2000 Florida Anthropological Society Inc.
The Florida Anthropological Society Inc. holds
source text of the Florida Anthropologist
considered the copyright holder for the text
all rights to the
and shall be
and images of
The Florida Anthropological Society has made this publication
available to the University of Florida, for purposes of
digitization and Internet distribution.
The Florida Anthropological Society reserves all rights to this
publication. All uses, excluding those made under "fair use"
provisions of U.S. Code, Title 17, Section 107 are restricted.
Contact the Florida Anthropological Society for additional
information and permissions.
0 1- 21
PUBLISHED BY THE
FLORIDA ANTHROPOLOGICAL SOCIETY, INC.
THE FLORIDA ANTHROPOLOGIST is published quarterly in March, June,
September, and December by the Florida Anthropological Society,Inc., do
Room 102, Florida State Museum, University of Florida, Gainesville, FL 32601.
Subscription is by membership in the Society for individuals or institutions in-
terested in the aims of the Society. Annual dues are $4.00; student members
$2.00. Requests for memberships and general inquiries should be addressed to
the secretary; subscriptions, dues, changes of address to the treasurer; orders
for back issues and manuscripts for publication to the editor; and newsletter
items to the president. Second class postage paid at Gainesville, Florida.
XXVI, No. 1
The Nalcrest Site, Lake Weohyakapka, Florida
Ripley P. Bullen and Laurence E. Beilman. .
Dalton Culture in Northeastern Arkansas
Dan F. Morse . . . .
Archaic Hafted Spokeshaves with Graver Spurs from the
Albert C. Goodyear . .. . .
. . 23
. . 39
OFFICERS OF THE SOCIETY
President George Magruder
Parkways Palms Apts., No. 235B
Indian Harbour Beach, FL 32937
1st Vice President John W. Griffin
46 St. George St., St. Augustine,FL 32084
2nd Vice President Benjamin I. Waller
3161 S.E. Ft. King Ave.,Ocala, FL 32670
Secretary Nan D. Magruder
Parkways Palms Apts., No. 235B
Indian Harbour Beach, FL 32937
Treasurer Donald L. Crusoe
National Park Service, P. O. Box 2416
Tallahassee, FL 32304
Editor-Resident Agent Ripley P. Bullen
102 Florida State Museum, University of
Florida, Gainesville, FL 32601
Three years: Dan D. Laxson
Two years: Yulee Lazarus
Ft. Walton Beach, Florida
One year: Wilma B. Williams
At large, for one year
Thomas H. Gouchnour
Richard D. Hagen
Wesley F. Coleman
THE NALCREST SITE, LAKE WEOHYAKAPKA, FLORIDA
Ripley P. Bullen and Laurence E. Beilman
During the past several years (1969-71) Beilman and to a lesser extent his friend,
Edward Wheeler--both of Nalcrest, Florida--have been collecting projectile points and
other chipped stone tools (Figs. 2-9) from the shallow waters of Lake Weohyakapka (Fig. 1,
Po-15) in south central Florida about 15 miles east of Lake Wales. Feb. 26, 1970, and
again in 1971, Beilman brought some of the specimens (Fig. 3, c) to the Florida State Mu-
seum for identification and arrangements were made for the Bullens to visit the site in
March 1971. Wheeler very kindly supplied water transportation for the ensuing investiga-
Also at this time Beilman' s collection was studied and photographed (Figs. 4-8) and
subsequently that of John Stegman (Fig. 9) was similarly included. Both of these are from
the main site, Po-15. Stegman also told us of finding a few similar specimens along the
eastern side of the lake (Fig. 1, x) but we were unable to locate this second site. Much of
the eastern shore of Lake Weohyakapka is low and swampy and where there is high land
there has been a lot of modification due to a real estate development and the building of a
pier. This area will not be referred to again.
In June 1970, the Bullens returned to Nalcrest where Ben I. Waller, an expert scuba
diver and amateur archaeologist from Ocala, Florida, joined the group. He very kindly in-
vestigated the bottom of the lake including the two deep "holes" indicated in Figure 1. The
bottom was covered with a thick deposit of fine grained sediments. Stone tools are not found
further out in the lake than 100 feet from shore or beyond a depth of water of 3-4 feet. In the
deeper holes Waller put his arm into the deposits up to his shoulder without feeling anything
hard. In no place was there any suggestion of fossil animal bones nor has Beilman in his ex-
tensive investigation of the shallower areas of the lake ever found any fossil nubbins.
Subsequently, in the winter of 1972-73, Beilman with the late John C. Hamilton, also
of Nalcrest, located another but smaller deposit of similar artifacts further south than the
main site (Fig. 1, Po-17). We have included illustrations of artifacts from Po-17 in Figures
10-12, in spite of the fact they are a little out of focus, to show the similarities in the two
collections. This, of course, substantially increases the quantities of very small tools which
are so typical of Beilman' s and Stegman' s collections. While scattered examples of such
tools are known for Florida, particularly in the Tampa Bay and Gainesville regions, the only
known heavy concentration comes from Lake Weohyakapka.
Many of the chipped tools from Nalcrest are extremely small, almost microlithic in
size. Some (Figs. 3, 09-19; 6, a-dd; 7, rows c-d; 9; 10, a-o) are so strange to Florida
archaeology as to indicate a more or less completely different culture--at least a different
lithic assemblage--than any previously recorded for the State. For this reason we have
comprehensively illustrated the specimens found at the Nalcrest site. A review of the liter-
ature indicated that many of the new forms are seldom illustrated and that comparative mate-
rial is extremely rare. Hence, for comparative purposes, when this paper was presented in
Florida Anthropologist, vol. 26, no. 1, March 1973.
a preliminary form at the 1971 annual meeting of the Eastern States Archeological Federa-
tion in Gainesville, Florida (Bullen and Beilman 1972), we asked Dr. Dan F. Morse of Ar-
kansas to present a paper on the then newly excavated Dalton period Brand site. The small
tool inventory from the Brand site is closer to that from Nalcrest than that of any other
site we have been able to locate.
Morse' s revision of his preliminary ESAF report follows this one. Also included in
this issue of the Florida Anthropologist is Goodyear' s revision of an article on "Hafted
Spokeshaves with Graver Spurs," which he originally tendered some time ago, both for
comparative purposes and because readers will be interested in his detailed treatment of
this tool which is surprisingly common at Nalcrest. Goodyear documents this typical Nal-
crest type tool for the Gulf coast region north of Tampa Bay. References in our text to
Morse or Goodyear refer to their articles in this issue of the Florida Anthropologist.
Lake Weohyakapka and Indian Sites
As located in Figure 1, there are four Indian sites on the west side of Lake Weohya-
kapka. Of these the shore sites, Po-14 and Po-16 are freshwater shell and dirt middens
producing Belle Glade Plain pottery (Fig. 2, a-c). Po-14 or the Nalcrest Midden is sit-
uated on a low natural levee of the lake. In places two feet thick, it parallels the shore for
perhaps a hundred yards. In the midden are found deer and turtle bones as well as small
snail and mussel shells. Po-16 or the Meyer' s site, of similar composition, is situated
on a high stabilized sand dune about 12 feet above the elevation of the lake and nearly 250
feet west of the present lake shore. The Meyers' site is small and seems to consist of
several heavy midden concentrations extending over about a half acre. Except for the pres-
ence of St. Johns as well as Belle Glade Plain sherds and its high elevation and small
size, it closely resembles the Nalcrest Midden site.
Locations of the two early sites, Po-15 and Po-17, are shown in Figure 1. These
are not, properly speaking, sites but are collecting areas. Specimens are found under
water in-narrow zones which parallel the present shore for a distance of about 1. 5 miles
for Po-15 and .4 miles for Po-17. They are found from the present shore outward a max-
imum of 100 feet or to a depth of 3 to 4 feet.
Tools appear to be in shifting tan sands but really are resting on an eroded, reddish,
indurated sand deposit of unknown thickness which supports both the specimens and the
shifting tan sands. In a few instances the tan sands have moved outward leaving specimens
behind. The consolidated red deposit occurs near the edge of the lake as soft "sandstone"
ledge remnants and, further west, under the present low sandy shore, as compact red
sand at about the same elevation.
The north, east and south shores of Lake Weohyakapka are low and south of Tiger
Creek and east of Weohyakapka Creek swampy. To the west of Po-15 is a narrow strip of
redeposited sand only about a foot or two above the lake surface (Fig. 3, b, foreground).
Immediately to the west is the eroded face ofstabilized dune sand (Fig. 3, b, background)
which in places reaches a height of 10 to 15 feet above the water of the lake. From north
to south along the side of Po-15, this dune increases in altitude and then decreases.
Shortly north of site Po-17, it gets high again but falls off south of site Po-17.
We dug tests in front of the main dune beside the mid-portion of Po-15 but only found
the compact red deposit mentioned above. Clearing the eroded face of the dune revealed no
dark zones, habitation layers, chips or other evidence of early man. We believe that these
BULLEN AND BEILMAN
high dunes used to extend substantially further to the east--where they or their lower
slopes were periodically used as camp locations by various prehistoric groups--and over
many years have, probably during periods of storms, been entirely eroded away and their
industrial contents spread along the present shore by the waters of Lake Weohyakapka.
It is quite likely that during the period of major occupation, possibly 7-10,000 years
ago, the water table was substantially lower than now and that Lake Weohyakapka was not
as yet in existence. The bottom of the lake is suggestive of a small filled-in lime sink.
If so there may have been a chert exposure present. Bellman' s collection does not contain
typical workshop debris but such could be present in the deeper parts of the lake, between
the 50- and 51-foot contours where it would be buried under several feet of silt. In any
case the physiography at Po-15 was quite different thousands of years ago. On the map
(Fig. 1) we have suggested possible brook confluences which, with a rise in the water table,
might produce Lake Weohyakapka as we know it today.
As shown in the illustrations (Figs. 2-9) there is a wide variation in chipped stone
artifacts from the Nalcrest site. From the projectile point typology it is evident that Paleo-
Indian, Dalton, Archaic, Florida Transitional, and Safety Harbor time periods (Bullen 1968)
are represented (Figs. 4-5, 8). An approximate quantification of these points is given in
Table 1. It is not our intention to describe these points as we feel the pictures adequately
reflect the situation.
