SIUE FIELD SCHOOL INVESTIGATIONS )

SIUE FIELD SCHOOL INVESTIGATIONS )
SIUE FIELD SCHOOL INVESTIGATIONS
IN THE LOCALE OF THE D. HITCHINS SITE (11MS1124)
(Please cite as: Electronic Version, http://www.siue.edu/ANTHROPOLOGY/, 2009)
Julie Zimmermann Holt
Miranda Yancey
Erin L. Marks
Southern Illinois University Edwardsville
March 2009
CONTENTS
Abstract……………………………………………………………...……………………2
Acknowledgments…………………………………………….…………………….…….3
Introduction………………………………………………………………….…………….4
Setting…………………………………………………………………...……………...…5
Previous research………………………………………………………………….……..7
Geophysical survey……………………………………………………………..………..8
Surface survey……………………………………………………………………...…….10
Shovel testing……………………………………………………………….……………12
Excavation methods………………………………………………………….…………..14
Excavation results………………………………………………………….…………….15
Features…………………………………………………………………………………..21
Ceramics…………………………………………………………………………..……..27
Chert artifacts……………………………………………………………………………28
Other lithic artifacts ………………………………………………………………….….30
Plant remains…………………………………………………………………………..…31
Animal remains…………………………………………………………………….…….33
Historic artifacts………………………………………………………………………….33
Summary and discussion………………………………………………………………....33
Conclusion……………………………………………………………………………….39
References………………………………………………..………………………………41
List of tables…………………………………………………………………………….44
List of figures………………………………………………………………………...…60
List of photographs……………………………………………………………………..75
1
ABSTRACT
The Anthropology Department of Southern Illinois University Edwardsville
conducted an archaeological field school in Hamel Township, Madison County, Illinois,
between May 15 and July 3, 2006. Field school investigations included surface survey
over approximately 30 acres of agricultural fields, an area which included previously
recorded sites 11MS1124, 11MS1125, 11MS1126, and 11MS1127; shovel testing at
11MS1125; and excavation at 11MS1124, the D. Hitchins site. Excavations focused on
the D. Hitchins Site (11MS1124) because it was predicted that Mississippian features
would be encountered there. Artifacts recovered from the plowzone during excavation at
11MS1124 date from the Early Archaic through Mississippian periods. In the
northernmost excavation unit, two shallow pit features appear to date to the early Late
Woodland period. A concentration of fire cracked rock was recovered from an
excavation unit farther south. In the southernmost excavation unit, a small circular post
structure and four pit features were encountered. Three of these four pits were excavated,
two of which were found to contain ceramics dating to the early Mississippian period.
Their proximity to the structure suggests that it and the adjacent pits also date to the
Mississippian period. The three excavated pit features contained maize, seeds of native
cultigens, and nutshell. These features are believed to represent a small Mississippian
camp used primarily to shelter people engaged in agricultural tasks during the early
Lohmann phase. Other activities undertaken by Mississippian occupants included
hunting, nut gathering, and possibly games and/or rituals.
2
ACKNOWLEDGMENTS
Deep and sincere thanks are extended to the following individuals: Mike
Hargrave, for doing the geophysical survey; Jim Heepke and Pat Martin, for working
around us as they planted and tended the crops; Henry Holt, for helping set up the grid,
giving advice in the field, backfilling, providing comments on this report, and too many
other things to list here; John Kelly, for helping us with the ceramics; Brad Koldehoff, for
examining the lithic materials and visiting us in the field; Sheryl Lauth, for helping with
several of the graphics; Marge Schroeder, for identifying the plant remains and helping
prepare the radiocarbon samples; and Greg Vogel, for providing a lecture on soils and the
good earthworm. SIUE’s Undergraduate Research Academy provided funding to Erin
Marks that paid for the paleoethnobotanical analysis and radiocarbon dating. Thanks to
Hong Wang and the ISGS for running a second date for us at no charge. Finally, the
biggest thanks go to the intrepid students who took the field school and kept on digging
no matter how hot it got.
3
INTRODUCTION
The Anthropology Department of Southern Illinois University Edwardsville
(SIUE) conducted an archaeological field school on the property of Henry and Julie Holt
in rural Hamel Township in Madison County, Illinois, between May 15 and July 3, 2006.
Field school investigations included surface survey over approximately 30 acres of
agricultural fields, an area which included previously recorded sites 11MS1124,
11MS1125, 11MS1126, and 11MS1127; shovel testing at 11MS1125; and excavation at
11MS1124, also known as the D. Hitchins site (Woods and Holley 1991). Julie Holt
acted as field director and instructor of record. Nine full-time students (Lori Belknap,
Linda Coats, Gregory Guntren, Lacy Heflin, Brian Kumpf, Sarah Linhart, Erin Marks,
Kye Miller, and Patrick Sullivan) and three part-time students (Nicholas Jahlbert, Valerie
Starr, and Miranda Yancey) enrolled in the course and acted as field crew.
The primary goal of the SIUE archaeology field school is to teach students
standard archaeological field methods. In addition, the field school offers research
opportunities to SIUE faculty and also students, who are encouraged to do original
research for their senior projects. Perhaps most important, the field school provides a
means for recording endangered archaeological sites, which are rapidly disappearing due
to development in Madison County. In the case of the area investigated by the 2006 field
school, there is little danger of development since the current landowners are committed
to preserving the site. In fact, the project area was taken out of cultivation following the
SIUE investigations, and efforts are underway to restore native habitat under the
Conservation Reserve Program. With this restoration project in mind, the field school
provided the opportunity to document the archaeological resources of the project area
while it was still cleared for agriculture and before it became hidden underneath prairie
grasses.
The field school achieved all of these goals. All students completed the course
successfully and received experience in all phases of archaeological fieldwork. In terms
of research, we were particularly interested in Mississippian use of 11MS1124, the D.
Hitchins site, which is why we conducted excavations in this area. The data recovered
provide significant new information about the Mississippian occupation at the D.
Hitchins site, and also information about early Late Woodland utilization of the uplands
of northern Madison County. As of this writing, two senior projects (Marks 2007;
Guntren 2008) and one master’s thesis (Yancey 2007) have been completed that utilize
data recovered during the field school investigations.
This report summarizes results of the SIUE field school undertaken on the Holt
property in the summer of 2006. Surface surveys yielded mostly debitage and FCR, but
artifact distributions led us to redefine the boundaries of several sites and record a new
site. Shovel testing at 11MS1125 indicated a low-density lithic scatter confined to the
plowzone of a single ridge top. Excavations were conducted at 11MS1124, the D.
Hitchins site, because it had the highest potential to contain intact Mississippian features.
Artifacts recovered by the SIUE field school from 11MS1124, however, date from the
Early Archaic through Mississippian periods. In the northernmost excavation block on
11MS1124, two shallow pit features appear to date to the early Late Woodland period.
Farther south, an undated concentration of FCR was recovered from another excavation
block. In the southernmost excavation block, a small circular post structure and four pit
4
features were encountered. The deepest pit and one shallow pit contained ceramics
dating to the early Mississippian period. Based on their apparent association with these
pits, the structure and the other pits are also assumed to date to the early Mississippian
period.
This report will begin with a description of the site setting and a summary of
previous investigations in the area. We will then detail our methods and results of all
phases of research. Finally, our findings will be compared with data from the greater
American Bottom. While the project area is deep in the uplands and outside of the
American Bottom technically speaking, it is clear from our research that the people who
used this area during virtually all periods of prehistory were engaged in the social arena
of the greater American Bottom.
SETTING
The portion of the Holt property investigated lies in the east half of the northeast
quarter of Section 6, Township 5 North, Range 7 West in Madison County, on a rolling
terrace located above an old channel of Sherry Creek (Figure 1). Today this channel
contains water only seasonally, since Sherry Creek was straightened and rerouted away
from the terrace, apparently in the late 19th century. The 1815 GLO map shows Sherry
Creek in this old, meandering channel, although it is unnamed on the map (Figure 2).
Cahokia Creek is named and appears much larger than Sherry Creek on the 1815 map.
The 1815 map also shows that the area investigated was in timber at that time, although
the prairie’s edge lies just northwest of the project area (less than 0.5 km away) and
prairie was also located ca. 1 km southeast of the area, on the other side of Cahokia Creek.
The 1873 plat map indicates that the project area was still wooded at that time. The land
was sold in 1875 to Henry Gunkel, who apparently cleared the land for agriculture. The
1892 plat map indicates that Sherry Creek was straightened and rerouted by that time.
Figure 1. USGS map showing project area
and previously recorded sites.
Figure 2. 1814 GLO map. Hamel township.
5
Prehistorically the creek channel below the project area would have been active,
joining with Cahokia Creek immediately south of the area investigated. The creek would
have provided both water and food (e.g. fish, frogs, and turtles). People living here
would have had easy access to forest resources (e.g. venison, nuts, turkeys, and wood),
edge resources (e.g. berries, bobwhite, and cottontail rabbit), and prairie resources (e.g.
prairie chickens and prairie grasses). Although the project area is located well into the
uplands, ca. 20 km north of the American Bottom, creeks like Cahokia Creek and Sherry
Creek dissect the uplands and create floodplain environments within them. Thus,
inhabitants of the project area had access to both upland and floodplain resources. Both
upland and floodplain soils are suitable for cultivation, and the area was probably
cultivated during later periods of prehistory, certainly during the Mississippian period.
The D. Hitchins site, 11MS1124, is situated on Hickory and Homen silt loams, which are
formed on till plains (Leeper 2004). Wakeland soils are found on the floodplain below
the D. Hitchins site to the east. Woods (1986) argued that Wakeland soils were selected
by Mississippian groups for agriculture based on a correlation between Wakeland soils
and Mississippian sites in the upland Cahokia Creek drainage. The forests which
occurred within upland drainages would have been easily cleared for agriculture using
slash and burn techniques. Away from the forested creeks, prairie soils were less likely
to be farmed prehistorically since breaking through dense prairie sods with stone hoes
would have been comparatively more difficult.
The social setting of the site is less clear, since relatively little archaeological
work has been done in the uplands north of Edwardsville. Woods’ survey (1986)
recorded over 100 sites in the upland Cahokia Creek drainage, but only 45 could be
assigned to a time period and none were known to be excavated prior to the 2006 SIUE
field school. Both Woods’ survey (see below) and the 2006 SIUE investigations suggest
that the project area was occupied repeatedly throughout prehistory, but most occupations
seem to have been short term. Woods’ team also surface collected 11MS341 (the
Kruckeberg No. 1 site) ca. 0.5 km east of the project area, which they identified as a
Mississippian village. Woods and Holley (1991) suggest that this village may have been
fortified, based on their interpretation of aerial photographs. They also suggest that both
11MS341 and 11MS1124 (the D. Hitchins site) date to the Moorehead phase. If the two
sites were contemporary, perhaps 11MS1124 was an extension of 11MS341, such as a
camp for hunting, farming, and possibly ritual activities. However, the Mississippian
features excavated by the SIUE field school at 11MS1124 more likely date to the early
Lohmann phase (the start of the Mississippian period) than the Moorehead phase (midlate Mississippian period). Besides the possible Mississippian village at 11MS341, it is
striking that no other long-term habitation sites dating to any time period are known to
exist in the uplands of Cahokia Creek.
While there is relatively little information about the nature of prehistoric
occupation in the upland Cahokia Creek drainage, there is an abundance of information
for the area where Cahokia Creek exits the uplands into the American Bottom at
Edwardsville (some 20 km south of the project area). Today, Cahokia Creek is
channelized there and heads due west for the Mississippi River, but prehistorically,
Cahokia Creek meandered southward along the base of the bluff, finally turning west as it
passed Cahokia and emptied into the Mississippi River. Many sites have been recorded
at both the top of the bluff and the base of the bluff in the Edwardsville area, and some of
6
these have been excavated (e.g., see Holt et al. 2005). It is clear that by the end of the
Late Woodland period, occupation along the bluff bordering the American Bottom in this
area was intense and long term. At this time it does not appear that occupation of the
uplands to the north was as intense or long term, with the possible exception of the
Kruckeberg No. 1 site, but the lack of excavation data from this area makes it impossible
to know this with certainty. In any case, it would have been an easy canoe trip from the
D. Hitchins site downstream to Cahokia (some 35 km via Cahokia Creek) and the
Mississippi River beyond.
PREVIOUS RESEARCH
Given that artifact collectors frequented the Hitchins’ farm for several decades
prior to its purchase by the Holts in 2003, the project area is probably better known to
avocational archaeologists than it is to professional archaeologists. Of these collectors,
only Dave Klostermeier’s collection is well known because Yancey (2005) documented it
as part of her senior project at SIUE. Klostermeier’s collection from the Hitchins’ farm
contains mostly lithic artifacts but also some ceramics. Yancey reports that projectile
points in the Klostermeier collection date to the Early Archaic, Middle Archaic, Late
Archaic, Middle Woodland, Late Woodland, and Mississippian periods. In addition,
Klostermeier has identified one point fragment within his collection as a broken Clovis
point (personal communication, July 2006); this identification has been confirmed by
Brad Koldehoff. Thus, projectile points in Klostermeier’s collection indicate that the site
was occupied or utilized during every prehistoric time period with the possible exception
of the Early Woodland period. Particularly notable within Klostermeier’s lithic
collection are Mississippian artifacts including a flare-tip celt and chunkey stone. The
chunkey stone or discoidal is a Cahokia type and crudely engraved with a weeping eye on
one side and a sun on the other side (Yancey 2005). Based on the presence of these nonutilitarian artifacts, Klostermeier believed that D. Hitchins was a Mississippian ritual site,
possibly containing burials. Ceramics collected by Klostermeier include several dozen
cordmarked grit tempered sherds, most of which were identified by Yancey as z-twisted,
and a few shell tempered sherds. These indicate occupation from the Late Woodland
through Mississippian periods; more precisely, one shell-tempered rim sherd collected by
Klostermeier probably dates to the Moorehead phase (Yancey 2005).
The first professional research done on the property was a survey by SIUE, which
was conducted in 1982 as part of Woods’ dissertation research (Woods 1986). Woods
recorded four sites on the Hitchins’ farm: 11MS1124, 11MS1125, 11MS1126, and
11MS1127. 11MS1127 was identified as a Late Woodland site; 11MS1125 and
11MS1126 were simply identified as prehistoric. The D. Hitchins site, 11MS1124, was
identified as a multi-component site, with artifacts recovered dating to the Early or
Middle Woodland, Late Woodland, and Mississippian periods. Notably, Woods reported
one rim sherd from an angled jar that was red-slipped with a smudged-black rim. This
rim, like the rim sherd in the Klostermeier collection, suggests a Moorehead phase
component at 11MS1124 (see Woods and Holley 1991:figure 3.4b). Woods (1986:90)
suggests that 11MS1124 was a Mississippian farmstead.
Yancey’s work with Klostermeier suggested that 11MS1124 is larger than
initially reported by Woods to the IAS, perhaps as large as 3 hectares (Yancey 2005).
7
However, we will argue below that the western portion of what Yancey referred to as
11MS1124 should be considered the northern end of 11MS1127. In addition, Yancey
(2005) documented another site ca. 150 m north of 11MS1124, where the current
landowners (Henry and Julie Holt) had observed debitage and FCR. We reported this site
to the IAS when we completed site forms for the present project, naming it the Gunkel
site to honor Henry and Katie Gunkel, who owned, lived on, and farmed the property
from 1875 until 1924. The site number assigned to the Gunkel site by the IAS is
11MS2336.
GEOPHYSICAL SURVEY
Our primary research goal was to determine the nature of the Mississippian
occupation at 11MS1124, the D. Hitchins site. Based on discussions with Klostermeier,
it was predicted that intact Mississippian features would most likely be located on the
eastern portion of the site. Remote sensing was used to guide our excavations in this area,
since this non-invasive method can be used to predict more precisely where features
might be located.
In April of 2006, Henry Holt located two survey pins set by surveyor Bill Lovsey
on the property when it was still owned by Don and Martha Hitchins. These pins marked
the northeast and southeast corners of the Hitchins’ pasture; today, the area is a
reconstructed wetland. Using these pins to establish a north-south base line, Holt set in a
hub 55 meters south of Lovsey’s southern pin. This point sat at the crest of a slope and
was later named N300 E300 on the SIUE grid. The survey instrument used was a
Realist-David White 4.5 inch transit, model TR300.
