The location of the Lawhorn site is in the St. Francis River valley in
an area that has had little archaeological investigation. Using the
subdivision of the Alluvial Valley (Phillips, et. al., 1951, Fig. 1) the
Lawhorn site comes within the Malden Plain, and lies between Crowley’s
Ridge and the Little River Lowland. The Cairo Lowland lies to the north
and east and is separated from the Malden Plain by the Morehouse
Lowland. All except the Morehouse Lowland were described, at least in
part, by Williams (1956). The Malden Plain area is less well known than
the other areas described by Williams. To the south of the Malden Plain
is the Lower St. Francis Basis which was discussed by Griffin in
Archaeology of Eastern United States (1952).
The site is relatively small and seemed to have no more than two
components represented. The materials that were obtained from the site
in the preliminary testing did appear to be somewhat unusual yet not
intrusive in the St. Francis valley area. One reason that the site was
picked for excavation is that it is relatively small in size making it
possible for more aspects of the site to be investigated within a
relatively short time. The excavation of the site was, in part, an
experiment to determine whether or not a group of serious amateur
archaeologists could produce information that would be of value to the
field of archaeology and professional and amateur archaeologists alike
on the necessary basis of working on weekends and during vacation time.
There were no funds for the excavation other than those provided by the
group of men interested in carrying out this experiment, and, as is
usual in such endeavors, though many worked on the project, a small
nucleus of four did most of the work. There were 8 to 10 persons who
worked as much as a year on the project. In spite of the changing crews
and the difficulties under which the different groups worked, a rather
full sample of the entire site was obtained. The total time spent
covered a period of four years in the field and two years of laboratory
work. The work from start to finish was under the direction of one
individual, the writer, and thus there was always an organization and a
continuity to it.
The methodology followed in the excavation is that of accepted hand
methods of archaeology (Fig. 46) controlled by making a map of the site
and excavating within a grid system and with depths provided by a
farmer’s level transit. All notes and excavation procedures were checked
with professional archaeologists and when some new problem arose a
professional archaeologist was contacted for advise or consultation
before the excavations progressed further. Since time was always at a
premium this consultation was many times by telephone and quite often
several long distance phone calls were necessary before some phase of
the work could be carried to a conclusion. Furthermore, the site was
ninety miles from the base station of those working on the project which
meant a round tip of 180 miles on each weekend or other excursion to the
site.
[Illustration: Figure 46. Cutting a Horizontal Profile in Square
27R32, Showing Use of Hand Tools]
The analysis and interpretation of the materials excavated pointed up
the necessity of obtaining many notes and saving all material from the
site. Part of the analysis was done prior to the finish of the
excavation and this was helpful in gaining fuller information from some
of the later parts of the excavation. Most of the analysis was done at
the Chucalissa State Park Museum, Memphis, Tennessee, through the
courtesy of the museum director, Charles Nash.
The site covered approximately four acres and the portions excavated
included the village living area with house sites, a plaza area, and
cemetery area. What was considered to be a fair sample was obtained from
all of these areas. Incidental to this sample was the sampling of the
earlier component since all excavations were carried to sterile subsoil.
In the analysis most of the time was spent on the pottery since it
seemed probable that the pottery was the most diagnostic in determining
the cultural position and the time of the site. It was broadly apparent
from the beginning of analysis that the pottery represented two distinct
traditions. The greatest part of it was the standard shell tempered
pottery of the area, Neeley’s Ferry Plain with a small percentage of
decorated types on the Neeley’s Ferry Plain base. The secondary type was
a sand tempered ware that was plain, cord marked and textile marked. The
sand tempered ware was representative of the early component on the site
and was of a type that has not been named for this particular part of
the lower Mississippi valley.
The sand tempered pottery had some variety and was divided upon the
basis of the size of the cord markings as well as the textile markings.