PROJECTILE POINTS FROM NALCREST SITE, Po-15
Typology Collections: Beilman Stegman Totals
Clovis-like (fragments) 4 4
Suwannee 1 1
Santa Fe 1 1
Beaver Lake 2 2
Dalton- like 2 2 4
Bolen Plain 5 5
Bolen Beveled 15 3 18
Archaic Stemmed* 62 8 70
Culbreath 3 2 5
Citrus 4 4
Hernando 1 1
miscellaneous side notched 5 2 7
Pinellas 6 6
Totals 107 21 128
Our attention here is directed primarily to a consideration of the small scrapers,
spurred tools, and other tiny artifacts (Table 2) which are so common at Nalcrest and, cur-
rently, so rare elsewhere in Florida. However, we would point out the relatively high per-
centage of Bolen points at both sites Po-15 (Table 1) and Po-17 (Fig. 11). It is exceeded
only by that of Archaic stemmed points which have not in the past been noticeably associ-
ated with these microlithic tools. This correlation is supported by the apparent absence or
rarity of the Nalcrest small tool assemblage at excavated Florida sites or at Paleo-Indian
collecting locations in Florida rivers. To date a Dalton period site--one producing Bolen
Beveled points--has not been dug in Florida.
Nalcrest Small Tool Assemblage (Table 2)
As this assemblage is so poorly known, it seems desirable to examine its known dis-
tribution. To that end we will add a paragraph after the description of each type giving
available distributional data for the 14 types listed in Table 2 for this complex. These
paragraphs will include chronological clues.
NALCREST TOOL ASSEMBLAGES, SITES Po-15 And Po-17
Typology Po-15 Po-17 Totals
Collections Beilman Stegman Beilman
Small or tiny tools
Tiny stemmed points 4 4
Twisted drills(?) 2 4 6
Tiny side scrapers or drills 2 3 6 11
Spurred chips 4 6 10
Small spurred cores 7 7
Small cores 1 7 8
Small concave end scrapers 47 16 4 67
Long concave end scrapers 5 5
Small ovate scrapers 32 11 2 45
Spurred scrapersb 11 11
Small hafted scrapers 3 3
Medium sized trianguloid
engraving tools 10 10
Flake drills 8 8
Small Waller knives 2 2
Larger or normal sized tools
Blanks or broken knives 9 2 11
Medium sized amorphous
scrapers or scraper knives 15 14 4 33
Hafted scrapers 7 3 2 12
Archaic stemmed drills 6 3 2 11
Clear Fork gouge 1 1
aSometimes spurred at corner.
bSmall or normal (thumbnail) size.
Tiny stemmed points (Fig. 3, c7-8) These are extremely small stemmed points which du-
plicate in shape well known Archaic forms. They are about 2. 5 cm. in length and propor-
BULLPEN AND BEILMAN
tionally smaller in width so that the apparent reduction in size from Florida Archaic
forms is uniform in all directions.
These tiny points are rather rare. There is a tiny stemmed point in the Simpson col-
lection at the Florida State Museum and Bullen has seen one or two others. Going further
afield, Prufer (1963:14-15; Fig. 11, 57, 60) lists two small wide-stemmed points as having
"been found in consistent association at several Late Paleo-Indian sites in the Ohio area."
Gagliano (1964:50-51, Fig. 17, c-d; 1967; 70-71 illustrates two tiny stemmed points for
Avery Island, Louisiana, but they came from a "surface" collection. Unfortunately no de-
finitive data is available at Nalcrest or elsewhere bearing on the function or date of these
miniature stemmed points. The above clues imply but certainly do not demonstrate these
tiny stemmed points to be pre-Archaic in date. If so they do not represent reductions in
size from Archaic forms.
Twisted drills or scrapers (Fig. 10, a-d) These small, about 2 cm. long, neatly chipped
tools are more flat on lower than on the illustrated upper sides. Characteristically their
major axes are rotated so that they do not lie flat on a table. While reminiscent of twisted
drills reported by Douglas Jordan (MacDonald 1968:22 a) for the Bull Brook site in Mas-
sachusetts, these are a different tool. Their use is not evident but they could function as
delicate scraping tools. ,
"Twist drills" are illustrated by Byers (1954: Fig. 92, f) for the Bull Brook Paleo-
Indian site in Massachusetts. MacDonald (1968:81, Fig. 22, a; Pl. X, f-g) illustrates
some twist drills from the Paleo-Indian Debert site in Nova Scotia. He mentions fluting or
attempts at fluting for his examples. There is no suggestion of this trait on our specimens.
While the Nalcrest tools included in this grouping resemble Byers and MacDonald' s illus-
trations of twist drills closer than they do any other illustrations we can find in the litera-
ture, there is a possibility that our tools are basically different and represent specialized
scrapers as opposed to drills.
Tiny side scrapers or drills (Figs. 2, g; 3, clZ-13; 7, cl7-20, dll-13; 10, e, ee-gg) -
These tools resemble broken off ends of perforators. Morphologically there are two vari-
eties, one with a trianguloid cross section, the other trapezoidal. In the latter case, the
steeply chipped scraper-like sides slope inward from the wider flat bottom to the narrower
flat top. They vary in length from 1. 5 to 4 cm., and taper longitudinally. Usually damage
from use prevents definitive statements about their ends. Sometimes the larger end has
been neatly chipped to produce a thinned curving butt, presumably for hating. The smaller
end also may exhibit chipping scars but they appear to be the result of use. It is believed
these specimens represent drills or perforators which were hafted in handles such as
hollowed-out antler tines.
These unifacial tools are fairly common but not plentiful. Our specimens duplicate
several found at Debert (MacDonald 1968: P1. X), Bull Brook (Byers 1954: Fig. 92, e), the
Reagon site in Vermont (Ritchie 1953: Fig. 89, 65, 69), and Paleo-Indian sites in eastern
Pennsylvania (Witthoft 1952: Pl. 2, 16). They were also present at the Dalton period Brand
site in Arkansas (Morse below: Figs. 3, m; 4, u-v). They are also found in northcentral
Florida but we have made no specific survey of provenience in that area.
Spurred chips (Figs. 2, d-e; 7, bl0, clO-11, f4; 8, w10-11; 9, x-z). A fairly common tool is
an ordinary ovoid chip, fairly thin and 2-3 cm. in maximum dimension, on the edge of which
spurs or engraving points have been generated by minute chipping. More often than not,
multiple spurs are present.
Similar engraving or scoring tools are known for many sites. Limiting ourselves to
small spurred chips (small flakes), we find them present at Brand (Morse below, Fig. 4,
l-o; Goodyear 1971: Fig. 18) in some quantity. They are also found at Avery Island (Gag-
liano 1967: Fig. 20, L), Debert (MacDonald 1968: Pl. 18, b), Bull Brook (Byers 1954:
Fig. 92, c), Lindenmeier (Roberts 1936: Pl. 9), Shoop site in Pennsylvania (Witthoft 1952:
Pl. 4, 6-8), Reagen (Ritchie 1953: Fig. 89, 39) in Vermont, and the Lecroy Paleo-Indian
site on the Tennessee River (Lewis and Kneberg 1956: Pl. 6, 1-9). In northcentral Florida
a multi-spurred chip (FSM-102719) came from a depth of 36-42 inches at the Darby Springs
site which is noted for a dominance of Bolen Beveled points. Another is recorded for the
Dalton zone at the Stanfield-Worley shelter in Alabama (DeJarnette et al. 1962: Fig. 41, D).
Also in the Dalton zone at this shelter were several larger flakes, ca. 5 cm. in size, with
engraving spurs (DeJarnette et al. 1962: Fig. 14, a-b, e-i).
Small spurred cores (Figs. 2, n-o, r-s; 7, d4, 9) These steep sided cores, 1.5 to 2.5 cm.
across, come in two forms. One is more or less oval in shape with a flat bottom and "round-
ed" top. They are about 1. 5 cm. thick and have one "corner" of the base "pulled out" to form
an engraving spur. The other type is squarish in shape with steep sides which tend to be a
little concave. Spurs are found on one or more of the projecting corners. This variety tends
to be a little thinner (about 1 cm. ) and to have a flattish top. Both varieties exhibit multiple
step fractures, usually along sides opposite spurs. While core-like in morphology, it is
doubtful because of their small size (thinness) that they were ever used for the striking off
Apparently these spurred cores were intentionally so made to serve as combination
planing and engraving tools. It is possible they were also used as wedges but we do not
think the evidence supports that usage. We have been unable to locate in the literature any
small spurred cores unless one from the Shoop site (Witthoft 1952: Pl. 4, 13) qualifies.
Small cores (Figs. 2, p-q; 3, oZ-23; 8, w4; 9, cc) These core-like tools resemble small
spurred cores but exhibit no engraving spurs. They tend to be a little larger than spurred
cores and are more often "turtleback" in shape. Perhaps they are unfinished spurred scrap-
ers. In their present form they would function as very small scraping planes.
The small steep-sided Nalcrest cores (Fig. 2, p-q) without spurs is rarely illustrated.
It is certainly present at the Brand site (Morse below, Fig. 6, i-J; Goodyear 1971: Figs. 15,
1-m; 16, m). A somewhat similar tool is illustrated for the Quad site near Decatur, Ala-
bama (Cambron and Hulse 1960:20, 24) which also produced numbers of Big Sandy I (Dalton
period or Bolen) points. This site should belong in the same time period as the Brand site.
Gagliano assures us (Personal communication) that they are present in Louisiana. They
certainly resemble his core scrapers (Gagliano 1967: Fig. 18, i-l). DeJarnette et al.
(1962:72) may be referring to the same tool when they write "Core Scraper: A core with one
or more edges apparently used as a scraper, or intentionally modified on one or more edges
to a scraping bit. Our Small Cores are similar to but much smaller than the one they
illustrate (DeJarnette et al. 1962: Fig. 43).