On April 15, 2006, Michael Hargrave of Engineer Research and Development
Center/Construction Engineering Research Laboratory (ERDC/CERL) conducted a
geophysical survey at 11MS1124 at
the request of Julie Holt (Photo 1).
Several field school students assisted
in the field work. A magnetic field
gradient survey and an electrical
resistance survey were conducted with
the goal of identifying possible
prehistoric features that would be
investigated by the SIUE field school.
Time permitted magnetic sensing in
six complete and two partial 20 x 20 m
squares, and electrical survey of one
complete and one partial 20 x 20 m
square. The squares chosen for the
survey were deemed most likely to
contain Mississippian features based
on topography (two flat “saddles” were
surveyed; a ravine in between these two “saddles” was not surveyed). The eastern edge
of the northeastern survey block ended approximately where the terrace suddenly drops
in elevation down to the floodplain of Sherry Creek.
8
Results of the electrical survey suggested two relatively large rectilinear
anomalies southeast of N300 E300 (see Figure 3). Hargrave suggested that these could
be possible structures but were more likely the intersection of plow furrows. The
magnetic survey revealed a number of smaller, generally circular anomalies, mostly
located in the southern survey blocks (see Figure 3). Hargrave indicated that some of
these could be prehistoric pits, while others were possibly the result of small fragments of
iron or dips in the plowzone. In general, Hargrave noted that the site was magnetically
“quiet.”
Hargrave and Julie Holt cored several anomalies and several non-anomalies for
comparison on the day of the survey. Soil compaction made it very difficult to penetrate
the ground with the core. Cores made inside and outside the rectilinear electrical
anomalies suggestive of structures showed similar profiles, indicating a plowzone 15-20
cm thick over subsoil. In the vicinity of N220 E300, two magnetic anomalies (possible
pits) and one control location were cored. Cores of the two anomalies showed no
evidence of features, indicating a plowzone approximately 30 cm thick over subsoil.
Ironically, the control core indicated a plowzone approximately 20 cm thick over 10 cm
of dark soil containing charcoal flecks over subsoil. That is, the control core suggested a
possible feature where none was expected. Subsequent coring 30 cm north and 30 south
of this possible feature showed no trace of it, indicating that whatever it was, it was small.
Holt and field school students attempted to core several more anomalies in early June but
were unable to penetrate the ground below the plowzone with the core.
Figure 3. Locations of magnetic and resistance anomalies at 11MS1124 recommended for ground truthing. Map by
Dr. Michael Hargrave of ERDC/CERL. Hargrave’s N120 W40 = SIUE’s N300 E300.
Several anomalies were selected for excavation. These anomalies included one
possible structure identified in the electrical survey and two possible pit features
9
identified in the magnetic survey. The possible structure did not materialize; however,
two shallow pits were found in this area. One possible pit feature excavated did indeed
turn out to be a pit feature, while the other magnetic anomaly excavated was found to be
a large concentration of FCR. Results of these excavations will be discussed in detail
below.
SURFACE SURVEY
Pedestrian survey was conducted over approximately 30 acres of an upland
agricultural field in May and June of 2006 (Photo 2). A surveyor walked down every
eighth planted row, resulting in
transects spaced 5.3 m apart. The
rows ran north-south, except where
the planter had turned corners.
The majority of the field was
surveyed on May 15 and 16. At
the time of the May survey, the
corn crop was only several inches
tall, but visibility was poor,
perhaps 25% at best. The poor
visibility was caused by soybean
residue left from the previous
year’s crop, which remained in the
field because the 2006 crop had
been drilled without tilling.
Students placed pin flags wherever a possible artifact was encountered. After the
instructor (Julie Holt) verified the find, students recorded the location, artifact type, and
count using a Garmin GPSMAP 76CS. Artifacts were left in the field unless they were
diagnostic to time period.
Most artifacts recorded during this controlled surface survey were debitage or
FCR. Only one diagnostic artifact was collected, a broken spear point tentatively
identified by Brad Koldehoff as a Middle Archaic Falling Springs point made of
Burlington chert. This point was recovered from 11MS1127, which Woods’ crew had
identified as a Late Woodland site. Beside the Archaic point, no other artifacts were
recovered from 11MS1127 (Figure 4). However, the landform containing 11MS1127
continues to rise in elevation toward the north, and it was along this ridge that most
artifacts were found during surface collection. Therefore, we submitted site forms to the
IAS that expand the boundaries of 11MS1127 to continue up the ridge to the north (see
Figure 5). Klostermeier’s collections on the peak of this landform primarily yielded
Archaic artifacts (see Yancey 2005), but Klostermeier also found one Late Woodland
ceramic vessel in this area. In sum, Archaic artifacts were found from the north to the
south end of this landform, Klostermeier found evidence of a Late Woodland component
toward the northern end of the landform (the highest point in elevation), and Woods’
crew also found evidence of Late Woodland use at the southern end of the landform.
Few artifacts were observed on the surface of 11MS1126. A small concentration
of artifacts was noted in an area of higher elevation west of 11MS1126, in the
10
southwestern corner of the area surveyed (Figure 4). This lithic scatter was previously
identified by Yancey (2005). Although the controlled survey made by SIUE in 2006
recovered nothing diagnostic from this site, Klostermeier’s collections indicate that it
contained Archaic occupations. Thus, we submitted site forms to the IAS identifying this
as an Archaic site. The IAS site number assigned was11MS2337.
The area surveyed on May 15-16 included only what had been planted in corn
prior to that date. The unplanted/unsurveyed portion of the field to the east included both
11MS1125 and 11MS1124, the D. Hitchins site. The landform containing 11MS1125
was shovel tested between May 17 and 23 (see below). On May 23, the eastern portion
of the field was planted in soybeans. Excavation of 11MS1124 was begun on May 25
(see below). A controlled surface collection was finally made of the area planted in
soybeans, including sites 11MS1124 and 11MS1125, on June 15. Artifacts were
recorded by students led by Yancey using the Garmin GPS. Only debitage and FCR were
observed, mostly in association with 11MS1124 (see Figure 4). Few artifacts were noted
on the surface of 11MS1125. However, by this date, visibility was extremely poor: the
soybeans were only about 6 inches tall, but weeds were dense and often taller than the
soybeans, resulting in visibility ranging between 5 and 25%. Two diagnostic artifacts
were recovered from the surface of 11MS1124, not during the controlled surface survey,
but prior to that date since they happened to be lying near excavation units. An early
Late Woodland Steuben point was found near N294 E302, and a Mississippian Cahokia
point was found near N230 E305. Archaic points were also recovered during our
excavations.
Figure 4. Surface collection results.
Figure 5. USGS map showing revised site
boundaries and newly recorded
sites.
Thus, our investigations confirm Woods’ survey in terms of the time periods
represented at 11MS1124, but Klostermeier’s collections also indicate a significant
Middle Woodland component at the site (Yancey 2005). The site forms we submitted for
11MS1124 redefined the boundaries of the site. Whereas the site boundaries as defined
by Woods indicated that the peak elevation to the west should be considered part of
11
11MS1124, we consider that area to be part of 11MS1127 for several reasons. First, it
exists on the same landform as 11MS1127, a ridge running from north to south. Second,
our controlled surface surveys clearly show a continuous scatter of artifacts from north to
south along this landform, whereas there is an area devoid of artifacts between the high
point at the north end of the ridge and 11MS1124 to the east. 11MS1127 as redefined is
primarily an Archaic site with evidence of Late Woodland use; 11MS1124 as redefined is
primarily a Mississippian site with additional evidence of Archaic, Middle Woodland,
and Late Woodland use.
Finally, we should note that Woods and Holley (1991) referred to 11MS1124 as
the D. Hitchens site, after the landowner at the time of Woods’ survey in the 1980s.
When we completed the site form for the new site found in our survey, and site revisit
forms for the sites found by Woods, we named the other sites on the property after
current and former landowners. Site 11MS1125 was named the M. Hitchins site to honor
Don’s wife and partner, Martha. Site 11MS1126 was named the John Weaver site after a
19th century landowner. Site 11MS1127 was named the Holt site after the current
landowners, who purchased the property from Don and Martha Hitchins in 2003. Site
11MS2337 was named the Franke site after Mr. and Mrs. Louis Franke. Louis Franke
bought the farm from Harry and Katie Gunkel in 1924, and Franke’s heirs sold it to Don
Hitchins in 1958.
SHOVEL TESTING
The finger ridge containing 11MS1125 was shovel tested on May 17-23 (Photo 3).
This ridge was shovel tested because the western slope of the landform, which descended
into a ravine, was to be the source of fill used to dam the ravine at its southern (lower)
end in order to create a small pond. Shovel test pits (STPs) were approximately 40 cm
wide and excavated until sterile subsoil was reached. All sediments were screened
through quarter-inch mesh and all cultural materials were collected. Profiles showing
sediment colors and textures were drawn for all STPs.
Photo 3. Shovel testing at 11MS1125.
Figure 6. Shovel test results, 11MS1125.
Transect A was laid in first, with STPs A1-A12 set 10 m apart (Figure 6). A1 and
A2 were located on the floodplain, A3 was located on the slope, and A4 was located
12
more or less at the crest of the ridge. A5 through A10 continued to rise gently in
elevation. A10 was located just inside the corn field which had already been surface
collected, so A10-A12 were not excavated.
Transect B was laid in 10 m east of A but offset by 5 m to the north (Figure 6).
STPs B1-B12 were set 10 m apart, with B1 on the floodplain, B2 on the slope, and B3
more or less on the crest of the ridge. The landform continued to rise gently north of B3
but was rolling, so that B5 and B10 were lower in elevation than adjacent STPs. Transect
C was laid in 10 m west of B but again offset by 5 m to the north. STPs C1-C13 were set
10 m apart, with C1 and C2 on the floodplain, C3 and C4 on the western side of the ridge,
and C5 and C6 on the crest of the ridge. The slope continued to rise gently in elevation
north of C6. C6-C13 were located in the corn field which had already been surface
collected, but we excavated them since they were close to area that would be borrowed.
Transect D was laid in 10 m west of transect C and was offset 5 m to the south so that it
was in line with transect A. STP D1 was located in an agricultural drainage ditch so was
not excavated. STPs D2-D16 were located in the corn field previously surveyed but were
excavated since they were located in the area to be borrowed. D2-D3 were located in the
floodplain, while D4-D16 were located on the side of the ravine.
STPs on the top of the ridge generally indicate a plowzone (≈10YR4/3 brown
silty clay loam) of varying depth (7-30 cm) over subsoil (≈10YR5/8 yellowish brown
silty clay loam). The plowzone of STPs A6-A9, B4-B5, B9-B10, and B11-B12 contained
low densities of debitage, ranging from one to five pieces of chert per STP (Table 1).
Additional STPs were laid in five m from these positive STPs (A4.5, 5.5, 6.5, 7.5, 8.5,
9.5, and 10.5; B3.5, 4.5, 8.5, 9.5, and 10.5; E5; G9.5 and 10; and F 8, 8.5, 9, 9.5, and 10).
The positive STPs located on top of the ridge can be used to define the boundaries of
11MS1125. The site is highly eroded, and no evidence of intact features was found. A
biface fragment was recovered from STP B5, but no diagnostic artifacts were recovered.
All chert was identified as either Burlington or glacial chert.
STPs located on the steepest slopes of the finger ridge and ravine generally
indicate a plowzone (≈10YR4/3 brown clay loam with gravels) of varying depth (15-30
cm) over subsoil (≈10YR5/8 yellowish brown clay loam). The plowzone of several STPs
(D4, D10, and D16) on the slope contained one or two possible flakes among glacial
gravels. Because the slopes of the ridge are also highly eroded from plowing and
colluviation, it is assumed that the few cultural materials found in the plowzone had
washed in from above.
STPs located on the floodplain generally indicate a plowzone (≈10YR4/4 dark
yellowish brown loam) of considerable depth (40-90 cm) over alluvial soils (≈10YR3/2
very dark grayish brown loam). The plowzone of STPs A1, C1, C2, D2, and D3
contained cultural materials; cultural materials were also found below the plowzone in C2,
D2, and D3. It appears that these sediments and cultural materials were deposited
relatively recently. Evidence of this is the occurrence of rusty nails in stratum C of STP
D2 at approximately 46 cm below ground surface. STP C2 contained possible flakes in
strata C and D, along with quantities of gravel. STP D3 contained a small flake in
stratum B, which also contained sand lenses. The gravels and sand lenses suggest that
sedimentation in this area is caused by a combination of both colluvial and alluvial forces.
Because no evidence of in situ cultural materials was found in the ravine on the
west side of the ridge, these clayey sediments were used in the fall of 2006 to build the
13
dam as planned. The resulting basin filled quickly with water and today contains a small
pond.
EXCAVATION METHODS
The excavation on 11MS1124, the D. Hitchins site, took place between May 24
and July 3, 2006. Excavation conditions were generally good, with rain days spent at the
SIUE Archaeology Lab washing and sorting artifacts. The last several days of the field
school, July 5-7, were also spent in the lab. This provided time to wash and sort the rest
of the artifacts, and we were also able to process all flotation samples.
As described above, in April Henry Holt used two survey pins previously set by a
professional land surveyor to establish a north-south base line running across 11MS1124
to create the grid used in the geophysical survey. On May 24 we returned to the same
two pins and used the same procedure to reestablish the baseline. SIUE field school
students supervised by Julie Holt set in a hub 55 meters south of the surveyor’s pin. This
point sat at the crest of a slope and was named N300 E300 on the SIUE grid. The survey
instrument used was a Realist-David White 4.5 inch transit, model TR300. Moving the
transit to N300 E300, we then shot in six more hubs moving south across 11MS1124,
placing them every 20 m. The placement of these hubs was later double-checked and
elevations were taken using a Sokkia SET 6E electronic total station.
Most excavation units were located to investigate geophysical anomalies thought
likely to indicate Mississippian period features. As features were found in some of these
units, it became necessary to expand them. As new units were opened to expose features
in entirety, they were typically given names to show which units they were attached to
(e.g., Unit DD was an expansion of Unit D; Unit KA was an expansion of Unit K; Units
LA, LB, and LC were an expansion of Unit L; and Units NA, NB, NC, ND, and NE were
and expansion of Unit N). Units O, P, Q, and R were opened as extensions to Units LA
and LB.
All excavation units were 2 x 2 m except for Unit DD, which was only 1 x 2 m.
The plowzone was removed from all units in a single level (stratum A, level 1) using
shovels. The base of the plowzone was troweled to look for features. If no features were
encountered, excavation was usually discontinued at the base of the plowzone. This is
typical archaeological practice in the uplands of the greater American Bottom: since the
upland loess was deposited around 12,000 years ago (Neely and Heister 1987), buried
sites are not expected in the uplands and so archaeologists ordinarily do not excavate
subsoil deposits. However, Van Nest (2002) has shown that Archaic period sites in
upland prairie and prairie-forest transition soils can be buried below the plowzone by the
activity of earthworms or other small soil fauna. On such sites, Archaic period artifacts
will accumulate within a “stone zone” on top of a Bt soil horizon, typically about 40 cm
deep in prairie-forest transition soils or somewhat deeper in prairie soils. Although
11MS1124 was forested historically, it is located close to the prairie-forest transition.
Because of this, we excavated samples of subsoil in multiple units. All excavated
sediments were dry-screened through quarter-inch mesh, except for samples retained for
flotation processing.
After features were defined by troweling, they were photographed and drawn in
plan view. Photographs were taken in color with a digital camera and in black and white
14
film with a manual camera, the latter being recommended for archival purposes. Plan
views were drawn on 20 lb. metric grid vellum at a scale of 1:10 cm. All features were
bisected; bisection lines were also drawn on the plan views during the mapping procedure
described above. Generally, the first half of the feature was removed in one level using a
trowel and dry-screened through quarter-inch mesh. After the first half of the feature was
removed, a 5-10 cm “window” was excavated around the profile edges so that the feature
boundaries would be clearly visible against the subsoil when the feature profile was
photographed. The feature profiles were photographed, again using a digital cameral for
color and a manual camera for black and white film, and then drawn on 20 lb. metric grid
vellum at a scale of 1:10 cm.
The second half of each feature was excavated using a trowel, removing each
stratum separately. Flotation samples were taken from each stratum encountered in the
second half of the feature; typically, samples were taken from the top of the feature and
also the bottom of deeper pit features. Flotation samples were usually 10 l; smaller
samples were taken when there was not enough sediment for a 10 l sample. All feature
sediments not saved for flotation were screened through quarter-inch mesh.