Only a small percentage was plain. Seemingly associated with the sand
tempered pottery was a lithic complex that has not been described
elsewhere. The prominent projectile point or dart point type that
appeared to be associated was less than 2½ inches in length and was
relatively thick. It had little or no shoulders and was approximately
twice as long as wide. There were no diagnostic points that were
associated definitely with this component but it is possible that the
one Motley point found on the site gives an indication of the time
period of the component. The Motley points supposedly date between 1300
and 200 B.C. A Uvalde point was also found on the site but there is no
assurance that this was not an accidental inclusion in the site. Four
Gary points were also found but it is just as possible that these may
have been associated with the later component on the site.
There were no features that could be definitely related to the early
component and it was necessary to separate the early component primarily
upon typology and secondarily upon superposition. One reason for this
was the fact that the site itself is on a very sandy bench of land and
it seems probable that the evidence of the early occupation was shifted
and many were exposed at the time the later occupation took place
causing a certain amount of mixing in the deposits. Furthermore,
earthquake activity had disturbed the deposits (Fig. 47). The old
surface that the site lies on is the A1 surface of the Ohio according to
Fisk (1944 Plate 15 sheet 1).
[Illustration: Figure 47. Vertical and Horizontal Profiles Showing
the Intrusion of Sand into Cracks in the Soil, Judged to be caused
by Earthquake Activity]
The sand tempered pottery is very similar to the Barnes series that has
been named in the Cairo Lowland area (Williams 1956). The ware also fits
the description of the Thomas Plain, Blue Lakes cord marked and Twin
Lakes fabric impressed (Phillips, et. al., 1951). It was decided not to
utilize the names of the types from either area since the descriptions
were not sufficient for it to be certain that the cord marked series
here could be equated with either the lower alluvial valley types or the
Cairo Lowland area. It does seem relatively certain, however, that the
early component on the site is representative of a period that may have
preceded the Baytown or Middle Woodland period or which may have been
contemporaneous with it. It is of interest that no Baytown Pottery
occurred on the site. The fact that Baytown pottery occurs in some
profusion on sites in the surrounding area might indicate that the early
component on the Lawhorn site is earlier than Baytown and that the site
was never utilized during the Baytown or Middle Woodland period.
It is possible that the sand tempered wares represented in the early
component on the site are a middle Woodland type and the evidence for
this is the fact that the Cairo Lowland area to the north, Williams
(1956, in his table 2, page 32) places Barnes Ridge where the Barnes
Series of Sand tempered types occur after Burkett and Hoecake and prior
to Black Bayou. This places it in a Middle Woodland position.
Furthermore on page 29 in Williams Table I he places the Barnes Ridge
component at LaPlant after the Hoecake. The Hoecake equates with
Woodland. Barnes Ridge, in his Table 3 page 38, is noted as Middle
Baytown. This would equate with Middle Woodland in the central
Mississippi valley.
If the situation here in the Malden Plain area is similar to that in the
Cairo Lowland and the sequence that Williams has suggested is correct,
this would explain why no clay tempered Baytown material occurred on the
site for this would be a component of the Middle Baytown period and
would be preceded and perhaps followed by the clay tempered Baytown
series.
The late component on the site, or the Mississippian component, is not
easily placed in either its cultural position or the time period in
which it existed. The pottery is rather distinctive and is wholly a
Neeley’s Ferry Plain and a few associated decorated types. The Neeley’s
Ferry Plain vessel shapes consist of water bottles of two types, the
long necked and the short necked and an unusual amphora water bottle;
plates with undifferentiated rims and undecorated with the exception of
red filming; shallow bowls; moderately deep, straight sided bowls; and
deep bowls both straight sided and those with an incurvate side similar
to some of the Caddo wares; and a usual jar form with strap handles and
relatively low rim. A few of the jars are decorated and the most
prominent decorated type is Wallace Incised or something very similar.
Another minor decorated type is Matthews Incised and there are sherds of
incised decoration which were not classified. These may possibly be
Barton or Kent Incised but the number of sherds having the decoration on
them and their small size was such that it was not considered desirable
to try to classify them. Painted wares were all under the general term
Avenue Painted and consisted of Old Town Red, Carson Red on Buff and at
least one sherd of Nodena Red and White.
