A north-south cross sectional profile along the R22 line shows a rise of
3.4 feet from the south base line to a point of maximum elevation some
400 feet north. The midden deposit shows a corresponding increase in
depth. The zero base line was on the south end of the site in an area
that appears to have been destroyed by an early St. Francis River
meander. Excavation was carried on from this point to square 40R22, a
distance of 400 feet north, but not to the northern end of the site
which is some distance beyond. The northern portion of the profile
indicated that this was an area of major house building activity while
to the south and just north of the area washed out by the St. Francis
River meander there is some indication of a court or open community
center. The ancient river meander was filled with a bluish sandy clay.
This ends near stake 9R22 with a datum elevation of 1.2 feet above the
base line.
From 5 to 8R22 there was a sandy deposit below the plow line which
probably represents a deposit from standing water as the meander
activity comes to a stop and filling in became a slow silting process.
From 8 to 17R22 there was a thin hard-packed deposit from .2 to .3 feet
in depth. Test pits through this section yielded almost no
archaeological material while just north of 17R22 the midden deposit
abruptly deepened. This area, almost 100 feet north-south, was also seen
on the east-west profile. It is this area that gives the impression of
being a plaza or community center (Fig. 2).
Starting just north of this open area there was a bank of clean sand and
the midden deposit below suddenly dipped downward until at 21R22 it was
quite pronounced with evidence of a sand boil. This disturbance was due
to the New Madrid earthquake and can be clearly seen in many areas of
this state. Here, the evidence for land subsidence, along with large
sand boils originating from considerable depth, is clear. This
earthquake evidence was no longer apparent north of square 29R22 and the
full undisturbed midden deposit of 1.5 foot depth gives evidence of the
Indian occupation. At square 40R22 the deposit was 1.8 feet deep and it
is at this point that houses 2 and 3 were found. The excavation was not
carried any farther north.
It is impossible to see absolute stratigraphic separation of
archaeological materials in an inspection of cross sectional profiles.
There was a general feeling among the excavators that the sand tempered
sherds were more numerous in the lower levels but no clear association
can be made. In several instances sand tempered sherds were found in the
underlying subsoil but never were any shell tempered sherds so found.
It is to be noted that the firebasins and houses built on subsoil had
only shell tempered pottery associations so it must be concluded that
the people responsible for this pottery lived here at a time when the
midden was non-existent. It is tempting to argue that evidence of an
earlier occupation by people making the sand tempered pottery and
perhaps dart points was washed away in some series of floods sweeping
the camp-site clean except for a few minor items left behind in the
newly silted sands. The earthquakes of recent times have played their
part in reshifting the materials in this deposit thereby completely
confusing such stratigraphic picture as may once have been present.
Two east-west cross sectional profiles are available for study, one at
the northern end of the site and the other toward the south end. The
northern portion of the deposit can be seen along the 32 line starting
at the levee at stake 32R2 with a deposit depth of 2.2 feet. From this
point to 32R10 the top of the subsoil was quite irregular and this
irregularity is apparent on the surface of the land as the midden
deposit follows the irregularities of the subsoil. It is probable that
this is the result of the New Madrid earthquakes.
It was in this area that the presence of sand tempered pottery in the
subsoil was first noted. A number of burials were encountered in this
section of the excavation. From 23R10 to 23R16 the drainage ditch has
removed all archaeological materials. At stake 32R19 the deposit was 1.6
feet in depth and continued to stake 32R25 where it was 2.2 feet deep.
This full depth of deposit continued to 32R28 where it started to taper
out until at stake 32R39, it was only .2 feet deep. While levee building
has destroyed the western edge of the site we know that it was at least
400 feet wide. This would indicate a town four or more acres in extent.
An east-west profile was also drawn along the R17 line from the levee at
stake 17R5 where the deposit is quite thin to 17R39 where it almost
tapers out.
At 17R8 the depth was 1.4 feet while immediately across the drainage
ditch the hard packed dark band becomes apparent with little or no
deposit either above or below it. This hard packed area runs eastward
for about 100 feet to 17R30. At 17R32 the deposit was again 2.2 feet in
depth but this thinned out rapidly at 17R32-39 as the eastern limits of
the village was reached.