EXCAVATION RESULTS
Units A-I were located in order to expose an electric anomaly that suggested a
possible prehistoric structure in size and shape (see Figure 3 and discussion of the
geophysical survey above). Together these nine units created an excavation block, the
northernmost excavated, that was 6 x 6 m (Table 2; Photo 4). The plowzone in Units A-I
was 20-25 cm deep and a 10YR4/2 dark
grayish brown silt loam. No evidence of a
prehistoric structure was encountered at
the base of the plowzone, nor was it clear
what had created the electrical anomaly.
Plow scars were oriented north-south,
while the anomaly was oriented NE-SW
and NW-SE. Two circular pit features,
Features 1 and 2, were encountered at the
base of Units A/B and D. Since Feature 2
extended into the west wall of Unit D, an
additional 1 x 2 m unit, Unit DD, was
opened to expose Feature 2 in entirety.
Features 1 and 2 will be described in
greater detail below.
FCR and debitage were the most common prehistoric materials recovered from
the plowzone in Units A-I. Debitage included two hoe flakes, one of Kaolin chert from
Unit B and one of Burlington chert from Unit I. A point fragment reworked into a
scraper was also recovered from Unit B. A Terminal Archaic Prairie Lake spear point of
Burlington chert was recovered from Unit G, and a point tip of unknown chert type from
Unit G also appears to be a Terminal Archaic point. Ceramics were uncommon in Units
A-I. Units A and B each produced a single grit-tempered sherd, possibly in association
with Feature 1. Eighteen grit-tempered sherds were recovered from Units D and DD in
15
association with Feature 2. Three of these were cord-marked, one of which was
identified as S-twist. The pottery within Feature 2 was identified as early Late Woodland
(see discussion of Feature 2 below).
Of Units A-I, subsoil was excavated in one unit, Unit G, to test for the presence of
a buried Archaic component (see Van Nest 2002 and discussion above). Unit G was
selected for sub-plowzone investigation since two Archaic points were recovered in its
plowzone. Below the plowzone, Unit G was excavated as a 1 x 2 m unit in arbitrary 2 cm
levels. The subsoil was generally a 10YR5/4 yellowish brown loam, increasingly clayey
with depth (see Figure 7). Two Burlington flakes and 222 g of FCR were encountered in
the first 2 cm level below the plowzone (Stratum B, Level 2), which still showed
evidence of plow disturbance. Small amounts of FCR were encountered in the next two
levels. Two small pieces of shattered Burlington chert (perhaps debitage or FCR) were
recovered from Stratum B, Level 7. By Stratum B, Level 8 (approximately 14 cm below
the base of the plowzone and 34 cm below ground surface), pebbles were no longer
encountered. After several more arbitrary levels without pebbles, excavation was
discontinued at approximately 20 cm below the base of the plowzone and 40 cm below
ground surface. No evidence of a buried “stone zone” (Van Nest 2002) was found. This,
along with the occurrence of Archaic artifacts mixed with Woodland artifacts in the
plowzone, can be taken as additional evidence that the D. Hitchens site was forested (Van
Nest 2002).
Figure 7. Profile facing south, Units G, H, and I.
Unit J was located to investigate a small magnetic anomaly. The plowzone in
Unit J was a 10YR4/3 brown silt loam and about 25 cm deep, yielding FCR, debitage,
and ceramics. Debitage included two Cobden chert flakes and two Mill Creek chert
flakes, one of which was a hoe flake. Ceramics were grit-tempered; some were cordmarked with an S-twist, while a few were Madison County Shale with a plain surface.
An anvil was also recovered from Unit J. At the base of the plowzone, plow scars were
visible running north to south, but there was no evidence of a feature that would have
been the source of the anomaly. After removal of the plowzone, the north wall profile
was mapped and excavation was discontinued.
16
Unit K was located over a white space on the magnetometer map (i.e., there was
no evidence of a magnetic anomaly in this location). The plowzone in Unit K was a
10YR3/2 very dark grayish brown silt loam, 20-35 cm deep. Cultural materials
recovered included FCR, ceramics, debitage, a Late Woodland arrow point of Burlington
chert, and a hammerstone. Ceramics recovered were mostly grit-tempered, some of
which were cordmarked and some of which were plain. Cordmarking, when examined,
was twice as often found to be S-twist rather than Z-twist. Several shell-tempered sherds
were also recovered. The base of the plowzone in Unit K was difficult to define during
excavation due to disturbance of the subsoil, so it was first defined when a clear line of
transition was noted in the profile. However, we were still finding artifacts and the soil
continued to be dark and mottled. Two burnt areas were noted, as well as several linear
stains which we hoped might be wall trenches. Given the possibility that we might have
been inside a Mississippian house, we called the entire unit Feature 4 at the base of the
plowzone and opened up Unit KA to the south. During removal of the plowzone in Unit
KA, we received visits from archaeologists (Henry Holt and Brad Koldehoff) and the
farmer (Jim Heepke). Based on discussions with Holt and Koldehoff, we determined that
the linear stains were not wall trenches. Heepke confirmed that they were likely caused
by a chisel plow, which goes deeper than an ordinary plow.
Unit KA contained even more artifacts: FCR, debitage, ceramics, a Cahokia point
of Burlington chert, and a possible grinding tool. Ceramics were mostly grit-tempered,
but again some shell-tempered ceramics were recovered. The base of the plowzone was
again difficult to define because the subsoil was disturbed, so we stopped excavation of
the plowzone where the line of transition was visible in the profile. We removed ca. 5
cm of disturbed soil below the plowzone as Stratum A/B, Level 2. Artifacts continued to
be found to the base of excavation. At the base of excavation in Units K and KA, several
possible posts were identified. These were excavated as Features 13 (in Unit KA), 14 (in
Units K and KA), 15 (Unit K), and 16 (Unit K). Features 13-16 will be described below.
Feature 4 was determined not to be a feature, so it will not be described below. Again,
we would note that sterile soil was never reached in Units K and KA: excavation was
discontinued because we ran out of time. It is likely that disturbances noted below the
plowzone in these units were the result of both cultural (chisel plowing) and natural
(bioturbation) factors, which brought prehistoric artifacts deeper into the profile than
where they were originally deposited.
Unit L was located in order to expose a magnetic anomaly suggestive of a
prehistoric pit. The plowzone in Unit L was a 10YR3/2 very dark grayish brown silt
loam, 20-35 cm deep. It contained significant quantities of cultural materials including
FCR, debitage, and ceramics. At the base of the plowzone, it was found that the source
of the anomaly was indeed a prehistoric pit, Feature 3. Feature 3 was circular in plan
view, extending into the northeast corner of the unit. In order to expose the feature in
entirety, Units LA, LB, and LC were opened (see Table 2). At the base of the plowzone
in Unit LA, another circular pit feature was encountered, Feature 8, which appeared to
extend into the west wall. Nine post molds forming a neat semicircle (Feature 9) were
found at the base of the plowzone in Units LA and LB. In order to expose the rest of
Feature 8, Unit O was opened west of Unit LA. Units Q and R were opened north of
LA and LB in order to look for more posts in association with Feature 9. Unit P was also
17
opened at this time, joining the excavation block that began with Unit L with Units K and
KA to the west.
At the base of the plowzone in Unit O, it was found that Feature 8 actually ended
at the west edge of Unit LA and did not continue into Unit O. A possible pit (Feature 11)
and a possible posthole (Feature 12) in Unit O were excavated as features, but these were
determined to be natural disturbances by their amorphous boundaries. At the base of the
plowzone in Units P and Q, another circular pit feature, Feature 10, was encountered. At
the base of the plowzone in Units Q and R, Feature 9 was found to consist of a complete
circle of posts; however, the posts in the northern half of the feature (in Units Q and R)
were difficult to define and not evenly spaced like the posts in the southern half of the
feature.
Units L, LA, LB, LC, O, P, Q, and R, along with adjoining Units K and KA, thus
formed the southernmost excavation block (Photo 5). The plowzone in these units
contained moderate quantities of FCR, debitage, and ceramics. Three broken spear
points of Burlington chert were found in the plowzone of Units L-R. These were
identified as a probable Terminal Archaic Prairie Lake point from Unit L, a probable
early Middle Archaic Jakie stem point from Unit O, and a Middle-Late Archaic Godar
point from Unit Q. A Late Woodland arrow point of Burlington chert was found in Unit
LB. Four Mill Creek hoe flakes, a Burlington hoe flake, and a Kaolin hoe flake were
recovered from Units L, LB, LC, Q, and P. A mano was found in Unit LA. Ceramics
recovered included grit-tempered sherds, some of which were cord-marked. When twist
direction was examined, S-twisting was about three times more common than Z-twisting.
A few grit-tempered sherds were identified as Madison County shale with a plain surface.
Shell-tempered sherds were also present but less common in the plowzone than in
features below. This is not surprising since shell-tempered sherds do not preserve well
compared to grit-tempered sherds. Red-slipping was visible on a few of the shelltempered sherds. In sum, these artifacts represent multiple time periods (from Archaic to
Mississippian) and various activities (e.g. hunting, farming, and food processing).
Features 3, 8, 9, and 10 are believed to date to the early Mississippian period and will be
discussed below.
After removal of the features and possible features in the southernmost excavation
block, subsoil was excavated in selected units. The eastern halves of Units LC, LB, and
R were excavated ca. 10 cm into the subsoil, as were the northern halves of Units P, Q,
and R; and the west half of Unit O. Excavation of this arbitrary level (labeled Stratum B
Level 2) created deeper profiles for the east, north, and west walls of the excavation
block, which were then drawn. It was during this excavation that Feature 18 was
recognized at the base of the plowzone in the northeast corner of Unit R. Feature 18
extended into both the north and east walls of Unit R and was not excavated since it was
discovered on our last day in the field. After discovery of Feature 18, we did not dig any
deeper into the northeast corner of Unit R in order to leave as much of the feature
undisturbed as possible. Feature18 is believed to be Mississippian in age and will be
discussed below. The subsoil was generally a 10YR5/6 yellowish brown mottled with
10YR4/4 dark yellowish brown loam, increasingly clayey with depth. A small number of
artifacts were found in the subsoil, but these presumably descended from the plowzone
via natural processes or plowing. Excavation was discontinued ca. 40 cm below ground
surface. No evidence of a buried “stone zone” (Van Nest 2002) was found. Along with
18
the occurrence of Archaic artifacts mixed with Woodland artifacts in the plowzone, this
can be taken as additional evidence that the D. Hitchens site was forested (Van Nest
2002).
Unit M was randomly located just outside of the area surveyed with the
magnetometer. The plowzone was a 10YR4/3 brown silt loam and 20-25 cm deep,
yielding FCR, debitage, and ceramics. The few sherds recovered were all grit-tempered,
three of which were cord-marked with an S-twist and one of which was of a Madison
County shale with a plain surface. Two Mill Creek hoe flakes were identified among the
debitage. After removal of the plowzone, the north wall profile was mapped and
excavation was discontinued.
Unit N was located in order to expose a magnetic anomaly suggestive of a
prehistoric pit. The plowzone was a 10YR4/4 dark yellowish brown silt loam, 25-30 cm
deep and containing debitage, ceramics, and large quantities of FCR. In fact, six file
boxes of FCR (over 20,000 g in weight) were recovered, mostly from the northeast
quarter of the unit. Some of this FCR was still in situ at the base of the plowzone,
including a large cracked boulder,
perhaps the size of a cantaloupe (see
Photo 6). This broke into pieces as it
was removed from the ground. It
appeared that this was the base of an
FCR-filled pit which had been
truncated by plowing. Presumably this
concentration of FCR was also the
cause of the magnetic anomaly, since
no pit similar to Feature 3 (which had
a similar magnetic signature) was
encountered. The age and purpose of
this FCR concentration is not clear. It
was not given a feature number since it
was mostly contained within the
plowzone, but the rocks in situ at the base of the plowzone were bagged separately (see
“Unit N-Ap rock” on Table 11).
The ceramics recovered from the plowzone in Unit N were mostly grit-tempered;
S-twist cordmarking was approximately three times more common than Z-twist.
However, compared to other units, a relatively high percentage of shell-tempered sherds
was also recovered, most of which had a plain surface, but one of which featured a brown
slip. Again compared to other units, relatively high numbers of exotic lithic materials
were also recovered. These included three Cobden chert flakes, one Mill Creek chert
flake, one Ste. Genevieve flake, and a fragment of hematite. Relatively high numbers of
cobble tools were also recovered, including three anvils, one possible grinding tool, and
two other cobble tools.
At the base of the plowzone in Unit N, a possible pit (labeled Feature 5) was
defined extending west from the in situ rock concentration (see Figure 8). The sediment
in Feature 5 was pale, only slightly darker than the subsoil (a 10YR 5/6 yellowish brown
mottled with 10YR4/3 brown loam). A darker stain containing artifacts in the southeast
corner of the unit was labeled Feature 6. Another dark stain in the northeast corner of
19
Unit N was labeled Feature 7. To further expose Features 6 and 7, Units NA, NB, and
NC were opened. These were initially 1 x 2 m, 1 x 1 m, and 1 x 2 m, respectively.
Upon excavation, the profile of Feature 5 showed dark patches with no clear
feature boundary, indicating that it was likely the result of bioturbation. Removal of the
plowzone in Units NA, NB, and NC indicated no clear boundaries for Features 6 and 7.
All three units were further expanded until each was a 2 x 2 m unit. Clear boundaries
were still not evident for Feature 6, so Units ND and NE were opened south of Units NC
and NB, with both of these units 2 x 2 m. With the addition of these units, the excavation
block that began with Unit N reached its final size of 6 x 4 m. A dark stain in the
southeast corner of Unit NE was named Feature 17. Feature 6 was amorphous in plan
view and large, covering Unit NB in entirety and extending into all adjacent units (Figure
8). Excavation of Features 6, 7, and 17 showed that they were all amorphous and shallow
in profile. It is believed that Feature 6 was a tree throw, and Features 7 and 17 were
probably also the result of bioturbation. We would point out that the southern border of
Unit NE was just two m north of Unit K. Thus, the bioturbation evident below the
plowzone in Units N and NA-NE continued in Units
K and KA. Features 5, 6, 7, and 17 will not be
described below since they were determined to be
natural rather than cultural in origin. Upon
completion of excavation in these units, a plan view
was drawn (Figure 8) and the west-facing profile of
Units N, NC, and ND was also drawn. (The profile
is not reproduced here since it simply shows the
plowzone over subsoil.)
Units NA-NE produced a quantity of
artifacts, including some below the plowzone as a
result of bioturbation. Artifacts included a
relatively large number of projectile points: three
Cahokia points, a Late Archaic Etly point, and an
Archaic point reworked into a scraper, all made of
Burlington chert. Another Cahokia point made of
Burlington chert was found on the surface just north
of Unit N. A Burlington hoe flake and three Mill
Creek hoe flakes were recovered from this excavation block. Other exotic cherts
recovered from Units NA-NE included seven Mill Creek flakes (in addition to the three
hoe flakes), three Choteau flakes, three Cobden flakes, three Cobden projectile point
fragments, one Ste. Genevieve flake, and one flake of Fern Glen. One piece of hematite
was also recovered. Ceramics were mostly grit-tempered; of these, only one was
identified as Madison County shale with a plain surface. Grit-tempered cordmarked
sherds were much more common; of these, S-twisting was slightly more common than Ztwisting. Shell-tempered sherds were also relatively common; these were plain in surface
treatment.
20
FEATURES
Features 1 and 2 were both shallow (8-9 cm below the plowzone), circular pit
features encountered in the northernmost excavation block (see Figures 9-10; Table 3).
Feature 1 was a 10YR 4/3 brown mottled with 10YR5/3 brown and 10YR5/4 yellow
brown loam flecked with charcoal. Few artifacts were recovered from Feature 1 – just
two Burlington or glacial chert flakes and two pieces of FCR. However, copious
charcoal including charred nut shells was visible in Feature 1 during excavation.
Analysis of the floated plant remains indicates that hickory nuts (Carya spp.) were most
common; other members of the Juglandaceae family identified included black walnut
(Juglans nigra). Wood charcoal was also common, with oak (Quercus sp.) the most
common wood identified. No seeds were present.