The water bottles had a sub-globular shape in most instances and had a
fillet around the base of the neck. They either had a flat base or an
annular ring base that was perforated. The shape in general compared
favorably with other vessels from the general St. Francis River area and
to some extent with water bottles from the Caddoan area to the
southwest. The steep sided bowls and the one incurvate sided bowl are
similar to some of the Caddo shapes but these shapes in the Caddo area
are associated with engraved wares and are upon much different paste.
The appendages on the vessels, both effigies and lugs and handles are of
aid in determining the relationship of the site to other areas. The
handles were perhaps the best guide since they seem to be copies of the
typical handles at Crosno, at Kinkaid and in the general lower Ohio
River valley. The handles have nodes at the top and in some instances
have a groove down the center or have two raised ridges on either side
of the strap handle. The strap handle is the most prominent on the site
although a very few loop handles do occur.
Judging from the pottery, the site is typically St. Francis River
Mississippian of the late B period (Phillips, et. al., 1951), and it has
some slight relationship to the Cairo Lowland area to the north and to
the Lower White River Basin to the south. Both of these relationships
might be expected due to the physical or geographical location of the
site between the two areas.
The other more diagnostic artifacts that were used in determining the
relationship of the Lawhorn site to surrounding areas were the
projectile points. The typical projectile point type is a small corner
notched arrowhead which has been named the Scallorn point in Oklahoma
and Texas and is supposed to date between 700 and 1500 A.D. It is also
very similar to the Table Rock Corner Notched arrowhead in southwestern
Missouri on the White River which is associated with a Shell tempered
pottery complex—the latest occupation in the area. Furthermore, the
arrowhead appears to be rather similar to the corner notched arrowheads
associated with the Matthews site and with the Mississippi cultures in
the Cairo Lowland area. Other important arrowhead types that might aid
in the association of the site elsewhere is the Bonham point which
supposedly dates between 800 and 1200 A.D. and which is found in the
Caddo area to the southwest. This point occurs only about a third as
frequently as the Scallorn point which seems to be similar to the
relationship expressed by the pottery from the site. Also occurring are
triangular points that are similar to the rough, ovoid triangular points
found at the Campbell site in the Little River Lowland of the Memphis
cultural sub-area, and to the triangular points that are common on
Mississippi sites throughout the central Mississippi Valley. One of
these triangular points, the Fresno, is supposed to date from 800 or 900
to 1600 A.D. Another, the Maude point, is noted to date from 1200 to
1500 A.D. These dates compare favorably with those for the ovoid points,
the Young, which date from 1200 to 1500 A.D. and the Catan, which dates
from 500 to 1800 A.D. Another triangular point that may be important is
one that can be classified readily as a Fort Ancient point and dates
from 1200 to 1600 A.D. All of these points tend to run through the
period 1200 to 1500 A.D., which is the general period that was
considered probable as the date of the site by analysis and
interpretation of the pottery. Therefore, the projectile point types
from the Late component appeared to verify the dating of the site in the
latter part of the B period as it has been established in the lower
alluvial valley by Phillips, Ford and Griffin (1951).
There are other items that are associated with the complex that might
aid in establishing its position and its relationship elsewhere. The one
pipe hints at relationships toward the Caddo area for it is similar to
those found at Spiro. The thumbnail scraper is similar to the ones found
on the Campbell site and other late sites in the Mississippi Valley, but
the small number of these suggest an earlier time period. These
miscellaneous items strengthen, rather than weaken the conclusions drawn
from the pottery and projectile points.
It is premature to suggest the sequence in this Malden Plain area of the
St. Francis River basin, but it might be helpful to present the sequence
here that will show the position of the Lawhorn components to the
probable sequence of the general area. The Southeast Missouri Area
Chronology Chart of Williams (1956, p. 38, Table 3) has been used, and
the probable position of the Lawhorn components are indicated and
underlined. The sequence is Williams (1956, Table 3) with the addition
of the underlined items.