Small Concave end scrapers (Figs. 3, cl5-20; 6, c-q; 7, c3; 9, a-h; 10: f-i) These are
unifacial tools which are trianguloid in shape and have a steeply chipped concave edge at the
wide end. Sides also have steeply chipped scraper-like edges while the pointed end tends to
be broken or dulled. Small concave end scrapers are usually 2-3 cm. in length and sub-
stantially longer than broad. Some, however, approach an equilateral triangle in shape.
Sometimes one corner of the concave edge has been chipped to form a spur while all edge
corners are sharp and could easily function as incising tools. Some of these corners clearly
BULLPEN AND BEILMAN
show wear. These tools could easily have been used to work wood or bone as Goodyear
(below) describes. The concave edge duplicates that of a spokeshave and the corners would
serve as incising or grooving tools. Undoubtedly they were hafted by means of the narrow-
er end in small bone, antler, or wooden handles.
Goodyear (below) describes these tools in detail and documents their presence in the
greater Tampa Bay area as well as at the Brand site. Five more are preserved in the
Simpson collection at the Florida State Museum. One (FSM-102560) from lamonia Creek
in Leon County has one corner of the bit end extended to form a spur. The second (FSM-
102617) from near Branford and the third (FSM-102605) from Little River in Suwannee
County are small, 2. 5-3. 2 cm., and approach an equilateral triangle in shape. Two others
(FSM-102289), one 2 and the other 3 cm. in length, came from Alachua Field near Gaines-
ville. This field is noted for the relatively large number of Bolen Plain and Bolen Beveled
points found there. Both the concave end scrapers and the Bolen points are heavily patina-
ted--both to the same degree. Similarly, small concave end scrapers are present in a
collection from the edge of a wet weather pond in Early County, Georgia (Singletary 1973).
Included in the Georgia collection are 12 Bolen type points (half are beveled), trianguloid
engraving tools, a small spurred core, and various uniface scrapers. All--both the Bolen
points, the concave end scrapers, and the other uniface tools--are heavily patinated and
have a chalky appearance like that typically found on Bolen points in Florida.
Both Morse (below) and Goodyear (below) document small concave end scrapers
at the Dalton period Brand site in Arkansas but in minimum quantities in contrast with
Nalcrest. It is not certain from the report whether or not they were present at Stan-
field-Worley in Alabama. However, DeJarnette et al. (1962: Fig. 40, E) illustrate a
broken tool that appears to be a broken Nalcrest concave end scraper.
Long concave end scrapers (Figs. 6, b; 10, 1-o) These duplicate small concave end
scrapers except for size. They are large, about 3.3 to 4.5 cm. in length.
We have not found examples of this longer variant in the available literature. For
Florida, one (FSM-102772) came from the 24- to 30-inch depth at Darby Springs. As
mentioned before this site also produced predominantly Bolen Beveled points.
Small ovate scrapers (Figs. 3, c24-27; 8, ul-4; 9, t-u, cc) These are ordinary small
scrapers, possibly scraper-knives, about 3-4 cm. in diameter. Some might be called
These tools are illustrated for Stanfield-Worley (DeJarnette et al. 1962: Fig. 42),
the Quad site (Cambron and Hulse 1960: Fig. 34), for northcentral Alabama (Haln 1969:
Figs. 43, w, y; 4: A-B, L) and the Brand site (Goodyear 1971: Fig. 15, 1-m). Larger
ones are more common,
Spurred scrapers (Figs. 2, f, 13) These are small thumbnail scrapers about 2 cm. in
size and 5 cm. thick with spurs or remnants of spurs at the corners of the wide scraping
edge. One illustrated specimen (Fig. 2, f) is aberrant in that it has a straight face and
four spurs. The more typical specimens (Fig. 13) are more ovate with a wide convex
steep scraping front edge, steep scraper-like sides that converge towards the rear which
tends to be flattish. Of the two illustrated examples, one (Fig. 13 b) has a sharp, extreme-
ly carefully chipped, scraper edge and sides which after chipping have been dulled and ex-
hibit step fractures. It was prepared for hafting but probably never used. The other (Fig.
13, a) exhibits many stepped fractures of its scraper edge while the sides show more ev-
idence of dulling than in the previous example. Presumably, this one was hafted and used
for some time before being discarded.
SFig. 13. Small, spurred, thumbnail scrapers.
a, with use wear (tiny stepped fractures on
convex face; b, with sharp unused edge.
Both have sides dulled for hafting.
a is 1.8 and b 2. 2 cm. maximum width.
Spurred snub-nosed or thumbnail scrapers are a Paleo-Indian trait found at Linden-
meier (Robert 1936: Pl. 8), Bull Brook (Byers 1954: Fig. 92, a), Delbert (MacDonald 1968:
Pl. 13, b), the Williamson site in Virginia (McCary 1951: Fig. 9, 7, 9, 12), and the Lecroy
site (Lewis and Kneberg 1956: Pl. V, 1-8). They are also present in fair quantities in the
more recent Brand (Morse below: Fig. 4, r) and Stanfield-Worley (DeJarnette et al. 1962:
Fig. 48, triangular end scraper) sites as well as at Nalcrest and at Bolen Point south of
Gainesville (Bullen and Wing 1968:94).
Small hafted scrapers (Fig. 3, c9-11) These rather rare specimens are about 2 cm. in
length and have side-notched stems suitable for hating in a manner similar to that of much
larger knives and projectile points. These are presumedly scraping tools but, it would seem,
they were for less careful work than that for which the other scrapers were made.
Small hafted scrapers--stemmed with hafting notches--currently are only reported
for Nalcrest. A short but much wider form is reported by Haln (1969: Fig. 7, N) for north-
central Alabama. Large forms, of course, are fairly common in Archaic horizons (Bullen
Medium sized trianguloid engraving tools (Figs. 2, i-1; 7, a5-6, e2; 8, k) These tools
vary from 2 to 5 cm. in length and have a thickness of about .7 to 2 cm. They are unifacial
with roughly chipped scraper-like sides and an intentionally made point or spur. In some
cases a graver blow is indicated (Fig. 2, j-k).
Presence of these tools in a Bolen (Dalton period) site in Early County, Georgia, has
been mentioned. What may have been trianguloid engraving tools are illustrated for Linden-
meier (Roberts 1936: Pls. 5, f-g; 7, g), Bull Brook (Byers 1954: Fig. 92, i4-5, jl), Del-
bert (MacDonald 1968: Pl. 15, e, h), McConnell (Prufer 1963: Fig. 10, 49) Avery Island
(Gagliano 1967: Fig. 17, E, H-I) and probably the Lecroy site (Lewis and Kneberg 1956:
Pls. 4, 1-3; 6, 21), and Shoop (Witthoft 1952: Pls. 2, 1-3; 3, 21, 25, 30) sites.
Flake drills (Fig. 8, _-p) About 4 cm. in length, these are crudely made from a blade
or other narrow chip as perforators. Their ends are dulled from use.
This simple tool is not illustrated in any of the reports mentioned above except for
the Lecroy site (Lewis and Kneberg 1956: P1. 4, 1-6). It may relate to the Archaic period.
In Florida two (FSM-102298) came from the Matthews Field site near Gainesville. They
were well patinated and the field has produced a fair number of Bolen points. However, it
has also produced many specimens belonging to the Archaic period. The latter are not,
however, as much patinated.
BULLEN AND BEILMAN
Small Waller Knives (Figs. 2, h; 8, r) These flake or blade tools have side notches pre-
sumably for hafting. Fine sharpening retouch appears along their edges but otherwise, ex-
cept for the notching and blunting of the bases, they are unworked. They resemble closely
except for size the Waller knive (Waller 1971).
Larger or Normal Sized Chipped Tools
Blanks or broken knives (Fig. 6, ee-mm) We have included these specimens here on the
chance they may represent a stage in the manufacture of small scrapers. Available exam-
ples are about 3 cm. across and 2 to 3 cm. perpendicular to the broken straight edge.
Medium sized amorphous scrapers or scraper knives (Figs. 7, e-h; 8, ul-4; 11, o-r) -
These miscellaneous scrapers, knives, or scraper-knives are commonly found in all
Florida Archaic period sites, especially workshops. They may also belong in the Paleo-
Hafted scrapers (Fig. 5, hh-nn; 12, 1-m) These tools--either especially made or re-
chipped from broken points--have Archaic tangs for hafting. At one site beside the Chatta-
hoochee River in West Florida they have been radiocarbon dated to 1200 B. C. (Bullen 1958).
Archaic stemmed drills (Figs. 8, h-j; 12, o-p) These expanded base drills are another
standard Archaic chipped stone tool.
Clear Fork gouge Not illustrated is a small but typical Clear Fork gouge. It is 4 cm.
long, 2. 5 wide at the bit, and 2 cm. thick. The steep bit edge is concave and exhibits use
chipping. The top of this tool has been hammered while the smaller end shows the result
Before closing this section, we should mention that one good-sized steatite vessel
fragment was also found in Site Po-15. Perhaps we should also advise that while we have
tried to include all significant chipped stone tools in the foregoing lists, we make no claim
for absolute completeness. For example, some broken tools with straight scraper-like
edges are illustrated in Figure 9 (o-s). We did not include them as special tools but perhaps
we should have done so. The same might be said of the spokeshave illustrated as dd in the
same figure. Our emphasis has been upon the small tools which are relatively new to
We have described and illustrated the small, predominantly unifacial, chipped stone
tools found at Nalcrest. Unfortunately, the site has completely eroded away so that excava-
tion is impossible and the many specimens salvaged by Beilman and others must be treated
as a surface collection. Two important questions remain. One refers to dating and the other
As the Nalcrest collection includes specimens from several archaeological periods
and relative dating at the site is impossible, clues as to their age must come from data ob-
tained at other sites. We have briefly sketched available distributional data.
Table 3 presents a synopsis of this data for nine small tool types. It seems evident
from this table that certain typical Paleo-Indian traits--tiny side scrapers or drills, spur-
red chips, and spurred-thumbnail scrapers; possibly trianguloid engraving tools--are linked
PROVENIENCE OF NALCREST SMALL TOOLS
Tools 0 U *D
Sites i 4 l lg le meu de
Lindenmeier x x ?
Bull Brook x x x ?
Debert x x x
William son x
Shoop x ?