Figure 9. Plan view, northern excavation block.
Figure 10. Profiles of Feature 1 (facing north) and Feature 2
(facing south).
We had difficulty defining the western edge of Feature 1. A 20 cm wide strip on
the west side of Feature 1 was excavated separately (labeled Stratum B) during
excavation of the second half of the feature. This could have been a zone of slumping
within the feature, or a disturbance inside or outside of the feature. Screening Stratum B
produced no cultural materials larger than ¼ inch. The 10 l flotation sample from
Stratum B produced the same types of nuts and wood found in Stratum A, but the density
of charcoal from Stratum A was 20 times greater than the density of charcoal from
Stratum B.
Feature 2 was a 10YR 4.5/3 brown mottled with 10YR 6/3 pale brown loam
flecked with charcoal. A large rim sherd was visible at the feature’s surface (see Photo 7);
all the sherds present in Feature 2 appear to represent perhaps 25% of a single early Late
Woodland jar. This jar was cordmarked with an S-twist, and its lip features interior
impressions. Other grit-tempered sherds in the plowzone from Units D and DD
presumably came from the same jar. In terms of lithic materials, just three Burlington or
glacial chert flakes were recovered from Feature 2. More FCR (27 fragments) was found
21
in Feature 2 than in Feature 1, but still it was not abundant. The types of plant remains
identified in Feature 2 were similar to those identified in Feature 1, but the density of
charred plant remains present was considerably lower in Feature 2 than in Feature 1.
Again the hickory/walnut family (Juglandaceae) was most common among the nutshell
identified. Wood charcoal again included oak (Quercus sp.), but walnut or hickory wood
(Juglans sp./Carya sp.) was more common. Again, no seeds were present.
Photo 7. Feature 2; pottery at surface in foreground.
Photo 8. Features 1 and 2, after excavation.
Based on the presence of the ceramics within it, Feature 2 surely dates to the early
Late Woodland period. It is reasonable to hypothesize that Feature 1 does also.
Evidence of this includes the proximity of the two features: they were 1.6 m apart (see
Photo 8). Evidence would also include the similarity of the two features in size and
shape. Both features contained low densities of artifacts, and identified plant types were
similar in both. In addition, an early Late Woodland Steuben point was found on the
surface very close to Feature 1. If funding for further analysis were available,
radiocarbon dating of the abundant charred plant remains recovered from Feature 1 could
be used to test the hypothesis that it too dates to the early Late Woodland period.
Photo 9. Feature 3, showing shell-tempered sherds in
southern situ in the profile and on the floor.
Photo 10. Feature 9 and surrounding features,
excavation block.
Feature 3 was the deepest pit excavated at the site, extending ca. 44 cm below the
plowzone, which is why it showed up so clearly in the magnetometer survey (see Figures
11-12; Table 3.) One zone of feature fill was identified, a 10YR 3/2 very dark grayish
brown silt loam. Grit-tempered sherds were most common in Feature 3. These were
frequently cordmarked, with Z-twisting about twice as common as S-twisting. Shell-
22
tempered sherds were less common, but several red-slipped shell tempered sherds were
found at the very base of the feature (see Photo 9). Z-twist cordmarking is typical of
Emergent Mississippian ceramics, while shell-tempered pottery is typically Mississippian.
The co-occurrence of the sherds in this pit, with shell-tempered sherds present on the
floor of the pit, suggests that Feature 3 dates to the beginning of the Mississippian period
or the end of the Emergent Mississippian period. Although debitage was less common
than pottery in Feature 3, a greater quantity of debitage was recovered from Feature 3
than from any other feature. All debitage from Feature 3 was identified as Burlington or
glacial chert. More FCR was also recovered from Feature 3 than from any other feature.
Plant remains recovered from Feature 3 included nut shell, maize, seeds of several native
cultigens (Polygonum, Chenopodium, and Hordeum), and wood charcoal. The nuts,
maize, and native cultigens are all harvested in the fall.
Figure 11. Plan view, southern excavation block.
Figure 12. Profile of Feature 3 (facing north).
Feature 8 was a circular pit located northwest of Feature 3 (see Figure 11), but it
was not as deep (23 cm; see Figure 13) and the feature fill appeared very different (a
10YR 3/3 dark brown silty loam flecked with charcoal and burnt clay). Few sherds were
recovered from Feature 8; these were all grit-tempered and frequently cordmarked, with
S-twisting more common than Z-twisting. The few flakes recovered from Feature 8 were
all Burlington or glacial chert. A small amount of FCR and burned limestone was also
recovered from Feature 8. Plant remains recovered from Feature 8 included nut shell,
maize, seeds of native cultigens (Polygonum and Phalaris), and wood charcoal. The
presence of Phalaris (maygrass) would suggest a spring harvest; whereas nuts, maize,
and Polygonum (knotweed) are harvested in the fall.
Feature 9 was a circle of posts located north of Feature 3 and east of Features 8
and 10, approximately 3 m in diameter (see Figure 11, Photo 10). The posts were small,
averaging about 7 cm wide and 6 cm deep below the plowzone (see Table 4). The
southern half of the structure was uncovered first. Posts in the southern half were much
23
clearer and easier to identify than those in the north half. This could have been because
excavation conditions were different (e.g., soil was drier) by the time we excavated the
north half. Alternatively, it could be that the north wall was rebuilt on one or more
occasions. The northwestern posts were particularly confused; since this is where winter
winds typically come from, it seems likely that these posts were replaced or reinforced.
Another possibility is that a seat or shelf could have been built into this corner, although
this seems unlikely in such a small structure. In total, there were as many as 22 posts in
the structure; these were all profiled (see Figure 14).
Figure 13. Profiles of Feature 8 and Feature 10 (both facing south).
Figure 14. Profiles of post molds in Feature 9.
A = post mold (10YR4/2.5 - 10YR4/4);
B = subsoil (10YR5/6 mottled with
10YR5/4).
Dating of post structures is difficult since typically few diagnostic artifacts are
recovered in clear association with them. In the case of Feature 9, small and crumbly
grit-tempered sherds were recovered from several post molds. Their presence suggests
that the structure postdates the Archaic period. Circular post structures are typical of the
Middle Woodland period in the American Bottom; however, Middle Woodland structures
are typically much bigger than 3 m in diameter (e.g., see Fortier 1993). Moreover,
although Dave Klostermeier found Middle Woodland artifacts perhaps 20 m south of
Feature 9, we recovered no Middle Woodland artifacts from the vicinity of the structure.
Early Late Woodland people in the American Bottom built post structures of varying
shape, and they were typically small like Feature 9. In fact, a circular post structure
excavated at the Cunningham site was even smaller than Feature 9 (see Fortier 1993:
Figure 6). Thus, Feature 9 could be early Late Woodland based on its style; however, no
artifacts securely dated to the early Late Woodland period were found in the vicinity of
Feature 9. A late Late Woodland (Patrick phase) or Mississippian age is more likely
given the artifacts we recovered from this area.
Typical domestic structures from the Late Woodland Patrick phase through the
Mississippian period feature semi-subterranean basins and are square or rectangular in
shape (see Fortier 1993). Clearly, Feature 9 is not a typical late prehistoric domicile.
However, circular structures are also found on Mississippian sites. Circular structures
with hearths are typically interpreted as sweat lodges whereas circular structures without
hearths are typically interpreted as storage facilities (see Fortier 2007). Given the
24
proximity of Feature 9 to pit Features 3, 8, 10, and 18 – which appear to encircle Feature
9 – it is believed that these features are contemporary in age. Ceramics from Features 3
and 10 suggest an early Mississippian or late Emergent Mississippian age. Similar
Mississippian structures excavated at the East St. Louis site have been interpreted as
small storage huts (Fortier 2007; Pauketat 2005). We will argue below that Feature 9 is
most likely an early Mississippian structure for sheltering food procurement and
processing activities.
Feature 10 was located just north of Feature 8, and it was essentially identical in
size and shape to Feature 8 (see Figure 13). At its surface, the fill in Feature 10 also
looked very similar to the fill in Feature 8 (see Photo 11); their fill was notably lighter
and less homogenous than the fill in Feature 3. The fill in Feature 10 was mostly a
10YR4/3 brown mottled with 10YR5/4 yellowish brown loam; however, there was a
darker lens (labeled Stratum B) toward the bottom of the feature described as a 10YR3/3
dark brown mottled with a 10YR4/4 dark yellowish brown loam flecked with charcoal
and ochre. This lens was floated in its entirety and turned out to have the greatest density
of charcoal relative to all other flotation samples. It contained nut shell, seeds of native
cultigens (Polygonum, Phalaris, Helianthus, and Iva annua), and wood charcoal; but
most of all, it was packed with maize. Fewer seeds were present in the other flotation
samples from Feature 10, but otherwise taxonomic representation was similar. The
presence of Phalaris (maygrass) would suggest a spring harvest; whereas other native
cultigens, nuts, and maize are harvested in the fall.
Photo 11. Profiles, Features 8 and 10.
Ceramics were mostly grit-tempered, including both cordmarked and plain
surfaces. Two grit-tempered sherds were identified as having an S-twist cordmark, while
one was a Z-twisted rim sherd, cordmarked to the lip. Three shell-tempered sherds were
also recovered. The Z-twist rim sherd and shell-tempered sherds were much larger than
most sherds, which were small and crumbly. The co-occurrence of these sherds suggests
that Feature 10 like Feature 3 dates to the beginning of the Mississippian period or the
end of the Emergent Mississippian period. Other materials recovered from Feature 10
were FCR, Burlington or glacial chert flakes, and burnt clay.
A sample of maize from the very bottom of Feature 10 was submitted to the ISGS
for radiocarbon dating. The age returned was 1750 ± 110 radiocarbon years BP (ISGS-
25
6040). This would suggest that Feature 10 dates to the Middle Woodland period, which
is impossible. First, maize on Middle Woodland sites is rare. Second, the ceramics
recovered from Feature 10 indicate that the pit dates to the early Mississippian or late
Emergent Mississippian period, which is consistent with the abundance of maize present
in the pit. No Middle Woodland artifacts were recovered in the vicinity of Feature 10.
Because of the problems with the age obtained on the first sample, a second
sample of maize from Feature 10 was submitted to the ISGS, this time for AMS dating.
The second sample came from the center of the feature. The age returned was 905 ± 20
radiocarbon years BP (ISGS A1263). This age is consistent with the early Mississippian
or late Emergent Mississippian ceramics recovered from the pit.
Figure 15. Profiles of Features 13, 14, 15, and 16.
Photo 12. Feature 18.
Features 13, 14, 15, and 16 were identified in Units K and KA (see Figure 11).
They ranged between 20 and 40 cm in diameter, and typically extended about 10 cm in
depth below the plowzone. Although they produced few cultural materials, their profiles
suggest they were cultural features, even though Feature 13 appeared to have a root or
rodent hole descending from the east end of its base (see Figure 15). Features 14-16 each
produced one or two small grit-tempered sherds, suggesting that they post-date the
Archaic period. Feature 15 produced one flake of Burlington or glacial chert. Features
15-16 also produced small amounts of FCR. Flotation samples were taken, but they have
not been submitted for paleoethnobotanical analysis. The function of these features is
unknown; perhaps they were post holes.
Feature 18 was identified northeast of Feature 9 (see Figure 11). Since it was
discovered on the last day of the excavation, it was not excavated (see discussion of Unit
R above). It extended into the north and east walls of Unit R, so its shape in plan view is
unknown. However, the fill in Feature 18 appeared very similar to the fill in Features 8
and 10, which lay on the other side of Feature 9 (Photo 12). (The fill in Feature 18 was
described as a 10YR 3/3 dark brown loam flecked with charcoal and red ochre.) Reexamination of the map produced by magnetometer survey (see Figure 3) shows that all
three features produced a faint magnetic signature. It is therefore speculated that this is
another pit like Features 8 and 10, which along with Feature 3 encircled the circular post
structure, Feature 9.
26
CERAMICS
Ceramics were identified by Miranda Yancey with assistance from John Kelly
and Julie Holt. Inspection of sherds was macroscopic, using a 10x lens. Temper type
and surface treatment were recorded whenever possible for all body and rim sherds;
additional observations regarding lip treatment, profile, paste, diameter, and
chronological phase or period were also made for rim sherds. All cordmarked sherds
were examined to determine twist direction. The method for examining twist direction
was to press white Sculpey clay into the sherd; then the impressions in the clay were
examined (after Drooker 1992).
A total of 3711 sherds weighing 4426.11 g were collected (see Tables 5 and 6).
The vast majority of these were highly fragmented sherds recovered from the plowzone.
In fact, the only sherds of any size were recovered from Feature 2 and from the base of
the plowzone in Unit N and adjacent units. In the case of Unit N and the adjacent area, it
would appear that the FCR concentration protected sherds found in disturbed soil at the
base of the plowzone by deflecting the plow as it passed. The rim sherd from Feature 2
and a rim sherd from Feature 10 were the only rim sherds recovered. Given the area
excavated, this is an extremely small ceramic sample and illustrates continuity in site
function over time: multiple lines of evidence suggest that occupation at D. Hitchins
during all periods was short-term and probably seasonal. Apparently, a complete set of
dishes (pots, pans, etc.) was not needed.
Given that most ceramics were recovered from the plowzone, there are few
published sites in the area for valid comparison since most American Bottom
archaeologists do not screen the plowzone. We screened the plowzone in the 2002 SIUE
field school excavation at AE Harmon (Holt et al. 2005). At that site we excavated some
44 square meters, uncovering three Late Woodland pits, a structure with a basin, and a
ceramic sample of 4215 body sherds and 95 rim sherds. At D. Hitchins we excavated
some 108 square meters but recovered fewer features and fewer sherds. Unfortunately,
we did not record the weight of the ceramics from AE Harmon, but given that sherds
from AE Harmon were generally larger, particularly those recovered from features, the
weight of that sample would have been substantially greater than the sample from D.
Hitchins.
Grit-tempered sherds were by far most common in the ceramic sample. Grittempered pottery was more commonly cordmarked than plain, and S-twisting was more
common than Z-twisting. The only substantially complete vessel found at the site was a
grit-tempered, early Late Woodland jar recovered from Feature 2; perhaps 25% of the jar
was recovered (see Photos 7, 13). This jar featured S-twist cordmarking up to the lip,
which is rounded and has interior lip impressions apparently made with a plain dowel
(see Koldehoff and Galloy 2006). The diameter of the jar’s opening was approximately
20 cm. It appears most like Jar Type 1, again using nomenclature from Koldehoff and
Galloy (2006), which is to say that it has an inslanting-outcurved profile (also see
Jackson 2007). It is clear from Jackson’s discussion that we should not attempt to assign
a single jar such as this to a phase given similarities in ceramics during the early Late
Woodland Rosewood, Mund, and Cunningham phases.
27
Photo 13. Early Late Woodland rim sherd, Feature 2.
Photo 14. Emergent Mississippian rim sherd, Feature 10.
The only other rim sherd recovered from the site was also grit-tempered and
cordmarked to the lip, but this was a Z-twisted sherd from Feature 10 (see Photo 14). Its
lip was squared with no visible decorations. The sherd was from a jar with an inslanting
profile, but the section of rim present was too small to determine the jar’s diameter. The
presence of Z-twisting suggests that the sherd is Emergent Mississippian, although
determining phase is again problematic. Only a few other sherds from Feature 10 appear
to be from the same jar, and none exhibit features diagnostic to phase. Two S-twisted
cordmarked sherds were also found in Feature 10 as were several shell-tempered sherds.
Given the small size of these sherds and incompleteness of the several vessels that they
represent, they appear to be sheet refuse that was swept into or washed into the pit. The
same could be said of the ceramics found in Features 3 and 8.
Shell-tempered sherds were found in Features 3 and 10. Shell-tempered sherds
were also relatively abundant in Units N and NC, in the vicinity of and south of the FCR
concentration (note that Features 5 and 6 in Table 5 were actually determined to be
natural disturbances; they were located within Units N and NC). Red-slip was visible on
a number of shell-tempered sherds, while brown-slip was positively identified on only
one sherd. The presence of shell temper suggests that these sherds are Mississippian or
perhaps late Emergent Mississippian, although determination of phase is again
problematic given that no shell-tempered rims were recovered.
A small quantity of burned clay was found in Feature 10. In total, 72 fragments
weighing 50.36 g were recovered. It seems doubtful that these fragments represent the
manufacture of ceramics given the small quantity of burned clay recovered.
