount of early Caddo material, including sherds, one Red River pipe stem, and one Alba arrow point. Early Caddo ceramic types collected were Weches Fingernail Impressed (n=2), Dunkin Incised (n=1 ), Davis Incised (n=1 ), and a Hickory Fine Engraved rim with a fl at lip (n=1 ).

CONCLUSIONS
The Henry Chapman site was a prehistoric Caddo habitation site much like other Caddo sites in the northern part of Smith County, and in the Sabine River basin. These sites were probably occupied by kinship-related groups who cooperated in the gathering, production, and consumption of local resources. No evidence was obtained by Mr. Whiteside's investigations confirming the cultivation of domesticated crops such as maize but investigations at other nearby sites suggest that agricultural activities supplemented the hunting and gathering of native plants in the Caddo diet. Evidence of large utility vessels at the site is indicative of the change by Caddo peoples after ca. A.D. 850 from hunting-gathering to a more sedentary lifestyle with the storage of foodstuffs, including domesticated plants. Large carinated bowls present at the Chapman site would seem to indicate that feasting by Caddo living there also played a role in their social life.
Excavations at the Chapman site seem to indicate the presence of three or more permanently constructed houses, either all occupied at one time or more likely it is the case that they represent a sequence of houses. There is evidence that one or more of the houses may have burned, either by accident or intentionally. These houses, following the settlement pattern of other small farmsteads, probably were not occupied for more than one decade at a time, and thus if the houses represent sequential use, the Henry Chapman site may have been occupied overall for ca. 30 years.
There were two burials reported from the Chapman site, although only one of these was documented in any detail. This burial, Feature 3, was a partial cremation and contained no obvious grave goods. A single radiocarbon date from the burial indicates the burial took place somewhere between AD 1280-1440. Cremations of this type are atypical in Caddo mortuary traditions, although an early Caddo cremation has been reported from the Boxed Springs (41 UR30) mound center on Big Sandy Creek near its confluence with the Sabine River. Artifacts in the grave fill in Feature 3 at the Henry Chapman site indicate that the burial took place during (or after) the principal Caddo occupation of the site. A date obtained from the burial appears to be anomalously too recent based on the ceramics recovered from excavations at the Chapman site.
In summary, the Henry Chapman site represents a local group of folks that lived in this part of East Texas between ca. A.D. 1000-1200 that are culturally different from other Caddo groups, although it appears to be linked to them in some fashion by direct contact or interaction. The Henry Chapman site may have been occupied by a certain local Caddo group that held on to certain traditions (such as distinctive styles of decorations on ceramic vessels) longer than did contemporaneous groups, or they resisted changes in utility ware decorative styles (such as brushing on vessels), longer than groups in other areas. It is uncertain what the nature of the social and cultural glue was that held these scattered Caddo groups in this area together. Hopefully, studies of other Caddo sites investigated by Mr. Whiteside and others on Prairie Creek and surrounding drainages will shed more light on how the Henry Chapman site fits into Caddo cultural history.