Brand x x x R x x
Quad x x
Stanfield-Worley x x 1 ? x x
Darby Springs, Fla. x
'. '.4 -;
Early County, Ga. x x x
R indicates rare
traits found in both the Paleo-Indian and succeeding Dalton periods. Waller knives and less
probably --due to their crude construction--flake drills appear limited to Paleo-Indian times.
BullSmall to tiny cores, whether or not spurred, small ovate scrapers, and the small concave
Debert x x x
Williend scrapers (hated spokeshaves) seem limited to sites of the Dalton time period.
Shoop x x ?
Lecroy x ? x 1
Nalcrest x x x x x R x X 2
In Florida, Georgia, and Alabama these small tools are associated with Bolen Plain,
Bolen Beveled, and the very similar Big Sandy I points of Alabama (DeJarnette et al. 1962:
48-49, Fig. 47, top row) and hence may be dated as belonging to the Dalton time period.
Brand x x x R x x
Quad x x
Ch arcoa l from th e Dalton zone at the Stanfield-Worle rock shelter has been dated at ich-
igan Darby Springsd 8, Fla0 400 B.P. (Samples and 1153, Darnette et al. 196
85-86)or around 7330 BGa. It may be assumed this date applies approximately to the Dal-
ton period occupation at Nalcrest. In this respect, it is interesting to note that points which
we would in Florida without hesitation class as Bolen points (Mason 196rare37, Fig. 3 were
traits found at the Simonsen bison kill in Iowa. They came from Level 7, where they were and less
probciated with Bison occidentalis, which is radiocarbon drillated appeal 6,471 B.C. or limited to Peo-Idi times.
Small to tiny cores, whether or not spurred, small ovate scrapers, and the small concave
end scrapers (hafted spokeshaves) seem limited to sites of the Dalton time period.
(Agogino and Frankf, Georgia, and Alabama these small tools are associated with Bolen Plain960).
Bolen Beveled, and the very similar Big Sandy I points of Alabama (DeJarnette et al. 1962:
48-49, Fig. 47, top row) and hence may be dated as belonging to this very early Florida
culture period is harder ton zone at the Stanfied-Worley rock shelteing or has beengraving tools mandated at Mich-
igan to 9, 640 h 450 and 8, 920 b 400 B. P. (Samples 1152 and 1153, DeJarnette et al. 1962:
85-86) *or around 7330 B. C. It may be assumed this date applies approximately to the Dal-
ton period occupation at Naicrest. In this respect, it is interesting to note that points which
we would in Florida without hesitation class as Bolen points (Mason 1962:237, Fig. 3) were
found at the Simonsen bison kill in Iowa. They came from Level 7, where they were asso-
ciated with Bison occidentalis, which is radiocarbon dated at 6, 471 B. C. or 8, 431 520 B. P.
(Agogino and Frankforter 1960).
The other question, regarding the function of these tools in this very early Florida
culture period is harder to answer. It is evident that incising or engraving tools and uni-
BULLEN AND BEILMAN
facial scrapers were very common and hence of great economic importance. It is further
evident, presumably, that these tools were primary tools used in the production of secon-
dary tools or of ornaments of bone and wood. The manufacture and decoration of atlatls
and spears as well as the development of animal traps and of new bird and fish procuring
techniques, made important by the loss of large Pleistocene animals as a food supply, may
explain the expansion of the Dalton period tool kit.
Certainly the emphasis on small, usually unifacial, tools at Nalcrest is impressive.
It seems to have been an important center in immediately post-Paleo-Indian times. Was it
a place of manufacture of small tools which were then distributed over a large area? Or
was it a manufacturing center for atlatls, leisters, and other specialized food procuring
tools? Obviously, Florida archaeology would profit tremendously from the excavation and
study of Paleo-Indian and Dalton period sites--sites like the Brand site where working
floors with differential distribution of tools could be critically analyzed.
The Nalcrest and Dalton period unifacial and spurred tools had their earliest devel-
opment during Paleo-Indian times. While typologically different, many unifacial scraping
tools were used in Archaic times. We would agree with DeJarnette et al. (1962:88) that the
Dalton period is transitional between the Paleo-Indian and the Archaic cultures and is
"best explained as a bridge between the two. "
Agogino, George A., and W. D. Frankforter
1960 A Paleo-Indian bison-kill site in northwestern Iowa. American Antiquity,
vol. 25, pp. 414-15.
Bullen, Ripley P.
1958 Six sites near the Chattahoochee River in the Jim Woodruff Reservoir area.
River Basin Surveys Papers, no. 14. Bureau of American Ethnology,
Bulletin 169, pp. 315-57. Washington.
1968 A Guide to the Identification of Florida Projectile Points. Florida State
Bullen, Ripley P., and Laurence E. Beilman
1972 The Nalcrest Site, Lake Weohyakapka, Florida. Bulletin of the Eastern
States Archeological Federation, no. 31, `pp. 9-10. Ossining.
Bullen, Ripley P., and Marjorie H. Wing
1968 A scraper with graver spurs from Florida. Florida Anthropologist,
vol. 21, nos. 2-3, pp. 94-95. Tallahassee.
Byers, Douglas F.
1954 Bull Brook--A fluted point site in Ipswich, Massachusetts. American
Antiquity, vol. 19, no. 4, pp. 343-51. Salt Lake City.
Cambron, James W., and David C. Hulse
1960 An excavation at the Quad site. Tennessee Archaeologist, vol. 16, no. 1,
pp. 14-26. Knoxville.
DeJarnette, David L., Edward B. Kurjack, and James W. Cambron
1962 Stanfield-Worley Bluff Shelter, Excavations. Journal of Alabama Archaeology,
vol. 8, nos. 1-2. University.
Gagliano, Sherwood M.
1964 An Archaeological Survey of Avery Island. Avery Island, Inc. Louisiana.
1967 Occupation sequence at Avery Island. Coastal Studies Series, no. 22.
Louisiana State University Press. Baton Rouge.
Goodyear, Albert C.
1971 The Brand site: The Dalton tool kit with intrasite analysis. Unpublished
Master' s Thesis. University of Arkansas. Fayetteville.
Haln, Mrs. Richard N.
1969 Catalog of Alabama tools. Journal of Alabama Archaeology, vol. 15, no. 1,
pp. 1-19. Moundville.
Lewis, Thomas M.N., and Madeline Kneberg
1956 The Paleo-Indian complex on the Lecroy site. Tennessee Archaeologist,
vol. 12, no. 1, pp. 5-11. Knoxville.
Debert: A Paleo-Indian site in central Nova Scotia. Anthropology Papers,
no. 16. National Museum of Canada.
McCary, B. C.
1951 A workshop site of early man in Dinwiddie County, Virginia. American
Antiquity, vol. 17, no. 1, pp. 9-17. Salt Lake City.
Mason, Ronald J.
1962 The Paleo-Indian tradition in eastern North America. Current Anthropology,
vol. 3, no. 3, pp. 227-78. Chicago.
Morse, Dan F.
1972 Dalton culture in northeast Arkansas. Bulletin of the Eastern States
Archeological Federation, no. 31, p. 13. Ossining.
1963 The McConnell site, a late Paleo-Indian workshop in Coshocton County, Ohio.
Scientific Publications of the Cleveland Museum of Natural History, new
series, vol. 2, no. 1.
Ritchie, William A.
1953' A probable Paleo-Indian site in Vermont. American Antiquity, vol. 18,
no. 3, pp. 249-58. Salt Lake City.
Roberts, Frank H. H., Jr.
1936 Additional information on the folsom complex. Smithsonian Miscellaneous
Collections, vol. 95, no. 10. Washington.
1973 Letter of 15 March with photographs of specimens from a wet meadow
pond in Early County, Georgia.
Waller, Ben I.
1971 Hafted flake knives. Florida Anthropologist, vol. 24, no. 4, pp. 173-74.
BULLPEN AND BEILMAN
W itthoft, John
1952 A Paleo-Indian site in eastern Pennsylvania: An early hunting culture.
Proceedings of the American Philosophical Society, vol. 96, no. 4,
pp. 464-95. Philadelphia.
Florida State Museum
0 1/2 1
Fig. 1. Contour map of bottom of Lake Weohyakapka locating Indian sites
and hypothetical stream channels.
14 NALCREST SITE
Fig. 2. Pottery from sites Po-14 and Po-16 and chipped stone specimens
collected by Beilman at Po-15 and donated to the Florida State
Museum (FSM- A161).
a-c, Belle,Glade Plain sherds; d-f, spurred tools; g, tiny drill-scrapers;
h, small Waller knife; i-1, trianguloid drill-scrapers; m-o, r-s, spurred
core-scrapers; p-q, tiny core-scrapers.
BULLEN AND BEILMAN
Fig. 3. View of shore and sand dune overlooking mid part of Nalcrest
site (Po-15) and specimens found during winter of 1970.
c-l, Santa Fe; 2, Clovis-like; 3-6, Bolen; 7-8, tiny stemmed points; 9-11,
tiny hafted scrapers; 12-13, drills; 14-20, small concave end scrapers; 21,
end and side scraper; 22-27, ovate scrapers (some with small spurs). a
J k I m f P
t u v x
a b c d a f
S k I m n o p
q r a t u v w
oa bb ce dd ff
h h tM Mnta
hh-jj, Culbreath; kk, Westo; and 11-mm, Sumter-like.
Fig.4. Points from Beilman 1969-71 collection, site Po-15.
a, reworked Clovis or Santa Fe; b, Dalton-like; c-n, Bolen
(c-j are beveled); o-q, Bradford-like; r-s, Greenbriar-like;
t y-gg, Archaic stemmed; v-w, Newnan; x, Hillsborough;
hh-jj, Culbreath; kk, Westo; and Il-mm, Sumter-like.
z aa bb cc
** ft gf
Fig. 5. Points and hafted scrapers, Bellman 1969-71, site Po-15.
a-w, small Archaic stemmed; x-y, Westo; z-cc, Citrus; dd,
Hernando; ee-gg, Bradford-like points;andhh-nn, hafted
BULLEN AND BEILMAN
i n 0 q r t
4e ft 1,
Jj kk II mm
Fig. 6. End scrapers and possible blanks, Bellman 1969-71 collection, Po-15.
a-n, v-z, small concave end scrapers; o-u, unfinished concave end scrapers,
aa-dd, very small Clear Fork gouges; ee-mm, blanks(?).