CHERT ARTIFACTS
Chert debitage and tools were analyzed by Erin Marks with guidance from Brad
Koldehoff. Visual inspection of chert artifacts was macroscopic. All chert artifacts were
identified to raw material when possible, examined for use wear, and finally counted and
weighed.
Some 5140 chert artifacts weighing 4194.21 g were collected (see Tables 7-9).
The most common chert types recovered were, not surprisingly, locally available (see
Table 7). “Local” types include Burlington chert, even though the closest outcrops of
28
Burlington are probably some miles distant along the Mississippi River bluffs. Glacial
chert is truly local, its source being chert cobbles found in locally exposed glacial till.
Given difficulties in macroscopically distinguishing between some Burlington chert and
local glacial cherts (see Koldehoff 2006), we did not attempt to separate out debitage of
one type vs. the other. For making tools, however, Burlington was clearly preferred (see
Tables 7-9). As Koldehoff (2006) notes, glacial cherts are of “limited utility.” They
were apparently not used to make formal tools at D. Hitchins.
By far the most common extra-local chert identified, according to weight, was
Salem chert (see Table 8). The closest known outcroppings of Salem chert are found in
southwest St. Clair County (Koldehoff 2006), but Ray indicates that “potential” deposits
might be found even closer, in northwest Madison County (see Ray 2007: Figure 8.49).
In terms of number, the most common exotic chert was Mill Creek chert, which comes
from southern Illinois. Mill Creek chert was commonly used to make hoes in the
Mississippian period, and in fact nearly half of the Mill Creek chert identified at D.
Hitchins consists of hoe flakes. Cobden (Dongola) chert was also relatively common;
this chert is also from southern Illinois. Several Cobden point fragments were found, as
was a nearly complete Steuben point made of Cobden.
Other extra-local cherts identified were Choteau, Ste. Genevieve, Kaolin, and
Fern Glen (Table 8). Outcrops of Chouteau chert are common in the lower Illinois River
Valley, and Fern Glen is also exposed near Grafton (Koldehoff 2006). The closest
known outcroppings of Ste. Genevieve chert are found in southwest St. Clair County
(Koldehoff 2006), but as with Salem chert, Ray indicates that “potential” deposits might
be found even closer, in northwest Madison County (see Ray 2007: Figure 8.51). Kaolin
chert, like Mill Creek and Cobden chert, comes from Union County in southern Illinois
(Ray 2007).
In sum, most cherts used at D. Hitchins were derived
from sources close by. Notably, all but one of the spear and
arrow points found at D. Hitchins were made of Burlington
chert. Other sources of chert were the lower Illinois River
Valley, southwest St. Clair County, and southern Illinois
(Union County). Surely the trade systems (routes,
negotiations, etc.) by which these “exotic” cherts were
obtained varied through time. It is likely, for example, that
Cahokia played some role in the procurement and
distribution of Mill Creek hoes during the Mississippian
period (e.g., see Brown et al. 1990). Perhaps similarly
formal systems of exchange existed for Cobden chert during
the Middle Woodland period. Cherts from closer sources in
the lower Illinois Valley might have been obtained directly
by quarrying expeditions.
All projectile points were recovered from the surface
or from the plowzone (Table 9; Photo 15). Two or more
Archaic points were found in all large excavation blocks. A
Steuben point was recovered adjacent to the northern excavation block; this was most
likely contemporaneous with the early Late Woodland pottery found nearby in Feature 2.
Two Late Woodland or Emergent Mississippian arrow points were found in the southern
29
excavation block. Three Mississippian Cahokia points were found in or next to the
central excavation block, and one was found in the southern excavation block. Thus,
Archaic points were found across the site, early Late Woodland points were found at the
north end of the site, Late Woodland-Emergent Mississippian points were found at the
south end of the site, while Mississippian points were found in the central and south part
of the site.
Other evidence of flaked tools includes hoe flakes, which are produced when
sharpening stone hoes. Most hoe flakes were of Mill Creek chert and most likely date to
the Mississippian period. Like the Cahokia points, these were also found in the central
and southern excavation blocks. Several hoe flakes of Burlington chert were also found;
one was found in each excavation block (north, central, and south). A number of utilized
Burlington flakes (flake tools) were recovered from the central and southern excavation
blocks, and two Mill Creek flake tools were recovered from the southern excavation
block.
It is noteworthy that none of the “exotic” cherts at D. Hitchins were found in
features, nor were any of the spear points, arrow points, or even flake tools. It seems that
these objects were not intentionally discarded by disposing of them in pits with other
garbage; they were more likely lost on the surface of the site. The same pattern was
noted in SIUE excavations at the AE Harmon site (see Holt et al. 2005).
OTHER LITHIC ARTIFACTS
Other lithic artifacts were also analyzed by Erin Marks with guidance from Julie
Holt and Brad Koldehoff. Visual inspection of lithic artifacts was macroscopic. All
lithic artifacts were identified to raw material when possible, examined for use wear, and
finally counted and weighed.
Cobble tools were recovered from the plowzone in the central excavation block
and the southern excavation block (where Emergent Mississippian or Mississippian
features were encountered), but none were recovered from the northern excavation block
(where early Late Woodland features were located). Cobble tools identified included a
mano, other possible grinding tools, anvils, a burnt sandstone abrader, and a
hammerstone (Table 10). A fragment of a cobble tool – a possible mano – was found in a
postmold of Feature 9, which is believed to be late Emergent Mississippian or early
Mississippian in age.
FCR (fire-cracked rock) was by far the most common artifact recovered from the
plowzone, from features, and also from disturbances determined to be natural rather than
cultural in origin (features in quotation marks on Table 11). The category “chert/FCR” is
chert that also appears to have been used as FCR (Table 11). The FCR concentration in
the northeast quarter of Unit N stands out: over 21,000 g of FCR was recovered from
Unit N, with roughly half of that bagged separately at the very base of the plowzone
when the concentration became obvious (“Unit N-Ap rock” on Table 11). Included in the
cluster still in situ at the base of the plowzone was a cracked boulder about the size of a
cantaloupe (see Photo 6). Adjacent units (Units NA, NB, etc.) also contained relatively
large quantities of FCR; presumably some of this came from the FCR cluster in Unit N.
The limestone recovered was all burned (Table 11), possibly in processing food
such as maize. The quantity was extremely small, which is not surprising given that the
30
closest source of limestone was probably the Mississippi river bluffs. The greatest
concentration was recovered from Feature 8, which contained just three pieces (34 g) of
burnt limestone.
Minerals recovered included small quantities of limonite and hematite (Table 11).
Limonite and hematite are minerals that were used by people during several periods in
Illinois prehistory; those recovered from D. Hitchens were all found in the plowzone,
leaving us unable to date them with any certainty. These could have been used in
pigment production (e.g., see Koldehoff and Galloy 2006), or they may not have been
used by site inhabitants at all.
PLANT REMAINS
Plant remains are present in charcoal samples opportunistically collected during
excavation at D. Hitchins, but these have not been examined. Instead, all plant remains
discussed here were recovered from features through flotation processing. Flotation
samples were taken from each stratum encountered in the second half of the feature
excavated; typically, samples were taken from the top of the feature and also from the
bottom of deeper pit features. Flotation samples were usually 10 l; smaller samples were
taken when there was not enough sediment for a 10 l sample. Flotation samples were
processed by field school students at SIUE’s Anthropology Lab. The volume of each
sample was measured and recorded before processing. The samples were then separated
into light and heavy size fractions using a Flote Tech water flotation machine.
Paleoethnobotanical samples from Features 1, 2, 3, 8, and 10 were identified by
Marge Schroeder using comparative collections at the Illinois State Museum. Light
fractions were separated using 2 mm, 1 mm, and .5 mm geologic sieves. For these
samples the > 1 mm, > .5mm and < .5mm size fraction were carefully scanned for seeds
and remains such as cucurbit rinds, bark, modern seeds, and wild beans at 10-30x
magnification. The >2 mm light fraction was separated into the categories of wood and
nutshell under low magnification (10-30X). Both categories were weighed, counted, and
identified to the lowest possible taxonomic level.
Features 1 and 2, both believed to date to the early Late Woodland period (see
above), contained nutshell and wood charcoal (Table 12). Seeds were noticeably absent
from these samples (Table 13), suggesting that early Late Woodland people were in this
wooded area specifically to collect and process nuts. The sample from the main zone
(Stratum A) in Feature 1 was especially dense, containing quantities of both nutshell and
wood. Among the nutshell, Carya spp. (hickory nut) was common, and Juglans nigra
(black walnut) was also present. Hickory was not specifically identified in Feature 2,
although most nutshells here too belonged to Juglandaceae (the hickory/walnut family)
and black walnut was again present. Notably, early Late Woodland people at D. Hitchins
were not selecting acorns to eat, although these would have been most abundant in the
area.
The most common wood identified in Feature 1 was Quercus (oak), while the
most common wood identified in Feature 2 was Juglans sp. (walnut) or Carya sp.
(hickory). This is what we would expect in the oak/hickory forest that typifies the
uplands above the American Bottom (cf. Johannessen 1984). Wood of the Ulmaceae (the
elm/hackberry family) was also identified in Feature 1.
31
Features 3, 8, and 10 are believed to date to the end of the Emergent
Mississippian period or the beginning of the Mississippian period; they all contained
nutshell, wood charcoal, maize, and seeds. These features contained a very different
floral assemblage compared to Features 1 and 2, most notably in the diversity of wood
taxa used and in the presence of domesticated plants as well as “weeds.” Feature 10
stands out among the three Mississippian features for having the greatest density of
charcoal, particularly from the small lens at the center of the feature (Stratum B).
One clear avenue of continuity in plant exploitation from the early Late Woodland
occupation to the early Mississippian occupation was in the selection of nut resources.
Hickory nuts and black walnuts were common in Features 3, 8, and 10, suggesting a
similar pattern of nut exploitation compared to the early Late Woodland samples. Again,
no acorns were identified, although they would have been common in the area.
People continued to select wood of oak, walnut, and hickory trees around the
beginning of the Mississippian period. However, besides these trees, a much greater
diversity of woody taxa suggests an opening up of the forest at this time. In fact, wood of
the elm/hackberry family is actually more common in Mississippian samples than oakhickory. Other taxa identified are Prunus sp. (cf. wild black cherry), Platanus
occidentalis (sycamore), and Acer sp. (maple/boxelder). Wood of Vitis sp. (grape vine)
was also identified in Feature 10; grape vine commonly grows on trees in the area. The
presence of grass stems and seeds, as well as seeds of Solanum ptycanthum (nightshade)
and Euphorbia maculata (spurge), also indicate forest clearance.
The reason why the woods were more open at this time is evident in the presence
of maize and seeds of indigenous cultigens. That is, the woods were cleared for farming,
most likely using slash and burn techniques (e.g., see Koldehoff and Galloy 2006). Zea
mays (maize) kernels and cupules were most common, especially in Feature 10; glumes
were also common. Only two maize embryos and one cob section were recovered; these
were all from Feature 10. The abundance of maize and variety of maize parts is strong
evidence that inhabitants of the D. Hitchins site were growing and processing maize at
the site.
Indigenous cultigens were generally less common than maize in the flotation
samples; however, a variety of species was present. Seeds of Polygonum erectum (erect
knotweed) were by far the most common among the native cultigens. Other native
cultigens identified (see Table 13) were Chenopodium (chenopod, or lamb’s quarters),
Phalaris caroliniana (maygrass), Helianthus annuus (sunflower), Hordeum pusillum
(little barley), and Iva annua (marsh elder). The rind of a cucurbit (squash) was also
identified, although these preserve poorly (see Table 12). While these seeds and the
squash rind were not specifically measured to determine domestication, these taxa are
commonly assumed to be cultivated at this time based on multiple lines of evidence.
Some, such as Chenopodium, show morphological evidence such as thinner seed coats;
some, such as maygrass, become abundant outside their natural ranges; and others are
simply more abundant in the archaeological record than they are expected to be in nature.
At the very least, the presence of these weedy species also indicates forest clearance, and
we can safely assume that humans were intentionally selecting and most likely cultivating
these plants. In the American Bottom, people continued to rely on these native cultigens
even after maize was introduced to the area and became a staple (e.g., see Johannessen
1984).
32
In sum, the floral evidence suggests that early Late Woodland people at D.
Hitchins had a fairly narrow purpose in mind: to collect and process hickory nuts and
walnuts in the fall. In contrast, early Mississippian people continued to collect hickory
nuts and walnuts, but they also cleared the forest to grow maize and a variety of native
plants. This is not to suggest that the early Mississippian diet was necessarily more
diverse than the early Late Woodland diet. For early Late Woodland people, the D.
Hitchins site was surely just one stop in a seasonal round, and they probably had gardens
elsewhere. Early Mississippian activities took place at D. Hitchins during multiple
seasons: seeds were planted in the spring, crops were tended in the summer, and crops
and nuts were harvested in the fall. Maygrass would be an exception to these
generalizations: it would presumably be planted in the fall for spring harvest. Some
maize may have also been harvested green in the summer, but this would leave little trace
in the archaeological record.
ANIMAL REMAINS
Animal remains were identified by Julie Holt using her comparative collection at
SIUE. Bone preservation was extremely poor due to the acidity of the forest soils at D.
Hitchins. Four mole humeri, one mole radius, and the distal tibia of a cottontail rabbit
were recovered from the plowzone. The excellent condition of these bones indicates that
they are modern. Less than one gram of calcined bone was recovered from Feature 3.
Unfortunately, this bone could not be identified to taxa, giving us no direct evidence of
prehistoric animal exploitation at the site.
HISTORIC ARTIFACTS
A small number of historic artifacts recovered from the plowzone during our
excavation at the D. Hitchins site attest to 20th and possibly 19th century activities at the
site. In the northern excavation block, a small piece of rusted iron and a small piece of
white ceramic were recovered from Unit A, and four brick fragments (weighing 52 g)
were recovered from Unit I. In the southern excavation block, a lead bullet was
recovered from Unit K and another rusted object (a bolt?) was recovered from Unit P.
There is no indication based on historic maps or considering these artifacts that a historic
structure ever existed in the area.
SUMMARY AND DISCUSSION
During our surface surveys in 2006, we revisited sites 11MS1124 (the D.
Hithcens site), 11MS1125, 11MS1126, and 11MS1127; and also recorded a new site,
11MS2337, in the southwest corner of the project area. Site forms were completed and
submitted to the IAS for all of these sites. Notably, our surface collections led us to
redefine the boundaries of 11MS1124 and 11MS1127. In addition, we were able to show
an Archaic presence at 11MS1127, which previously was reported to be Late Woodland
in age.
Archaic use of the project area was extensive. Archaic peoples definitely made
use of 11MS1124 and 11MS1127, as demonstrated by the presence of Archaic period
33
spear points collected during surface collection of 11MS1127 and excavation of
11MS1124. In addition, it is reasonable to suggest that the lithic scatters at 11MS1125,
11MS1126, and 11MS2337 could also date to the Archaic period, given the lack of
ceramics noted at those sites. Of course, hunting parties probably visited those locations
during all time periods, given that the Klostermeier collection has yielded projectile
points from the Paleoindian period to the Mississippian period, with only Early
Woodland period projectile points possibly absent.
It is less clear how intensive Archaic use was of the project area. There was no
indication of Archaic features in our shovel testing of 11MS1125, nor was there any
indication of Archaic features in our excavations at 11MS1124. However, if shallow
features had been present at either site, they could have been obliterated by plowing. In
addition, 11MS1125 appears to be heavily eroded. Dave Klostermeier (personal
communication, 2005) has also suggested that there probably were once Archaic features
at the highest point in elevation, the north end of the lithic scatter that we have redefined
as 11MS1127. If so, they also would have been plowed and eroded away.