18 NALCREST SITE
a' .J itB EBEBGi~
Fig. 7. Various types of scraping tools, Beilman 1969-71 collection, Po-15.
BULLEN AND BEILMAN
a b d f g h
K I m n o p
5 6 t a ie 1u t3 14 13 3 3 7 I12 1 n- ZI is 14 I
Fig. 8. Additional tools, Beilman 1969-71 collection, Po-15.
a-f, Pinallas points; g, trianguloid point similar those associated with Deptford
pottery in northern Florida; h-j, Archaic drills; k-p, crude drills; q, incising
tool; r, Waller knife; s-t, crude scrapers or knives; u-w, various types of scrapers.
Fig. 9. Specimens in Stegman 1970-71 collection, site Po-15.
an, very small Clear Fork gouges; o-u, various end and ovate scrapers;
a-n, very small Clear Fork gouges; o-u, various end and ovate scrapers;
v-bb, scraper-like tools with engraving spurs; cc, ovate scraper; and
BULLEN AND BEILMAN
o b c
X Y zI
bb cc dd f
Fig. 10. Specimens in Beilman 1972-73 collection, site Po-17.
a-e, tiny bifacially chipped artifacts with a twist slong their major axes; f-i,
tiny concave end scrapers; j-k, small triangular end scrapers; 1-P, long con-
cave end scrapers; q-dd, various end and side scrapers; ee-g, fragments of
NOT ' 4 WPT
ii 1 FHPSHAT ESHER INETPTE
Hi Hi plate
Tel (it~ tint tigt >a
OF TR 4 is
Fig.11. Points and knives, Beilman 1972-73, site Po-17.
a- -d, Bolen; e- i, unclassified Bolen- or Dalton-like; a n, Ar-
Chaic stemmed points; o-q, ovate and qsymmetric trianguloid
knives; r, crude knife or scraper.
Fig. 12. Scrapers and drills, Beilman 1972-72, site Po-l7.
a-c, f-g, end scrapers; d-e, large concave end scrapers;
h, side scraper; ah â€œJs ovate or circular scrapers; k, scra-
per with worn spur; l-m, hafted scrapers; n, scraper on
blade; O-p, Archaic â€˜drills; q, narrow Waller knife; Qr, drill;
8-t, unique object with worked edges. (q, Hamilton collection)
DALTON CULTURE IN NORTHEAST ARKANSAS
Dan F. Morse
Arkansas Archeological Survey
To some readers, using the word "culture", in referring to "Dalton culture", may
seem presumptuous. For in much of the Southeast, Dalton refers to a particular lance-
olate style of projectile point. The probable association of unifaces and other tools as ex-
emplified first in Graham Cave (Logan 1952) and then at the Stanfield-Worley shelter (De-
Jarnette et al 1962) has not resulted in a basic understanding of cultural behavior, at least
not in the p published literature. However, both investigations do represent an important be-
ginning toward understanding Dalton tool kits. The Ford-Redfield survey, called "Dalton
Project 1960-61", represents the initial effort at collecting data on Dalton points and pos-
sible associated tools in the northern delta of the Mississippi River (Redfield 1971). This
effort was continued in 1967 with the advent of the Arkansas Archeological Survey and in-
tensified in 1970. Lack of money and personnel and the need to be attentive to other ar-
cheological problems has meant a lessening of field research concerning Dalton by the
Survey. However, the 1970 field discoveries and the 1970-1971 laboratory analyses have,
we feel, given us ample justification to speak of Dalton culture.
The L' Anguille Phase of Dalton Culture
There is a repetitive pattern in the L'Anguille, St. Francis, and Cache River drain-
ages (Fig. 1) of two basic kinds of sites producing Dalton points. They seem to be base
settlements arid butchering camps. The pattern is most easily observed in the drainage of
the upper St. Francis River north of Marked Tree and in the L'Anguille River region.
Crowley Ridge, an erosional remnant 1/2-12 miles wide, divides these two areas with
most of the St. Francis River area northeast of the L'Anguille River drainage area (Fig. 1).
On the eastern edge of the present St. Francis River drainage at 8000-6000 B.C. were re-
lict channels of an old braided Mississippi River, apparently active during earlier fluted
point times, and further east the Mississippi River itself. The L'Anguille drainage at that
time included the meandering L'Anguille River and numerous relict braided channels of
the Mississippi River when it flowed here some 30,000 years ago.
West of the L' Anguille drainage is the Cache River which at 8000 B.C. may have
been a combined Cache, Black and St. Francis river and may have involved all of the
Ozark drainage north of the White River. Fluted points, for instance, are concentrated
along both sides of this ancient channel. The Daltonsettlement pattern is more complex
in this drainage and may involve smaller band territories plus a shifting drainage pattern
(and hence shifting sites) through time. One major difficulty is that we do not know when the
Black River began to be restricted to the edge of the Ozark Highlands or when the St.
Francis River finally broke through Crowley Ridge to flow east of the ridge.
In the L' Anguille River drainage north of Forrest City are 38 recorded Daltom sites--
21 per cent of the total sites known in the drainage. This is subject to revision whenever
our huge collections are finally analysed. An area of about 900 square miles is involved
(Fig. 1). Actually the potential exploitative reaches of a group based in this drainage area
Florida Anthropologist, vol. 26, no. 1, March 1973.
is closer to 1200 square miles. We feel that this area is essentially the territory of a
single family band between about 8000 and 6000 B. C.
In the north central part of the L'Anguille drainage is the Lace site. "This is
the most important and impressive Dalton site examined by this project" (Redfield 1971:
42). Within four miles are two smaller sites which produce similar varieties of artifacts,
and help identify a very concise area of base settlements. But the Lace site is obviously
the largest, highest, most important and longest used of all of these sites. Unfortunately
for us it has also been the longest and most intensively farmed because of its prominence
and little remains of it today. Ample evidence of artifact manufacture, particularly of Dal-
ton points, is present. In addition a large variety of end scrapers suitable for hide prepara-
tion has been found. This latter implies female activity at the site. Adzes are also common
and indicate a considerable amount of wood working. Everything points to a relatively
large area intensively used for most of what we feel is the totality of Dalton activities.
The other sites examined to date in the L' Anguille River drainage are strikingly dif-
ferent from Lace. Less than a quarter acre in total extent, Dalton points and a variety of
unifaces suitable for butchering and bone tool manufacture are present in large numbers.
But point reforms, flakes of bifacial retouch, serviceable adzes, and acute-angled end
scrapers are rare to absent. These sites are known to be very numerous while base settle-
ments are exceedingly rare. We call these smaller sites "butchering stations. The Brand
site (Goodyear 1971; Fig. Za) is a butchering station located 5. 5 miles from the Lace site.
Although only a quarter acre in total size, we could excavate but a total of 120 square me-
ters because of the time required to dig by natural stratigraphic levels and working floors
(Fig. Zb). Still, 1979 Dalton artifacts, including 305 Dalton points and point fragments, and
an estimated 50, 000 flakes were recovered. The Dalton tools were found mainly deposited
on top of a water laid soil and capped by a wind blown soil (Fig. 2c-g). Artifacts were com-
mon and occurred in mutually exclusive oval clusters, called working floors, measuring 8
to 16 square meters in extent. Luckily, the site does not seem to have been used extensively
(perhaps no more than a dozen times) and the working floors do not overlap. From 73 to 90
tools are definitely associated with each cluster. Cumulative curves indicate virtually iden-
tical percentages of tools in each cluster which strongly implies that the same basic activity
is involved at each working floor. Shelters (possibly lean-tos for shade) may have been pre-
sent and each cluster appears to have been a temporally discrete activity lasting only a few
days and involving only a few males.
The major importance of the Brand site to Southeastern archaeologists is as follows:
(1) Natural stratigraphy does exist and excavation can be accomplished by peeling off nat-
ural levels. (2) Working floors are present at these early sites and can be defined by care-
ful excavation. (3) In addition, there are significant inter-site differences, as exemplified
by Lace and Brand, and we not only have the potential data concerning artifact types other
than projectile points, but can now recognize tool kits resulting in considerable insight into
cultural behavior. The L'Anguille phase artifacts are now interpreted as the remains of an
exogenous patrilocal family band. Similar bands living in the nearby drainages probably
were associated and even intermarried with the L' Anguille drainage band, and probably
eventually will be included within the same phase because of archaeological similarity.
Stone Artifact Categories
This is not meant to be a complete listing of stone tools known or thought to be Dalton
in age in the tradition of Bordes (1961) and de Sonnerville-Bordes (1960). Such a listing now
is possible and an excellent beginning has been done by Goodyear (1971) although his list is
based on a butchering station and is not complete by any means. Such a list, of course, will
be valuable for discovering inter-site differences. We now recognize, although not as sys-
tematically as we want at this time, a variety of tool kits suitable for wood working, skin
processing, stone artifact manufacture, butchering and bone modification.
Denticulate Dalton point.
In northeast Arkansas there are a large number of points collectively called "Dalton".
There is a great deal of variability in the shape of both stem and body and an attempt was
initiated to type these. We soon discovered that the body variations are due to resharpening
a basically similar shape. Then we felt we should not formalize types based on basal shape
until we could better control our data. Major variations of the base involve a shortened
stem with flaring wide ears, approximating a side-notched appearance, and a prominently
shouldered stem often with a very deep basal concavity. Another stemmed variety appears
to be actually fluted rather than thinned. What used to be called a "lanceolate Dalton" is
now considered to belong to an earlier family of points which includes Quad-like forms and
is called Golondrina in Texas and Coldwater in Mississippi (Brown 1926:134). There are
some unique unnamed points which range up to almost 30 mm in length and appear to belong
within the Dalton family.
The basic characteristics of the typical denticulate Dalton point are a lanceolate point
with ground, parallel to concave, lateral stem edges and a ground, concave basal edge which
has been well thinned. It is important to recognize the various stages of manufacture and use.