Our excavations on 11MS1124, the D. Hitchins sites, were undertaken with
particular interest in Mississippian use of the site. However, in addition to finding
Archaic points across the site, we also found what are probably two early Late Woodland
features, both shallow pits, in the northern excavation block. One of these features
(Feature 2) yielded a single early Late Woodland jar, while the second feature (Feature 1)
yielded no diagnostic artifacts. A Steuben point was found on the surface close to
Feature 1, supporting the assertion that it too is early Late Woodland in age. In addition,
the two features were close to each other, similar in size and shape, and contained a
similar suite of plant remains. These plant remains suggest early Late Woodland people
were in this area to collect and process hickory nuts and walnuts during the fall. Logic
and the spear point suggest that they also would have taken the opportunity to hunt while
they were here. The wood taxa present suggest the area was then an oak-hickory forest,
as would be expected. Presumably this was just one stop in a seasonal round for early
Late Woodland people, who likely grew native cultigens elsewhere. Although the small
number of flakes in Features 1 and 2 were all Burlington chert, the use of Cobden chert to
make the Steuben point shows that these people were involved in a larger social arena.
These features are apparently the first early Late Woodland features to be
excavated in the uplands of Cahokia Creek, and in the uplands of the northern American
Bottom. The closest excavated early Late Woodland sites are the Cunningham site
(Meinkoth et al. 2001), which is a bluff base site some 15 km from the D. Hitchins site,
and the Widman site, a bluff top site some 20 km distant (see McElrath and Fortier 2000).
The Cunningham site is notable for being one of the larger early Late Woodland sites
excavated in the American Bottom; it was interpreted as a “single-component
horticultural hamlet” (Meinkoth et al. 2001:185). It is also notable that it could not
clearly be assigned to either the Rosewood or Mund phase. That is, its ceramics bear
traits that could fit into either phase, but radiocarbon dates mostly place the Cunningham
site within the Mund phase. Jackson (2007) summarizes recent efforts by the Illinois
Transportation Archaeological Research Program (ITARP) to redefine these phases. He
proposes that Cunningham be considered a separate phase contemporary with Mund, at
least until additional data are recovered which might better clarify relationships between
Rosewood, Mund, and Cunningham. In any case, it is clear from Jackson’s discussion
34
that it would be imprudent to assign D. Hitchins to a phase, given that only one jar was
recovered from the site and given continuity in ceramic traits throughout the early Late
Woodland period.
The Widman site was a small site, with only 11 features excavated, whereas some
124 features including four structures and a single burial were excavated at the
Cunningham site (McElrath and Fortier 2000). McElrath and Fortier (2000:114) suggest,
“The smaller residential camps… which are ubiquitous in bluff-top and upland
environments, were no doubt tied to larger communities such as Mund or Cunningham.”
The D. Hitchins site was probably one such small camp. It could have been tied to the
Cunningham site itself, but we have no direct evidence for this: first, we cannot be sure
the two sites are contemporary, given our inability to assign D. Hitchins to a phase, and
second, surely there were areas equally rich in nuts and game closer to Cunningham than
D. Hitchins. McElrath and Fortier suggest that upland areas were probably used
primarily as hunting territories in the Middle Woodland period, but “became a focus of
horticultural production” during the early Late Woodland period (2000:115). In the case
of D. Hitchins, horticultural production took place elsewhere. There is no evidence of
gardening or forest clearing at the site during the early Late Woodland period.
In contrast, there is rich evidence of forest clearance and crop production at D.
Hitchins during the early Mississippian period. Paleoethnobotanical remains indicate a
greater diversity of trees in the area, as well as grasses and other plants that would be
expected in cleared areas. Among the crops grown, maize was most abundant, but
knotweed and other native cultigens were also grown. Hoe flakes, most commonly of
Mill Creek chert, provide additional evidence of Mississippian agriculture. Woods (1986)
suggests Mississippians chose this area because floodplain (Wakeland) soils below the
site were well suited to Mississippian agriculture, but we would add that the upland soils
in the area are also suitable for farming (see Yancey 2007 for a formal site catchment
analysis with respect to soils). Whereas the fertility of the floodplain soils would have
been replenished by annual floods, the fertility of upland forest soils would have been
dependent on the type of fallowing system typical of slash-and-burn agriculture. The
upland prairie soils a short distance from the site probably would not have required
fallow periods, or certainly much shorter fallow periods, but they would have been more
difficult to cultivate with stone hoes. While we did not find any celts during our
excavation, several broken celts in the Klostermeier collection probably date to the
Mississippian period and provide additional evidence of forest clearance (see Yancey
2005).
Other evidence shows that early Mississippian people at D. Hitchins did more
than just farm. Like their early Late Woodland predecessors, they continued to collect
hickory nuts and walnuts. We would point out that neither group collected acorns,
although they would have been most abundant in the area. The presence of Cahokia
points indicates Mississippian people also hunted, probably targeting animals that preyed
upon their cultivated crops and nut crops as well. The bones of any animals they hunted
were not recovered during our excavations, at least in part because of the poor
preservation that results from acidic soils. However, we would point out that less than a
gram of calcined bone was recovered from a Mississippian feature. If Mississippian
hunters had consumed their game here, it seems likely that more calcined bone would
35
have been recovered. Abundant charred plant remains were recovered. Perhaps any
game they killed was consumed elsewhere.
The small amount of ceramics recovered from Mississippian features suggests
that Mississippian people did not live at the site. As suggested above, they did not need a
full set of dishes, pots, or pans. The lithic assemblage also indicates limited activities
beyond farming, hunting, food processing, and maintenance of tools needed for those
activities. We did not uncover any evidence of Mississippian style housing – that is, we
found no wall trench structures – to suggest that people were actually living here during
the Mississippian period. We excavated only a small portion of the D. Hitchins site – less
than 5% of it – and it is certainly possible that wall trench structures are present in
unexcavated portions of the site. Nevertheless, strong corroborating evidence that there
was no intensive long-term Mississippian occupation at D. Hitchins was provided by the
SIUE survey of the early 1980s (see Woods 1986). In comparison with 11MS341, the
Kruckeberg No. 1 site, D. Hitchins produced significantly fewer Mississippian materials.
As a result, Woods (see Woods 1986 and Woods and Holley 1991) identified 11MS341
as a village and 11MS1124 as a farmstead. Based on his years of collecting both sites,
Dave Klostermeier (personal communication, 2005) agrees with the hypothesis that
11MS341 was a Mississippian village and 11MS1124 saw much lighter use by
Mississippians. However, Klostermeier thought that D. Hitchins could have been a ritual
site, perhaps a burial site. We will consider this idea below.
Woods and Holley (1991) suggest that both 11MS341 and 11MS1124 (the D.
Hitchins site) date to the Moorehead phase. However, the Mississippian features
excavated by the SIUE field school at 11MS1124 more likely date to the Lohmann phase
(early Mississippian period) than the Moorehead phase (mid-late Mississippian period).
A single rim sherd from Feature 10 was grit-tempered and cordmarked to the (squared)
lip with Z-twist cordage. However, shell-tempered sherds were also found in Feature 10,
and red-slipped shell-tempered sherds were found on the floor of Feature 3. Thus, the
small sample of ceramics we recovered from features in the southern excavation block
included both grit-tempered sherds and shell-tempered sherds, most of which were small
and appeared to be sheet refuse either swept or washed into the features. John Kelly
(personal communication, 2006) suggests that the assemblage of sherds from Features 3
and 10 could be considered either terminal Emergent Mississippian or early
Mississippian. Kelly also points out that assigning the assemblage to a phase is difficult
given the lack of comparative excavation data from the area.
Although two Late Woodland (Emergent Mississippian) arrow points were found
in the southern excavation block, a Cahokia point was also found in the southern
excavation block, and four more were found nearby in the central excavation block.
Seven Mill Creek hoe flakes were recovered from the southern part of the site, and three
more were recovered from the central excavation block. These data confirm
Mississippian use of the area. Moreover, an AMS date from Feature 10 indicates that the
pit, and presumably associated features, dates to approximately 905 radiocarbon years BP
(see above).
The features found in the southern excavation block consist of a small circular
post structure encircled by four pits, three of which were excavated. One of these,
Feature 3, was a relatively deep pit best suited for storage. The other two excavated pits,
Features 8 and 10, were more shallow and better suited for food processing. Given the
36
abundance of charred maize, especially in Feature 10, it is easy to imagine that these pits
were used to dry maize for storage. The presence of charred seeds of native cultigens and
nuts suggests that these features were also used to dry other plant foods as well. Feature
3 could have been used to store food, or it could have been used to store seed for next
year’s planting. The latter explanation seems more likely, given that we found no direct
evidence that Mississippian people actually lived at this site.
The circular post structure, Feature 9, is not a typical Mississippian house, which
would be a wall trench structure with a basin. Circular structures on Mississippian sites
are usually interpreted as either ritual structures (e.g., sweat lodges) or storage facilities
(e.g, see Fortier 2007). As mentioned above, Klostermeier believed that D. Hitchens was
a ritual site because of several unusual Mississippian objects he found at the site. These
included a Cahokia-style discoidal (chunkey stone) crudely engraved with a weeping eye
on one side and sunburst on the other, a very small washer-like discoidal (ca. 4 cm in
diameter) with a hole ground through its center, and a flared celt (Yancey 2005). While
the lack of an internal hearth could be taken as evidence that Feature 9 was not a sweat
lodge, a shallow hearth could have been obliterated by plowing, and significant quantities
of FCR were found nearby in the central excavation block. However, the most
convincing evidence that Feature 9 was not a ritual structure is its association with food
processing and storage pits. We would not expect such activities to take place outside a
sweat lodge, menstrual hut, or other ritual structure.
The alternative interpretation, that Feature 9 was a storage facility, also seems
unlikely. A storage facility would seem likely at a Mississippian village, but not at a site
like D. Hitchins without evidence of a residential population of any size. Plant remains
indicate the site was in use or at least visited throughout the spring, summer, and fall, but
there is no evidence of winter occupation. Above ground storage is vulnerable to theft
and would not make sense unless a site was occupied by a sizable group of people year
round to defend it. Below ground storage in pits like Feature 3 makes more sense at a
site that might be abandoned seasonally. We believe that Feature 3 may have been used
to store seed over winter that would have been planted at the site in the spring.
Rather than ritual or storage, then, we believe that Feature 9 was a temporary
shelter for the activities that took place at D. Hitchins during the early Mississippian
period. The most important of these activities was probably farming. Still, we don’t feel
the term “farmstead” is appropriate to describe D. Hitchins unless and until a residential
structure (i.e., a wall trench structure) is found at the site. “Field house,” a term
suggested by Finney 1993 for ephemeral, isolated Mississippian structures, might seem
more appropriate. The field houses identified by Finney (1993) were similar to
Mississippian residential structures. Some exhibited posts rather than wall trenches, but
they typically featured rectangular basins. In fact, Finney (1993) questions why the effort
would have been made to dig a basin for a temporary shelter for farmers. In the case of
Feature 9 at D. Hitchins, it better fits our expectations of a field house since it did not
feature a basin. It otherwise is similar to the field houses identified by Finney: its
associated features yielded few cultural remains and animal remains were essentially
non-existent.
However, we would reiterate that agricultural activities were not the only
activities to take place at D. Hitchins during the early Mississippian period. Nuts were
also gathered here, and the area was probably a good spot for hunting. It would appear
37
that some processing of crops, both cultivated and nuts, also took place here. In short, we
believe that Feature 9 was a temporary shelter for farmers, hunters, and gatherers. The
term “field house” still seems most appropriate, though, given that farming would have
been the most important of these activities in terms of its economic contribution.
The low density and low diversity of artifacts suggest that people were not living
at D. Hitchins. This would imply that the crops and possibly meat produced here were
consumed elsewhere. The closest Mississippian village, the Kruckeberg No. 1 site
(11MS341), is believed to date to the Moorehead phase (Woods and Holley 1991). The
Moorehead phase falls later in the Mississippian period, so early Mississippian produce
from D. Hitchins must have been sent somewhere else. It is likely that D. Hitchins was
also farmed during the Moorehead phase (Woods and Holley 1991), so its fields probably
produced food for the village at Kruckeberg at that time.
Other than Kruckeberg, there are no known Mississippian villages in the uplands
of Cahokia Creek (Woods and Holley 1991). The John Fox site (11MS1108), located
about 1200 m west-northwest of the D. Hitchins site, is believed to date to the
“Lohmann/Stirling time span” (Woods and Holley 1991:53), which might make it
contemporary with the early Mississippian component at D. Hitchins. However,
Yancey’s (2007) site catchment analysis shows that the soils at John Fox had even higher
agricultural productivity than soils at D. Hitchins. With no apparent access to running
water, Yancey suggests that John Fox was probably also a “temporary farming base”
rather than a habitation site (2007:82). That is, people at the John Fox site, like those at
D. Hitchins, were probably growing food for consumption elsewhere.
A Mississippian village was present at AE Harmon (11MS136), which was
situated on the bluff edge above Cahokia Creek some 20 km south of the D. Hitchins site
(see Holt et al. 2005). However, there’s no reason to think that people living at AE
Harmon would have needed to bring in food from elsewhere, given the size of the
settlement and productivity of local soils. A more likely destination for crops grown at
both D. Hitchins and John Fox would therefore be Cahokia (Marks 2006). Many
archaeologists have suggested that upland sites helped feed Cahokia. For example,
Pauketat (e.g., 2003) suggests that the need to feed Cahokia was probably one of many
factors that led to the development of farming villages he terms “the Richland Complex”
in the uplands west of Cahokia during the early Mississippian period.
Thus, we consider Features 3, 8, 9, 10, and 18 at D. Hitchins early Mississippian
rather than late Emergent Mississippian because we believe farmers, hunters, and
gatherers here were likely producing food for Cahokia. As is the case with the Richland
Complex, we hypothesize that this occurred around the time of the “Big Bang” that made
Cahokia the dominant center of the American Bottom and beyond. In this scenario, we
are left with several possible explanations for why the ceramics at D. Hitchins show both
Mississippian traits and “old-fashioned” Emergent Mississippian traits. Pauketat
suggests the reason may have been “resistance” to new trends emanating from Cahokia
by “Mississippian farmers [who] had agency” (2003:56). The problem with this
explanation is that it seems inconsistent with the idea that farmers built villages in the
uplands in part to feed Cahokia. If they were resistant to Cahokia, why feed Cahokia?
The alternative suggestion is usually that upland folks out in the hinterlands were a little
slow to pick up on the latest styles coming out of Cahokia. Comparing urban and rural
fashion trends today, that seems a more likely explanation. In the case of D. Hitchens,
38
however, the explanation may even be simpler: why take your newest dishes out to the
field house? People heading out to work some distance from their home village would be
more likely to take their old dishes along rather than their new ones.
We are not suggesting that rituals did not take place at D. Hitchins. It seems
likely that Mississippian farmers prayed for (and against) rain just as farmers do today.
However, the rituals suggested by the Klostermeier collection probably took place during
the Moorehead phase. Klostermeier and Woods (1986) found Moorehead phase ceramics
at both D. Hitchins site and 11MS341. As viewshed analysis by Yancey (2007)
demonstrates, villagers at 11MS341 would have been able to look down upon the D.
Hitchins site, as well as several other Mississippian sites in the area. Some of these sites
could have been simple hunting and farming locales, as it appears D. Hitchins was at the
beginning of the Mississippian period. D. Hitchins itself could have become a burial
ground during the Moorehead phase as Klostermeier has speculated, although we found
no evidence for it.
Finally, we are left wondering about the purpose of the big pile of FCR in the
central excavation block. While the rock cannot be dated, the high percentage of shelltempered pottery at the base of the plowzone in proximity to the FCR concentration, and
the Cahokia points and Mill Creek chert found in the plowzone, suggest that it might date
to the Mississippian period. If it dates to the early Mississippian period, we would
assume the FCR was used to dry crops. If it dates to the Moorehead phase occupation,
perhaps it was used in food processing, or perhaps it was used in a sweat lodge or other
ritual context. Either scenario must remain speculative given the poor context of the
feature. Still, a Mississippian age seems most likely. Archaic points were also found in
the plowzone in this area, but we can’t imagine why Archaic hunters would have
stockpiled such an enormous amount of FCR.
CONCLUSION
The SIUE field school achieved its primary goal of providing archaeological
training to students of anthropology. In addition, it provided the opportunity to record
new information about an area in the
uplands north of the American
Bottom that is poorly known to
professional archaeologists. The
data we uncovered provided research
opportunities for both students and
faculty (see Guntren 2008; Marks
2007; Yancey 2007). Thus, all of
our goals were achieved.