The point was made for use and along the way aborted examples were discarded. Preforms
for Dalton points are lanceolate in shape and usually significantly larger than the mean of
complete points (Fig. 3a-b). Preforms range from percussion flaked oval forms with
straight or unmodified bases to a pressure chipped form ready to be serrated and basally
and laterally ground. We think the reforms are made on large, thick flakes detached from
the Pleistocene gravels of Crowley Ridge and weathered nodules of chert eroded from the
Boone and Pickens formations in the Ozark Highlands. Most of the aborted reforms were
used as scrapers and/or knives. Occasionally, one is even resharpened.
A NEW DENTICULATE DALTON POINT has convex body edges and serrations mea-
suring as long as Z mm. (Fig. 3c). Once these are worn down, probably on bone and antler,
the point is reserrated as a RESHARPENED DENTICULATE DALTON POINT (Fig. 3d).
The point normally has a "right-handed" bevel at this stage. The first resharpening causes
about a 4 mm. decrease in body width from the initial stage. Each successive resharpening
(including wear) reduces the width about 2-4 mm. (Morse 1971a-b and Goodyear 1971:59).
Experimentation has demonstrated that a well-made Dalton point can be resharpened and
serrated 3 times, if beveled, with a total width loss of only 8-9 mm. There seems little
doubt that these points are serrated knives. We have evidence of a very extensive bone
technology which almost certainly involved the manufacture of bone points and eyed needles,
and a serrated chert knife would be ideal for sawing off the ends of long bones in addition to
cutting during the butchering of deer. Whether the Dalton point was hafted on a foreshaft
rather than in a knife handle is not known but it is possible that these Dalton knives were
occasionally actually used as projectile points.
After the point could no longer be resharpened as a serrated knife, it was made into
a drill-shaped point which may or may not be beveled (Fig. 3e). The edge angle became
increasingly steeper during the resharpening and the EXHAUSTED DALTON POINT or final
stage may have been a scraper. Goodyear (1971:61) feels that the point was either no longer
hafted or was hafted differently during this stage. More experimentation and microscopic
observation need to be done here. The point was modified but we do not now know why.
Much that has been observed seems to be due to the retouching rather than to use.
Further modifications of exhausted or aborted Dalton points produced END SCRA-
PERS (Fig. 3i), BURINS (Fig. 3f-g), and CONCAVE-SIDED SCRAPERS (Fig. 3h). Usually
the body was transversely rechipped to a convex, acute angled and sometimes gouge-like
edge. One base has been observed made into an end scraper (Fig. 3j). Burination of the
tip is relatively common, with most bladelets removing the lateral edges and some over-
riding the face of the tip. The result is a chisel-like tip. Wear, resulting probably from
longitudinally grooving long bones and hollowing out antler, is easily recognized. The wear
plus patterning of bladelet scars argue conclusively for calling these tools burins. Another
modification is chipping a concavity in the lateral edge of a broken Dalton point. Rarely a
point can have a short drill-like tip at the transverse end. Yet another variation is a re-
chipped cylindrical body broken from an exhausted Dalton point (Fig. 3m).
This represents one of the earliest true adzes in the world and undoubtedly reflects
a basic cultural response to making shelters at permanent base settlements and possibly
dugout canoes (Morse and Goodyear 1972). The adz is basically oval in outline with a gouge-
like bit and heavily ground lateral edges (Fig. 3q). Haft polish is present on the polls of
many specimens and wear polish is always restricted to the obverse or convex face of the
bit. CHISELS are basically smaller, thinner adzes which may have been hand held, but
size and shape criteria to differentiate chisels and adzes have not been worked out yet.
There seems little doubt that these tools, called adzes, were hafted and used as adzes
rather than axes. More surprising is how common they are. At Brand, 72 fragments were
recovered and none of the specimens were being used as adzes. They represent specialized
modifications of the exhausted and broken adz form. With the possible exception of the Clear
Fork gouge, comparable tools have not been recognized outside this northeast Arkansas area.
ADZ PREFORMS (Fig. 3p) are present at base settlements as are the NEW ADZ and
RESHARPENED ADZ. EXHAUSTED ADZES have a bifacially battered bit as if used as a
chopper. Modification of the adz is usually into a core or a knife. Occasionally one is modi-
fied into a scraper or a wedge.
Goodyear (1971:87-88) describes two general classes of bifaces, segregated on the
basis of thickness, which may have been exhausted knives. All we can say is that such tools
exist and do not seem to be very important. They may be made on aborted reforms and ex-
hausted adzes and if nothing else indicate the maximizing by Dalton culture of their resources.
Cylindrical bifaces may have been made from the exhausted Dalton point, but some
may be made from flake reforms (Fig. 3k and 1). A suspected rare Dalton biface, based
on surface provenience and chipping pattern, is an oval-shaped knife (Fig. 3r). When found,
they concentrate at base settlements. True crescents are also present on Dalton sites
(Fig. 3o and n; both are broken in half). They seem to be specialized knives but late Archaic
remains are also at these sites and eccentric flints seem to be part of the northeast Arkan-
sas and southeast Missouri late Archaic pattern.
END SCRAPERS include steeply retouched varieties as well as transversely utilized
and/or lightly retouched flakes. These latter tend to be made on circular and oval-shaped
flakes. There is a great deal of variety in the class of steeply retouched end scrapers.
Most are retouched to a triangular or TRAPEZOID shape (actually, fan-shaped). This
shape allows the tool to be hafted easily by placing within a hollowed-out antler handle
without having to bind it in. The craftsman could have one handle and.a variety of end
scrapers. Variations of the working edge include a variety of edge angles. Acute angles
are thought to indicate skin preparation and steep angles to indicate wood or bone working
(Wilmsen 1970). As the scraper was resharpened, the angle naturally steepened.
These trapezoid end scrapers are retouched on all edges and are often laterally and
basally ground. The working edge is usually CONVEX (Fig. 4p-r) and rarely CONCAVE
(Fig. 4t). Graver spurs, both double and single, are present on many specimens and are
pointed corner extensions of the working edge (Fig. 4r-t). Some may merely be the result
of preparing a wide scraper for a narrow handle. Flakes used for end scrapers range from
true blades (Fig. 4h) to flakes of primary decortication (Fig. 4p). At or near the working
edge, the reverse surface of the flake tends to abruptly angle downward from the long axis
of the flake.
One blade, either detached from a biface core or a cobble being made into a poly-
hedral core is pointed (Fig. 4b). The tip and adjacent lateral edges have been used. The
base is chipped into a short stem which is laterally ground. The bulb of percussion has
been chipped away. Another similar blade has an adz-like bit and also has been prepared
for hafting (Fig. 4c). A 13 cm. long true blade has the bulbular end chipped and ground for
hafting, a steeply retouched end, and is laterally retouched and used (Fig. 4a).
CIRCULAR end scrapers tend to be circular in outline. Most have two retouched
working edges opposite each other. Occasionally steep retouch is on all edges and some-
times involves a graver tip. Rarely, SQUARE end scrapers occur. These all appear to
have been hand held.
SIDE SCRAPERS (called knives if bifacially worn) also include laterally utilized
and/or lightly retouched flakes and steeply retouched specimens. Side scrapers and knives
are cruder as a class than the end scrapers and few exhibit much preparation.
Present is a hafted BACKED BLADE made on a cobble being prepared for blade pro-
duction or a biface core blade (Fig. 4d-f). The bulbular end is prepared for hafting. One
known specimen is even stemmed and "fluted" (Fig. 4d). The natural back on a steeply
ridged flake (usually of secondary decortication) is used or one edge is chipped and ground.
The opposite edge is steeply retouched and tends to be concave and sharply angled to the
long axis of the flake. Exhausted tools have an almost 900 angled working edge. In Louisi-
ana and South Carolina, similar working edges are found on side-notched points (Webb,
Shiner and Roberts 1971: Fig. 10, d and f; Michie 1968).
TRUE BLADES (Fig. 4g-k) were used and/or lightly retouched for use as knives and
scrapers. The true blade is not only twice as long as wide--the criterion for blade--but
also exhibits two or more blade-like scars on the obverse surface which are parallel to the
blade' s long axis (Bordes and Crabtree 1969:1). This is an excellent indication that blade
production is deliberate and not accidental. Goodyear (1971:120) feels that blades were
snapped to provide a backing for the hand. Another characteristic Dalton flake knife is
made on a wedge-shaped flake with a cortex backing (Fig. 6e-g). These were selected for
and the pebble core (Fig. 6h-l) was probably chipped to deliberately produce them. The
other flakes chosen to make into scrapers and knives are basically core trim flakes which
do not now appear to differ significantly from those in later assemblages.
A fairly common artifact at Dalton sites is the GRAVER (Fig. 4 l-o). Combination
tools with end scrapers and graver tips have been discussed. Flakes were chosen and
graver tips retouched. In many cases, a large part of the flake' s edge was retouched.
Multiple gravers are present. Some tips are very fine while other tips are relatively large
and compare favorably with a used and/or burinated Dalton point tip. There is general
agreement that the presence of gravers indicates fine incising and shallow perforating
(e. g., eyed needles) in bone, antler and/or wood.
CONCAVE SIDED SCRAPERS (Fig. 4 1 and n) or spokeshaves are of fairly common
occurrence at Dalton sites. Those made on end scrapers have already been described.
Many are laterally steeply retouched on flakes, but most are simply utilized and/or lightly
retouched flakes. Some artifacts tend to be similar to European "notches" and "denticulates".
Small oval artifacts made on small blades are found in a Dalton context (Fig. 4 m
and n). They may be retouched on all edges and pointed as a graver or retouched trans-
versely and laterally for multiple functions. Some very fine bone and wood work must
have been done.
PIECES ESQUILLEES (MacDonald 1968) are small blocky flakes with evidence of bi-
polar flaking (Fig. 6 a and b). Simultaneous pressure was exerted from both ends. They
were manufactured by placing one end of the core (Fig. 6c) on a stone anvil (Fig. 5a) and
striking the other end with a stone hammerstone (Fig. 5d). This creates sharp angles at
one or both ends and an essentially blocky, rectangular-shaped flake.
They are characterized by having a primary or hammered platform and a secondary
or lower platform. The piece esquillee was probably used as a wedge to split longitudinally
grooved long bones. They not only occur in Dalton assemblages but also have been recog-
nized in all stone and bone working stages in northeast Arkansas. At the Brand site where
bone manufacturing was being done, this was a very common tool and was made at the site.
Small abraders were made of sandstone, ironstone and rarely from a pumice-like
stone. The most common type is a FLAT ironstone slab, sometimes with visible stria-
tions on one or both faces. GROOVED abraders (Fig. 5b) can be segregated into narrow
(3-9 mm) and wide (10-20 mm) classes. (Goodyear 1971:143). The thinner grooves sug-
gest bone abrading and some of the larger grooves may be for grinding the lateral edges
of adzes, antler cylinders, or for wooden foreshafts. The former are often multiple
grooved and the latter singly grooved. NOTCHED abraders (Fig. 5c) are found almost ex-
clusively at base settlements. Notches have been worn along and are perpendicular to one
or more edges of the abrader. Notches on the thicker sandstone abraders fit the lateral
edges of a Dalton point. Those made on a thin slab of ironstone fit the basal edge of Dalton
Cobbles here include weathered chert, quartzite, and sometimes hard sandstone
rocks probably quarried from gravel deposits and only minimally modified. The most
common is an end and/or edge BATTERED HAMMERSTONE. These occurred in cobble
concentration at Brand with ANVILS (Fig. 5a) and CHOPPERS. The anvils are pitted and
striated from percussion. Choppers with crude cutting edges on one end probably were
involved in butchering and possibly cracking bone for marrow. PITTED HAMMERS (Fig.
5d) almost certainly were used with pitted anvils for the manufacture of pieces esquillees.
The pits may occur naturally from use and often both faces of a hammerstone will be
pitted and most of the edge battered.
Some large sandstones associated definitely with Dalton artifacts have deep, smooth
holes rather than pitted areas (Goodyear 1971:138-139). Some of these appear to be care-
fully drilled. Possibly these are for abrading the butts of shafts, awls, points or needles,
or maybe were involved in fire making. For now, we will reserve a statement concerning
Two types of cobble abraders have been recognized. Both probably were used to pre-
pare the edges of cores and reforms for knapping. Neither has been directly associated
with Dalton but both are present on the surface of Dalton sites. One is the EDGE GROUND
COBBLE (Fig. 5g) and the other is an EDGE STRIATED AND GROOVED COBBLE (Fig. 5
e and f). Quartzite is commonly used to grind chert striking platforms for the detachment
of flakes. The best examples known are the grooved quartz cobbles found in knapping kits
in Illinois Hopewell tombs (Perino 1971).
Since there is such a variety of bifacial and unifacial tools at Dalton sites, a con-
siderable amount of cores and flakes is expected. In addition, the debitage greatly supple-
ment, and even suggest ideas of how and where tools were used and manufactured.
The major three kinds of cores are called PEBBLE, (Fig. 6 e-h), BIPOLAR (Fig. 6c)
and BLADE (Fig. 6 m-p) cores. A variety of small flakes useful as immediate tools were
detached from pebble cores. One important flake is wedge-shaped with a natural, wide cor-
texed back and was selected for use as a backed knife or scraper (Fig. 6 e-f; White 1963:5).
Pebble cores probably exist throughout the prehistoric sequence and are definitely manu-
factured during the Mississippi period. At Brand they constitute a major activity, probably
to be associated with butchering.
Bipolar cores are result of manufacturing pieces esquilldes. Often it is difficult if
not impossible to distinguish between piece esquillee (Fig. 6 a and b), core (Fig. 6c) and
flake (Fig. 6d). Blade cores are of two kinds. One is a large biface prepared to detach a
single blade with the biface then made into an adz (Fig. 6p). Polyhedral or multiple faceted
cores for the production of true blades are rare but often can be reconstructed from the
blades, core preparation flakes (Goodyear 1971:116-121) and core rejuvenation flakes.
Flakes can be categorized as DECORTICATION (White 1963:5)--including SHATTER,
PRIMARY, and SECONDARY. Shatter flakes often result from breaking a cobble to see
how good it might be for chipping. A lot of these and cobbles with corners or ends removed
would indicate primary knapping activities, presumably at or near a source. Primary de-
cortication flakes (with cortex over all of the obverse face) and secondary decortication
flakes might indicate preparation of a cobble or pebble for tool or core manufacture. BI-
POLAR flakes (or columnar spalls) indicate the probability of making pieces esquillees.
There are FLAKES OF BIFACIAL RETOUCH (Bordes 1961:6) and FLAKES OF UNIFACIAL
RETOUCH. They are recognizable by virtue of the striking platform which is the edge of a
worn tool or edge prepared preform. Flakes of bifacial retouch exhibit the edge of the bi-
face. They can be from a point (lateral or basal thinning flake), adz (lateral or bit) or
One question often asked is "what did they hunt?" On the basis of the reconstructed
environment, whitetail deer were plentiful. Mastodon may have been present as well as
tapir, ground sloth and other open hardwood forest animals but we have no direct evidence
of their presence during Dalton times. However, we find a pattern of exploitation which
best fits the hunting of deer. At Brand, cobbles for cores and tools probably were stored
at a convenient hillock just before the hunt and the kill CARRIED to the butchering site.
The working floors or squatting areas are smaller than a mastodon or a buffalo. The con-
centration of tools also indicate butchering beneath a leanto of some sort. These sites
were used repeatedly--Brand at least 5 and possibly as many as 12 times. Bone tool
manufacture also took place at Brand. My understanding of most big game kill sites is that
these events took place at separate loci. In addition to all this evidence deer bone was
found in large quantities at Stanfield-Worley and the few preserved fragments at Brand
were deer. However, we expect that there was a great deal more variety to the Dalton
diet. The environment was rich and numerous animal, fish and plant foods were available.
Too much emphasis in the literature has been placed on "Big Game Hunting" without re-
gard to environmental considerations for Paleo-Indian in the eastern United States.
Another question is whether Dalton is "early Archaic" or "late Paleo-Indian". To
date this impasse hasn't really hindered our investigation. The basic lithic assemblage is
Paleo-Indian in nature and there are obvious correlations to Upper Paleolithic. The definite
trend in the lower Mississippi Valley is to call Dalton (and a southern variant known as San
Patrice) Paleo-Indian. However, in northeast Arkansas, we are dealing with two major pat-
terns of distribution and probably exploitation. Fluted points are concentrated along two ma-
jor rivers, indicating a tight riverine orientation during the terminal Pleistocene. Dalton
points are found on all land surfaces known to be inhabitable before and at the end of the
Pleistocene. In addition the transitional points such as Coldwater and Quad also occur on
Dalton sites, indicating the shift in settlement pattern occurred right at about the end of the
Pleistocene. The pattern continues through a period of Hardin and Cache River points after
the Dalton period. Furthermore, Dalton assemblages include the true adz and there can be
no doubt of a strong wood working tradition.
Recent field and laboratory work in northeast Arkansas has revealed a rich lithic in-
ventory for Dalton. Artifacts have been functionally related to each other and a rich bone
and antler industry postulated. We expect there were a number of wooden utensils judging
by ample evidence of wood working tools. Tools with a combination of functions and reworked
tools indicate a maximizing of stone resources. Experimentation has been used to test con-
cepts of tool manufacture and function with good results. Furthermore we have recognized
inter-site differences in artifact assemblages. Butchering combined with bone and antler
tool and blank manufacture took place at small sites which were satellite to a base settle-
ment. There is a possibility of permanent structures and/or dugout canoes. Finally we
have demonstrated the existence of working floors in natural depositional levels.
1961 Typologie du Paleolithique Ancien et Moyen. Publications de L' Institut de
Prehistoire de L' Universite de Bordeaux, Memoire No. 1. Bordeaux.
Bordes, Francois, and Don E. Crabtree
1969 The Corbiac Blade Technique and Other Experiments. Tebiwa, Vol. 12,
No. 2, pp. 1-21. Pocatello.
Brown, Calvin S.
1926 Archeology of Mississippi. Mississippi Geological Survey. University,
DeJarnette, David, et al
1962 Stanfield-Worley Bluff Shelter Excavations. Journal of Alabama Archaeology,
Vol. 8, Nos. 1-2. University.
1960 Le Paleolithiqte Superieur en Perigord. Bordeaux.
Goodyear, Albert C.
1971 The Brand Site: The Dalton Tool Kit with an Intrasite Analysis.
Unpublished Master' s thesis, University of Arkansas. Fayetteville.
Logan, Wilfred D.
1952 Graham Cave. Missouri Archaeological Society, Memoir No. 2. Columbia.
MacDonald, George F.
1968 Debert: A Paleo-Indian Site in Central Novia Scotia. National Museum of
Canada, Anthropological Papers No. 16. Ottawa.
1968 The Edgefield Scraper. Chesopiean, Vol. 6, No. 2, pp. 30-31. Norfolk.
Morse, Dan F.
1971a Recent Indications of Dalton Settlement Pattern in Northeast Arkansas.
Proceedings of the Twenty-Seventh Southeastern Archaeological Conference,
Bulletin 13, pp. 5-10. Morgantown, West Virginia.
1971b The Hawkins Cache: A significant Dalton Find in Northeast Arkansas.
Arkansas Archeologist, Vol. 12, No. 1, pp. 9-20. Fayetteville.
Morse, Dan F., and Albert C. Goodyear
1972 The Significance of the Dalton Adz in Northeast Arkansas. Manuscript on
file with Arkansas Archeological Survey. Fayetteville.
1971 Dalton Project Notes Vol. 1. University of Missouri, Museum of Anthro-
Webb, Clarence H., Joel L. Shiner, and E. Wayne Roberts
1971 The John Pearce Site (16CD56): A San Patrice Site in Caddo Parish,
Louisiana. Bulletin of the Texas Archeological Society, Vol. 42, pp. 1-49.
White, Anta M.
1963 "Analytic Description of the Chipped-stone Industry from Snyders Site,
Calhoun County, Illinois. In Miscellaneous Studies in Typology and
Classification by Anta White, et al, pp. 1-70. Anthropologica Papers
No. 19, Museum of Anthropology. Ann Arbor.