Today the project area has
been taken out of agricultural
production and planted with native
grasses, flowers, shrubs, and trees
(Photo 16). The sites in the project
area are protected to the extent that
they are no longer reduced by
39
plowing, and prairie grasses provide excellent erosion control. The roots of shrubs and
trees that have been planted could cause some damage if additional intact archaeological
features are present at D. Hitchins. However, and most importantly, the project area is
protected from development by the current landowners. While the sites it contains have
effectively become invisible beneath this prairie/savannah reconstruction, they are
arguably among the best protected sites in this part of the rapidly developing American
Bottom.
40
REFERENCES
Brown, J. A., R. A. Kerber, and H. D. Winters
1990 Trade and Evolution of Exchange Relations at the Beginning of the Mississippian
Period. In The Mississippian Emergence, edited by B. D. Smith, pp. 251-280.
Smithsonian Institution Press, Washington.
Drooker, P. B.
1992 Mississippian Village Textiles at Wickliffe. University of Alabama Press,
Tuscaloosa.
Finney, F. A.
1993 Spatially Isolated Structures in the Cahokia Locality: Short-Term Residences or
Special-Purpose Shelters? Illinois Archaeology 5:381-392.
Fortier, A. C.
1993 American Bottom House Types of the Archaic and Woodland Periods: An
Overview. Illinois Archaeology 5:260-275.
Fortier, A. C. (editor)
2007 The Archaeology of the East St. Louis Mound Center, Part II: The Northside
Excavations. Transportation Archaeological Research Reports, No. 22. Illinois
Transportation Archaeological Research Program, University of Illinois at
Urbana-Champaign.
Fortier, A. C., R. B. Lacampagne, and F. A. Finney
1984 The Fish Lake Site. American Bottom Archaeology FAI-270 Site Reports Vol. 8.
University of Illinois Press, Urbana.
Guntren, G.
2008 Using Experimental Archaeology to Reconstruct the Circular Post Structure from
the D. Hitchins Site. Unpublished senior project, Department of Anthropology,
Southern Illinois University Edwardsville.
Holt, J. Z., S. L. Moore, T. Evans, and C. Buskohl
2005 SIUE Field School Investigations at the A. E. Harmon Site (11MS1136).
Unpublished report on file at the IHPA and available for download at
www.siue.edu/ANTHROPOLOGY.
Jackson, D. K.
2007 A Reevaluation of Rosewood Phase and Initial Late Woodland Sites in the
American Bottom Area. Paper presented at the Midwest Archaeological
Conference, Oct. 4-6, University of Notre Dame, South Bend, Indiana.
41
Johannessen, S.
1984 Paleoethnobotany. In American Bottom Archaeology, edited by Charles
Bareis and James Porter, pp. 197-214. University of Illinois Press, Urbana.
Koldehoff, B.
2006 Chipped Stone Resources of Monroe County. In Late Woodland Frontiers:
Patrick Phase Settlement along the Kaskaskia Trail, Monroe, County, Illinois, by
B. Koldehoff and J. M. Galloy, pp. 367-376. Transportation Archaeological
Research Reports, No. 23. Illinois Transportation Archaeological Research
Program, University of Illinois at Urbana-Champaign.
Koldehoff, B. and J. M. Galloy
2006 Late Woodland Frontiers: Patrick Phase Settlement along the Kaskaskia Trail,
Monroe, County, Illinois. Transportation Archaeological Research Reports, No.
23. Illinois Transportation Archaeological Research Program, University of
Illinois at Urbana-Champaign.
Leeper, R. A.
2004 Soil Survey of Madison County, Illinois. USDA Soil Conservation Service
Marks, E.
2007 Who Fed Cahokia: An Analysis of the D. Hitchens Site. Unpublished senior
project, Department of Anthropology, Southern Illinois University Edwardsville.
McElrath, D. L. and A. C. Fortier
2000 The Early Late Woodland Occupation of the American Bottom. In Late
Woodland Societies, edited by T. E. Emerson, D. L. McElrath, and A. C. Fortier,
pp. 97-121. University of Nebraska Press, Lincoln.
Meinkoth, M. C., K. Hedman, and D. L. McElrath
2001 The Cunningham Site: An Early Late Woodland Occupation in the American
Bottom. Transportation Archaeological Research Reports, No. 9. Illinois
Transportation Archaeological Research Program, University of Illinois at
Urbana-Champaign.
Neely, R. D. and C. G. Heister
1987 The Natural Resources of Illinois. Special Publication 6, Illinois Natural History
Survey. Illinois Department of Energy and Natural Resources, Champaign,
Illinois.
Pauketat, T. R.
2003 Resettled Farmers and the Making of a Mississippian Polity. American Antiquity
68:39-66.
42
Pauketat, T. R. (editor)
2005 The Archaeology of the East St. Louis Mound Center, Part I: The Southside
Excavations. Transportation Archaeological Research Reports, No. 21. Illinois
Transportation Archaeological Research Program, University of Illinois at
Urbana-Champaign.
Ray, J. H.
2007 Ozarks Chipped-Stone Resources: A Guide to the Identification, Distribution, and
Prehistoric Use of Cherts and Other Siliceous Raw Materials. Missouri
Archaeological Society Special Publications No. 8. Missouri Archaeological
Society, Springfield.
Van Nest, J.
2002 The Good Earthworm: How Natural Processes Preserve Upland Archaic
Archaeological Sites of Western Illinois, U.S.A. Geoarchaeology: An
International Journal 17(1):53-90.
Woods, W. I.
1986 Prehistoric Settlement and Subsistence in the Upland Cahokia Creek Drainage.
Ph.D. dissertation, Department of Geography, University of WisconsinMilwaukee.
Woods, W. I. and G. R. Holley
1991 Upland Mississippian Settlement in the American Bottom Region. In Cahokia
and the Hinterlands: Middle Mississippian Cultures of the Midwest. University
of Illinois Press, Urbana.
Yancey, M.
2005 The D. Hitchens Site: A Survey and Analysis of a Multicomponent Site in the
Uplands Adjacent to the American Bottom. Unpublished senior project,
Department of Anthropology, Southern Illinois University Edwardsville.
2007 The D. Hitchens Site: An Interpretation of Mississippian Space. M.S. thesis,
Department of Geography, Southern Illinois University Edwardsville.
43
LIST OF TABLES
Table 1. Cultural materials recovered from shovel test pits. Weight (wt) is in grams.
STP
A1
A6
A6.5
A7
A8
A9
A9.5
B4
B4.5
B5
B7
B9
B9.5
B10
B11
B12
C1
C2
C2 – Stratum C
C2 – Stratum D
D2
D2 – Stratum C
D3
D3 – Stratum B
D4
D10
D16
E5
F8
F9
G9.5
Debitage
#
2
2
3
4
1
1
2
2
2
5
1
1
1
1
1
3
1
1
3
2
Debitage
wt
3.56
0.30
0.79
1.16
0.29
0.58
0.89
3.81
0.61
15.84
0.29
0.22
0.55
1.41
0.19
3.36
0.44
0.19
2.59
0.71
FCR
#
Historic
#
1
1
3
1
1
2
1
1
3
1
1
1
0.18
0.10
4.08
0.19
0.23
1.08
0.14
0.41
0.10
1
Note: Cultural materials were recovered from the plowzone unless otherwise indicated.
44
Table 2. Unit coordinates and sizes.
Unit
A
B
C
D
DD
E
F
G
H
I
J
K
KA
L
LA
LB
LC
M
N
NA
NB
NC
ND
NE
O
P
Q
R
SW Coordinate
N292 E302
N292 E304
N292 E306
N290 E302
N290 E301
N290 E304
N290 E306
N288 E302
N288 E304
N288 E306
N218 E300
N220 E308
N218 E308
N214 E312
N216 E312
N216 E314
N214 E314
N214 E298
N228 E305
N228 E307
N226 E307
N226 E305
N224 E305
N224 E307
N216 E310
N218 E310
N218 E312
N218 E314
45
Unit Size (m)
2x2
2x2
2x2
2x2
1x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
2x2
Table 3. Feature data. Volume is in liters; other measurements are in meters.
Feature Length (m) Width (m)
Depth (m)
Volume (l)
Plan
Profile
Function
Age
1
1
1.17
0.085
37.8
circular
inslanting
nut processing
early Late Woodland
2
1.42
1.38
0.075
55.6
circular
inslanting
nut processing
early Late Woodland
3
1.24
1.08
0.435
387.1
circular
belled/inslanting
storage
early Mississippian
8
1
0.85
0.23
79.7
circular
inslanting
food processing
early Mississippian
9
3.08
2.78
0
NA
circular
NA
shelter
early Mississippian
10
0.9
1.12
0.23
159.1
circular
vertical/inslanting food processing
early Mississippian
13
0.4
0.36
0.1
11.3
circular
vertical
post hole?
unknown
14
0.16
0.2
0.09
2.3
circular
vertical
post hole?
unknown
15
0.27
0.265
0.1
3.1
circular
inslanting
post hole?
unknown
16
0.27
0.33
0.145
6.4
circular
inslanting
post hole?
unknown
18
unknown
unknown
unknown
unknown
unknown unknown
food processing?
early Mississippian?
Note: Data obtained from feature summary forms. Volumes calculated using formulae from Fortier et al. 1984: Figure 22.
46
Table 4. Postmold data (obtained from profiles).
Postmold
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
average
Diameter (cm)
6
6
7
9
7
8
8
10
8
7
9
8
8
5
9
5
5
7
9
8
6
6
7
Depth (cm)
5
4
3
8
5
9
6
9
8
6
5
7
6
5
8
4
4
5
6
8
5
2
6
47
Table 5. Ceramic data, non-features. Weight (wt) is in grams.
UnitStratum
Grit/Unknown
Grit/ Plain
Grit/Plain
MCS
Grit/CM
Grit/S-twist
Grit/Z-twist
Shell/
Unknown
Shell/Plain
Shell/Red
Shell/Brown
ShGt/Plain
Grog/Plain
Unidentified
#
#
#
#
wt
#
wt
#
wt
#
#
wt
#
#
#
#
wt
#
wt
2
7.01
1
5.78
7
8.14
5
9.51
5
10.84
4
4.63
2
1.29
3
1.62
12
18.76
1
1.37
10
7.89
1
1.29
6
0.8
3
11.2
1
5.65
11
14.93
7
10.88
1
2.44
6
3.06
27
75.06
5
13.11
17
7. 08
1
1.05
2
6.59
2
0.49
2
0.49
5
7.37
2
0.36
3
4.23
3
4.38
2
0.56
wt
wt
wt
A-Ap
1
0.46
B-Ap
1
1.09
D-Ap
11
17.68
DD-Ap
4
2.59
G-Ap
3
0.98
H-Ap
5
4.3
I-Ap
7
7.71
J-Ap
106
76.34
2
2.85
K-Ap
153
142.78
4
7.8
KA-Ap
243
167.02
KA-A/B
39
20.99
L-Ap
120
117.23
7
18.2
LA-Ap
110
91.6
1
1.67
LB-Ap
38
25.29
2
1.8
LB-B
5
7.53
5
9.4
LC-Ap
295
148.37
24
31.41
LC-A/B
18
9.32
3
4.71
M-Ap
11
13.87
N-Ap
54
59.89
N-B
9
5.42
NA-Ap
128
140.22
NA-B
11
10.34
4
3
1
2
6
5.61
1.52
1.87
1
9
7.6
2.29
20.69
14.35
1
0.59
13
24.62
10
41.3
2
2.85
1
1.08
13
26.93
8
18.18
3
7.55
wt
3
5.48
37
14.64
1
1.28
3
19.1
9
5.67
13
51.56
18
8.54
4
7.32
48
wt
1
wt
21.24
wt
NB-Ap
130
107.66
22
NB-B
18
11.63
NC-Ap
81
69.74
16
ND-Ap
179
138.43
ND-B
18
14.23
NE-Ap
127
95.77
NE-B
28
14.26
O-Ap
74
82.65
P-Ap
277
200.56
3
P-B
5
2.13
Q-Ap
227
154.32
Q-B
7
R-Ap
19.91
10
22.31
7
15.74
3
9.91
1
1.75
2
21.91
39.07
2
5.14
2
4.29
7
17.02
6
10.34
11
21.02
4
6.68
3
8.29
1
0.79
16
33.58
2
1.38
8
10.86
5
4.47
1
11.29
5
15.81
2
0.68
2
0.84
2
2.63
18
71.29
7
8.09
4
4.44
2
7.19
5
5.22
3
2.34
1
0.37
4
3.45
5
20.52
2.95
23
33.81
3
10.92
20
8.14
16
10.54
1
3.4
2
1.02
6
5.79
5
10.05
2
11.79
32
16.85
3
7.1
11.35
1
0.66
1
3.33
52
63.96
5
7.87
1
8.47
R-B
3
11.85
3
12.62
Total
2598
2049.56
214
409.1
84
232.02
39
31.31
85
149.68
11
17.61
Brown = brown-slipped surface
CM = cordmarked, twist direction undetermined
MCS = Madison County shale
35
103.57
144
Red = red-slipped surface
ShGt = shell and grit temper
49
60.32
4
9
1
3
72
168.18
6
4.9
1
21.24
2
7.19
1
2.4
2
3.69
Table 6. Ceramic data, features. Weight (wt) is in grams.
Feature
2
3
“5”
“6”
8
9
10
“11”
14
15
16
Total
Grit/Unknown
#
wt
43
126.98
147
101.68
4
2.72
11
19.9
2
6.52
6
1.24
46
49.98
4
1.56
1
0.2
2
0.73
3
1.49
269
313
Grit /Plain
#
wt
1
4
3
8
3.55
18.85
4.4
26.8
Grit/CM
#
wt
29 278.91
19 31.46
18
3
65.25
2.49
7
9.53
76
387.64
Grit/S-twist
#
wt
5
156.23
4
17.39
4
33.84
12
46.02
2
1
2
0.83
1
29
1.54
257.85
Grit/Z-twist
#
wt
7
1
2
4
40.8
1.42
16
10.36
1
8.89
Shell/
Unknown
# wt
Shell/Plain
# wt
Shell/Red
# wt
Unidentified
# wt
5 14.18
1 3.83
8 77.43
1 0.99
2 1.39
1 0.18
2 6.89
1 0.09
15
77.47
3 1.57
CM = cordmarked, twist direction undetermined
Red = red-slipped surface
Note: Features in quotation marks were ultimately determined not to be cultural features.
50
8 24.9
8 77.43
2 1.08
Table 7. Local cherts. Weight (wt) is in grams.
Provenience
Feature 1
Feature 2
Feature 3
“Feature 5”
Feature 8
Feature 10
“Feature 11”
Feature 15
Unit A-Ap
Unit B-Ap
Unit C-Ap
Unit D-Ap
Unit DD-Ap
Unit E-Ap
Unit F-Ap
Unit G-Ap
Unit G-B2
Unit G-B7
Unit H-Ap
Unit I-Ap
Unit J-Ap
Unit K-Ap
Unit KA-Ap
Unit KA-A/B
Unit L-Ap
Unit LA-Ap
Burlington/Glacial
Debitage
#
wt
2
0.47
3
6.17
91
50.37
5
3.97
11
4.76
28
10.01
2
1.42
1
0.16
75
213
72
59.77
129
54.5
99
54.79
30
12.64
90
47.4
112
74.06
78
29.8
2
0.55
2
0.82
68
49.19
82
52.37
171
81.23
253
199.63
294
164.44
29
12.17
220
138
227
164.71
Burlington
Core
#
wt
Burlington
Flake Tool
#
wt
Burlington
Reworked
Flake
# wt
Burlington
Hoe Flake
# wt
Burlington
Scraper
# wt
1
1
2
3
4.52
2
4.26
21.86
20.49
1
Burlington
Biface
# wt
Burlington
Point
Fragment
# wt
Unknown
Chert
Biface
# wt
3.73
0.92
3
1
1
1
19.3
12.58
25.18
Unknown
Chert
Point
Fragment
# wt
1
2
1.96
18.02
1
2
12.55
1
9.76
6.73
1
51
1.03
3.03
Unit LB-Ap
Unit LB-B
Unit LC-Ap
Unit LC-A/B
Unit M-Ap
Unit N-Ap
Unit N-B
Unit NA-Ap
Unit NA-B
Unit NB-Ap
Unit NB-B
Unit NC-Ap
Unit ND-Ap
Unit ND-B
Unit NE-Ap
Unit NE-B
Unit O-Ap
Unit O-B
Unit P-Ap
Unit P-B
Unit Q-Ap
Unit Q-B
Unit R-Ap
Unit R-B
Total
124
8
158
6
97
337
17
314
31
256
28
207
230
14
236
40
191
10
258
6
116
2
145
6
5013
86.99
4.07
72.5
0.74
84.81
175.62
16.95
247.55
19.15
115.76
19.72
143.99
304.57
21.5
109.6
34.94
135.08
10.87
166
2.17
133.46
5.27
88.89
2.62
3489.22
1
1
9.34
1
3.78
2
1
17.14
14.49
2
35.66
2
64.91
2
10.61
1
18.72
1
1
5.79
1.92
2
32
1
27.74
1
18
23.01
320.5
1
1
11
63.97
1
0.82
2
3.52
5
12.1
4.53
1.14
1
4.09
1
1.39
8
14.91
15.4
1
3.1
1
1.07
3
5.09
Note: Features in quotation marks were ultimately determined not to be cultural features.
52
1
3.73
2
19.44
3
9.76
1
3.03
Table 8. Extra-local cherts. Weight (wt) is in grams.
Unit-Stratum
Unit B-Ap
Unit DD-Ap
Unit J-Ap
Unit L-Ap
Unit LA-Ap
Unit LB-Ap
Unit LC-Ap
Unit M-Ap
Unit N-Ap
Unit NA-Ap
Unit NC-Ap
Unit ND-Ap
Unit NE-Ap
Unit O-Ap
Unit P-Ap
Unit Q-Ap
Unit Q-B
Unit R-Ap
Total
Salem
#
wt
Salem core
#
wt
1
42.08
Mill Creek
#
wt
1
4
1
1.17
13.08
2.71
2
2
20.56
8.92
1
1.99
9
47.26
Mill Creek
Hoe Flake
#
wt
Cobden
Flake
#
wt
1
1
2
0.3
0.81
1
1
2
1
3
1
2
Mill Creek
Flake Tool
#
wt
1
1
2
26.37
68.45
10
2.59
0.3
0.98
0.35
5
3.06
1.95
2
5.66
1
0.2
1
12.53
2
2
9.47
9.47
1
1
1
0.32
0.68
3
0.91
1
2
Choteau
#
wt
Ste.
Genevieve
#
wt
1
0.26
1
0.71
1
1
0.12
0.42
4
1.51
Kaolin
Hoe Flake
#
wt
1
0.54
Fern Glen
#
wt
0.5
2.05
4.95
1
1
0.34
0.37
10
15.79
12
4.1
1
1
1
0.24
0.83
1
1.01
3
1
3.55
0.44
4
4.13
4
3.99
0.36
0.62
1
53
Cobden
Point
Fragment
#
wt
2
0.31
0.85
1
1.18
1
1.18
Table 9. Projectile points from 11MS1124. Weight is in grams; other measurements are
in millimeters.
UnitStratum
SUR 2
SUR 3
G-Ap
G-Ap
K-Ap
KA-Ap
L-Ap
LB-Ap
NA-Ap
NC-Ap
NC-Ap
NC-Ap
ND-B
O-Ap
Q-Ap
Time Period
Mid-Late
Woodland
(Steuben)
Mississippian
(Cahokia)
Terminal
Archaic
(Prairie Lake)
Terminal
Archaic?
Late Woodland
(arrow)
Mississippian
(Cahokia)
Terminal
Archaic?
(Prairie Lake?)
Late Woodland
(arrow)
Mississippian
(Cahokia)
Mississippian
(Cahokia)
Late Archaic
(Etly)
Mississippian
(Cahokia)
Archaic
early Middle
Archaic
(Jakie stem?)
Middle-Late
Archaic
(Godar)
Chert Type
Cobden
Weight
4.96
Maximum
Width
17.96
Burlington
1.11
broken
29.43
3.85
Burlington
8.15
28.70
41.91
8.09
unidentified
3.88
broken
broken
6.81
Burlington
1.35
12.79
28.05
4.30
Burlington
1.54
16.34
25.38
5.48
Burlington
6.39
29.21
broken
8.86
Burlington
4.21
broken
36.11
7.12
Burlington
0.63
broken
broken
3.41
Burlington
0.30
12.12
17.75
1.78
Burlington
28.55
40.19
broken
11.20
Burlington
3.46
broken
broken
6.09
Burlington?
5.29
25.92
8.80
Burlington
10.32
22.72
reworked
into a
scraper
61.42
7.25
Burlington
2.57
broken
broken
7.25
SUR 2 = Surface near Unit A
SUR 3 = Surface near Unit N
54
Maximum
Length
39.54
Maximum
Depth
6.45
Table 10. Cobble tools. Weight (wt) is in grams.
Provenience
Feature 9
Unit J-Ap
Unit K-Ap
Unit KA-Ap
Unit L-Ap
Unit LA-Ap
Unit LB-Ap
Unit N-Ap
Unit NE-Ap
Unit R-Ap
Total
Mano
#
wt
Anvil
#
wt
1
Possible
grinding tool
#
wt
Sandstone
abrader
#
wt
Hammerstone
#
wt
328
1
1
1
132
1060
516
1
1
Cobble tool
#
wt
1
126
516
3
1206
1
444
4
1534
2
1504
1
472
472
1
132
2
1272
3
1398
Table 11. FCR and other lithic artifacts. Weight (wt) is in grams.
Provenience
Feature 1
Feature 2
Feature 3
"Feature 5"
"Feature 6"
Feature 8
Feature 10
"Feature 11"
Feature 15
Feature 16
Unit A-Ap
Unit B-Ap
Unit C-Ap
Unit D-Ap
Unit DD-Ap
Unit E-Ap
Unit F-Ap
Unit G-Ap
Unit G-B2
Unit G-B3
FCR
#
1
14
248
29
97
29
74
12
3
2
132
144
587
702
28
90
195
71
22
1
wt
22
200
1300
1706
2146
632
558
136
2
2
802
1036
1578
4444
576
1414
1330
736
222
6
Burned
Limestone
# wt
Chert/FCR
#
wt
1
1
13
65
58
68
3
12
10
12
13
18
92
49
15
4
133
73
5
2
113
93
208
19
22
172
111
2
1
55
34
Limonite
# wt
Hematite
#
wt
Unit G-B4
Unit H-Ap
Unit I-Ap
Unit J-Ap
Unit K-Ap
Unit K-B
Unit KA-Ap
Unit KA-A/B
Unit L-Ap
Unit LA-Ap
Unit LB-Ap
Unit LB-B
Unit LC-Ap
Unit LC-A/B
Unit M-Ap
Unit N-Ap
Unit N-Ap rock
Unit N-B
Unit NA-Ap
Unit NA-B
Unit NB-Ap
Unit NB-B
Unit NC-Ap
Unit ND-Ap
Unit ND-B
Unit NE-Ap
Unit NE-B
Unit O-Ap
Unit O-B
Unit P-Ap
Unit P-B
Unit Q-Ap
Unit Q-B
Unit R-Ap
Unit R-B
Total
4
191
122
322
440
2
365
137
568
313
162
28
1366
18
343
629
49
36
512
42
616
30
422
323
45
743
87
284
76
1662
1276
3638
5438
5
5924
660
4980
4266
2234
128
3450
260
2614
10764
10330
390
6242
572
7688
638
2865
4786
294
5072
544
4304
80
11
928
10
232
9
11950
2912
104
4558
152
2568
84
120326
62
133
70
216
178
318
169
217
19
166
176
129
25
228
360
307
190
3
67
106
11
218
3
222
248
49
2612
442
34
141
28
224
3
259
6
57
1
168
2
280
2
62
3
318
300
59
371
83
518
18
412
3
158
1
254
1
559
3
150
3
6159
6651
1
1
1
4
0.63
1
4.57
1
0.04
2
4.61
2.9
1
0.92
3
6
4.45
8.9
24
58
Note: "Unit N-Ap rock" consists of rock from the FCR cluster at the base of the plowzone in Unit N. Rock
from "Feature 5" was also in association with this FCR cluster. Features in quotation marks were
ultimately determined not to be cultural features.
56
Table 12. Paleoethnobotanical raw data. Counts of >2 mm specimens; all samples were 10 liters.
Feature
Sample location
Charcoal wt.
Charcoal density (g/10 l)
Raw ct. or est., >2mm fraction
Count per 10 l
Hickory nuts (Carya spp.)
Juglandaceae (hickory/walnut)
Black walnuts (Juglans nigra)
Total nuts
Wood
Bark
Grass stems
Corn kernels
Corn cob sections
Corn cupules
Corn glumes
Corn embryos
Cucurbitaceae rind
(squash/gourd rind)
Galls
Other seeds >2mm
Unknown
Nutshell allocations:
Juglandaceae to hickory
Juglandaceae to black walnut
1
1
2
3
3
8
8
Strat A top
6.873
6.873
224
224
56
42
P
98
119
7
Strat B top
0.386
0.386
16
16
4
5
P
9
6
top
0.837
0.837
27
27
0
7
1
8
19
top
1.101
1.101
38
38
1
4
12
17
11
2
bottom
1.479
1.479
62
62
2
9
16
27
19
1
4
9
top
1.684
1.684
79
79
2
4
24
30
34
2
bottom
1.506
1.506
46
46
7
6
8
21
19
1
10
4
1
1
1
2
1
6
1
10
Strat A
top
1.488
1.488
101
101
10
7
13
30
24
2
35
10
P
10
“Strat B”
8.556
8.556
534
534
10
19
19
48
63
3
2
240
1
101
44
1
10
Strat A
bottom
3.183
3.183
188
188
8
9
9
26
10
42
1
5
7
0.308
3.692
P
1
8
0.308
3.692
P = presence only among charcoal <2mm.
57
2.8
3.2
3.043
3.957
category %
%ubiquity
52.63
(n=10)
35
5
1
27.093
27.093
1315
1315
100
112
102
314
324
16
8
404
1
150
52
2
23.88
24.64
1.22
0.61
30.72
0.08
11.41
3.95
0.15
90
100
100
100
100
60
30
70
10
70
60
20
3
20
8
2
9
0
3
22
19
0.00
0.23
1.67
1.44
10
10
20
50
6.552
12.45
4.235
4.765
65.246
46.754
P
100
P
1
Totals
nut taxon %
(after
allocations)
47.37
Table 13. Carbonized seeds. Counts of specimens >0.5-mm; all samples were 10 liters.
Feature
Sample location
Charcoal wt. (>0.5mm)
All Seeds 2 to 0.5 mm (raw counts)
Chenopodium sp. (chenopod)
cf Elymus sp. (wild rye grass)
Euphorbia maculata (spurge)
Helianthus annuus (sunflower)
Helianthus/Iva (sunflower or marsh elder)
Hordeum pusillum (little barley)
Iva annua (marsh elder)
Panicum sp. (panic grass)
Phalaris caroliniana (maygrass)
Poaceae (grass family cf Panicum?)
Polygonum erectum (erect knotweed)
Polygonum or Chenopodium (endosperm frag.)
Scirpus? (sedge)
Solanum ptycanthum (nightshade)
unidentifiable fragments
Total seed taxa identifications
Seed count per gram charcoal >0.5mm/10 liters
1
1
2
3
3
8
8
Strat A top
6.873
Strat B top
0.386
top
0.837
top
1.101
bottom
1.479
top
1.68
bottom
1.506
0
0
0
2
20
3
1
7
10
Strat A
top
1.488
10
“Strat B”
8.556
22
4
171
20
3
2
3
10
Strat A
bottom
3.183
62
5
1
1
2
3
2
11
0
0
0
0
0
2
0
0
58
4
16
10.818
5
1
0
0
7
4.648
1
2
4
2
5
4
18
12.097
5
3
9
5
94
20
16
16
155
18.116
1
2
24
38
11.938
Totals
% taxon
ids
%ubiquity
(n=10)
13.68
0.43
1.28
1.28
1.28
0.85
1.28
2.56
6.84
2.14
57.26
0.85
0.43
9.83
40.0
10.0
10.0
20.0
10.0
10.0
10.0
20.0
40.0
10.0
50.0
10.0
10.0
30.0
27.093
285
32
1
3
3
3
2
3
6
16
5
134
2
1
23
51
234
5.762
Table 14. Wood charcoal subsampling. All samples were 10 liters.
Feature
1
Sample location
Strat A top
Wood >2 mm, subsample to id
22
Acer sp. (maple/boxelder)
Carya sp. (hickory)
Juglans spp. (blk walnut/butternut)
1
Juglans sp./Carya sp.(walnut or hickory)
Platanus occidentalis (sycamore)
Prunus sp. (cf. wild black cherry)
Quercus (general oaks)
Quercus, red oaks group
11
Quercus, white oaks group
5
Ulmaceae (elm/hackberry family)
3
Ulmus americana (Am. Elm)
Vitis sp. (grape vine)
ring porous (type indeterminate)
1
diffuse porous (type indeterminate)
1
unidentifiable type wood
Wood taxa ids.
20
1
Strat B top
6
2
top
20
3
top
11
3
bottom
19
8
top
24
8
bottom
17
10
Strat A top
22
1
10
“Strat B”
23
10
Strat A bottom
10
2
3
3
4
1
3
7
2
1
4
3
6
1
2
8
3
10
7
10
1
4
2
3
2
3
11
3
5
1
14
3
2
1
7
20
7
1
13
2
4
2
7
11
1
8
1
20
59
12
Totals
174
1
3
9
7
2
6
4
23
5
39
8
13
21
21
12
120
% taxon
ids
%ubiquity
(n=10)
0.8
2.5
7.5
5.8
1.7
5.0
3.3
19.2
4.2
32.5
6.7
10.8
10.0
10.0
40.0
10.0
10.0
30.0
10.0
50.0
10.0
70.0
20.0
10.0
LIST OF FIGURES
Figure 1. USGS map showing project area and previously recorded sites.
60
Figure 2. 1814 GLO map. Hamel township.
61
Figure 3. Locations of magnetic and resistance anomalies at 11MS1124 recommended
for ground truthing. Map by Dr. Michael Hargrave of ERDC/CERL. Hargrave’s N120
W40 = SIUE’s N300 E300.
62
Figure 4. Surface collection results.
63
Figure 5. USGS map showing revised site boundaries and newly recorded sites.
64
Figure 6. Shovel test results, 11MS1125.
65
Figure 7. Profile facing south, Units G, H, and I.
66
Figure 8. Plan view, central excavation block.
67
Figure 9. Plan view, northern excavation block.
68
Figure 10. Profiles of Feature 1 (facing north) and Feature 2 (facing south).
69
Figure 11. Plan view, southern excavation block.
70
Figure 12. Profile of Feature 3 (facing north).
71
Figure 13. Profiles of Feature 8 and Feature 10 (both facing south).
72
Figure 14. Profiles of post molds in Feature 9. A = post mold (10YR4/2.5 - 10YR4/4); B
= subsoil (10YR5/6 mottled with 10YR5/4).
73
Figure 15. Profiles of Features 13, 14, 15, and 16.
74
LIST OF PHOTOGRAPHS
Photo 1. Geophysical survey at 11MS1124: Dr. Mike Hargrave with magnetometer in
the foreground; Miranda Yancey and Kyle Miller testing electrical resistance in the
background.
75
Photo 2. Surface survey, 11MS1127.
76
Photo 3. Shovel testing at 11MS1125.
77
Photo 4. Opening up Units A-I.
78
Photo 5. Excavation of southern block (Units L, LC, O, LA, LB, KA, P, Q, R, and K).
79
Photo 6. FCR concentration in Unit N, central excavation block.
80
Photo 7. Feature 2; pottery at surface in foreground.
81
Photo 8. Features 1 and 2, after excavation.
82
Photo 9. Feature 3, showing shell-tempered sherds in situ in the profile and on the floor.
83
Photo 10. Feature 9 and surrounding features, southern excavation block.
84
Photo 11. Profiles, Features 8 and 10.
85
Photo 12. Feature 18.
86
Photo 13. Early Late Woodland rim sherd, Feature 2.
87
Photo 14. Emergent Mississippian rim sherd, Feature 10.
88
NA-Ap
SUR-3
SUR-2
L-Ap
NC-Ap
KA-Ap
LB-Ap
K-Ap
NC-Ap
SUR-1
O-Ap
NC-Ap
G-Ap
Q-Ap
G-Ap
ND-B
Photo 15. Projectile points. SUR-1 was collected from 11MS1127; all other points were collected from
11MS1124.
89
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Related manuals

Download PDF

advertisement