Wilmsen, Edwin N.
1970 Lithic Analysis and Cultural Inference: A Paleo-Indian Case.
Anthropological Papers of the University of Arizona, No. 16. Tucson.
Arkansas Archeological Survey
State University, Arkansas
February 29, 1972
r ,---" -- -----
.i t Drm"r s e
11 Of" E .CI8;f AtynY
A LACC Site
Fig. 1. Map of northeastern Arkansas locating sites and Dalton culture regions.
Fig. 2. The Brand site (a), excavation by natural levels (b), and in situ artifacts (c-g).
Fig. 3. Dalton bifacial tools.
a-e, stages in manufacture and resharpening of Dalton points; f- -j, m, modification of
Dalton points into other tools, arrows indicate number and direction of burin blows; k-1,
narrow, punch-like bifaces; n-o, crescents; p- q,adz preform and finished adz; r, knife.
I, O. 1 9 1 2
g h i j k n a
Sq r s t u v
Fig. 4. Dalton unifacial tools.
C C M
C 0M 1 2 3
Fig. 5. Dalton abrading and battering tools.
a, anvil associated with hammer (d) in Feature 6, Brand site; b, grooved abrader;
c, notched abrader probably used for grinding bases of points; d, hammer; e, grooved
preform or core abrader; f, notched core abrader; g, edge ground cobble used to abrade
reforms or core edges.
e f g
CM 1 2 3
S I a
Fig. 6. Dalton cores and other flake tools.
a-d, two Pieces Esquill'es, a core, and a columnar spall; e-g, naturally backed
flakes used as scrapers or knives; h-1, pebble cores probably used to manufac-
ture naturally backed flakes; m-p, various kinds of blade cores.
ARCHAIC HAFTED SPOKESHAVES WITH GRAVER SPURS
FROM THE SOUTHEAST
Albert C. Goodyear
A few years ago, in the course of examining collections of local amateurs and through
personal surface collecting, an obvious type of flaked stone tool was noted that occurred
mainly on the central coast region of peninsular Florida, primarily Pinellas County.
Though none of the specimens (Fig. 1, d-q) were excavated from a site, an Archaic period
was indicated, by using a presence or absence system of scoring sites for cultural periods.
Two years after this study the author departed for the University of Arkansas where
he became involved with research concerning a Dalton period site. The Brand site (Good-
year 1971) a Dalton butchering station in northeast Arkansas, yielded some of the same
tools previously encountered on the west coast of Florida. Because of the typological sim-
ilarity (Fig. 1, a-c) it was felt that a comparison between the two assemblages was timely.
This was particularly true since the excavated Dalton materials were able to illuminate
various technological and cultural aspects of the tools heretofore undetermined for Florida.
It might also be added that the author has observed this same type of tool present in the col-
lections of the Institute of Archaeology and Anthropology in Columbia, South Carolina.
Since the spokeshaves associated with the Dalton occupations of northeast Arkansas
are better described and because the excavated Brand site yielded more information re-
garding their temporal placement and probable function they will be presented first.
Hafted Spokeshaves from the Brand Site, Poinsett County, Arkansas
The Brand site, 3P0139 in the files of the Arkansas Archeological Survey, was a Dal-
ton site located on a small windblown hillock which was formed on what was probably a ter-
race of an old relict stream chahnel of the Mississippi River when it flowed to the west of
its present course before the end of the Pleistocene (Roger T. Saucier, personal communi-
cation). The Brand site was quite rich in artifacts with 305 representations of Dalton points
recovered (139 of which were complete enough to measure) several end scrapers (with and
without graver spurs on the bit), side scrapers, bifaces, gravers on flakes, hafted adzes,
spokeshaves on flakes, flakeknives, true blades used as knives, and the bipolar wedges or
pieces esquillees also known in the Old World as outils ecailles. The hafted spokeshaves
under discussion were a rather minor tool type in terms of numbers (4) as were other
steeply chipped spokeshaves (5).
One of the more important features encountered at the Brand site were 5 working
floors on which the majority of the tools and points were located. These working floors or
use areas were spatially discrete from each other and their distribution over the site sug-
gested they were temporally separated. The 5 working floors were very similar in terms
of the types and numbers of stone tools which argues for functional autonomy for the work-
ing floors. Because of the generally male orientation of all the stone tools, with the possi-
ble exception of the end and side scrapers, and because of the small floor area comprised
Florida Anthropologist, vol. 26, no, 1, March 1973
by each working floor, suggesting that only part of the band was present, the site was in-
terpreted as a butchering and bone processing station used by small groups of males. The
latter is important in our functional considerations of the hafted spokeshave with graver
The hafted spokeshave can be described as having a roughly ovoid-triangular shape
from the obverse view with one end tapered by predominant unifacial retouch to give a
stemmed effect (Fig. 1). The working edge is slight to deeply concave, formed by steep
unifacial retouch, and the variation in the amount of concavity is probably attributable to
repeated resharpenings. There are usually two graver spurs on either side of the working
concavity but one spur is usually more pronounced. The sides of the stem were heavily
ground on all 4 Brand site specimens indicating the tool was prepared for hating. Grind-
ing or smoothing for hafting on chipped stone tools can be distinguished from wear polish
by a dull scoured surface which does not reflect light well. Polishing, on the other hand,
is usually restricted to the edge exposed to the surface being worked and presents a shiney,
glossey surface that reflects light. Hafting is also suggested by the small size of the tools
(Table 1) as it would be difficult to use the implements with much force unless they were
hafted to permit greater mechanical leverage.
The scraping or working concavity bore little polish on the Brand specimens, and the
same was true of the underneath or reverse face. There was no reverse face wear chippage
or nibbling. The graver spurs in all cases were either battered or polished and one had
parallel striations running in the direction of the spur' s axis which were observed through
a 30X binocular microscope. Semenov (1964) has presented information regarding the value
of wear striations to archeologists who wish to reconstruct tool functions and the method of
use. In this case the parallel striations on the graver spur indicate the spur was being used
in a linear movement either in a uni-directional fashion or in a back and forth position. It
should also be stated that the graver spurs on these spokeshaves are considerably more
sturdy than the usual graver spur fixed upon flakes. The flakes used as preform flakes for
these spokeshaves are somewhat thick (Table 1) providing a denser spur than simple flakes
with marginal retouch.
Another observation recorded for the hafted spokeshave was the edge angle located in
the working concavity. Wilmsen (1970) has advanced the concept of edge angle as a possible
indicator of function using various Paleo-Indian unifaces of North America. Though only
dealing with a sample of 4 there was great consistency in the Brand site spokeshave edge
angles with 3 specimens measuring approximately 75 degrees and one with an inclination of
about 74 degrees. According to Wilmsen (1970:71), tools with high edge angles (66-75 de-
grees) would probably have been used for working wood, bone, and heavy shredding.
Hafted Spokeshaves from the Central Florida Gulf Coast
A total of 17 specimens were examined from the Florida Gulf coast, one from as far
north as the town of Wacissa in the Florida panhandle. It is not known which site produced
the spokeshave from Wacissa, hence, it was not included in the site study. Two tools were
found on early dredge-up sites along the Withlacoochee River near Inverness and one was
found on a sand dune on the Weeki Wachee River near the Florida Presbyterian Church
Camp. The remaining 13 specimens were from Pinellas County sites. This doubtlessly re-
flects heavier collecting in the Pinellas County area rather than any significant geographical
distribution. None of the Florida hafted spokeshaves were examined under a microscope nor
was the edge angle in the working concavity measured. Photographs were taken, however,
and to judge from the shapes of the Florida and Arkansas assemblages they are practically
Fig. 1. Hafted spokeshaves.
a-c, from Arkansas; d-q, from Florida.
I I I 1
identical (Fig. 1). It should also be mentioned that the hafted spokeshave is not found solely
on the west coast of Florida as Ripley P. Bullen (personal communication) has observed
several similar specimens in the Florida State Museum's research collections which orig-
inated in other parts of the state.
While none of the Florida spokeshaves were recovered from excavations, various
accompanying data, as suggested early, indicate that these tools are preceramic in time.
Of the 10 sites that are known to have produced these tools, all have also produced typical
Florida stemmed Archaic points such as the Levy, Putnam, Marion, and Alachua (Bullen
1968). Four of the sites have yielded lanceolate points that should date back to Paleo-Indian
times. By cultural periods the early Archaic (6 out of 10), and Archaic (10 out of 10) seem
to predominate. Six of the 10 sites are non-ceramic and all 10 sites have produced lithic
artifacts of preceramic origins. Though 4 of the 10 sites have produced some pottery, it
can be stated with some confidence that these spokeshaves have never been seen by the
author on a purely ceramic site.
The geologic context in which these tools have been found should also be mentioned.
In 8 of the 10 sites the implements were located in a dry, sandy soil that varied from pure
white to goldish-yellow in color. This indicates that a recent or dark midden is not present
and that the sandy soils, particularly the yellowish-gold sands, are fairly old. It is also in-
teresting to note that all but one of the 10 sites in question are located on or near present
day bodies of fresh or salt water.
All but Z of the Florida specimens were made on fairly thin flakes. The 2 exceptions
were made from reworked distal portions of broken projectile points. Fourteen of the 17
tools were created by steep unifacial retouch on the stem margins. Three others showed
slight bifacial marginal retouch, in addition to the predominant unifacial retouch, which did
not result in a sinuous cutting edge but probably served to further "back" or blunt the stem
margins for hating. Though not observed under a microscope, 3 specimens were very dull
on the stem margins paralleling the Brand site specimens that had been ground. Grinding
for hafting is further indicated by the base of the stem which was noticeably ground or smooth-
ed in two cases. The Florida tools also have the spur projections on either side of the working
concavity and one spur is unusually more pronounced than the other (Fig. 1, d-j). It was ob-
served that 3 tools exhibited spurs that were quite smooth to the touch and one spur pos-
sessed a high degree of polish. Unfortunately, not all of the Florida assemblage was sys-
tematically examined about the spurs and shaving concavity for wear patterns.
The age of the Florida spokeshave is not as easy to pin down as in the Arkansas case
but certain data are present that are pertinent. First, though not infallible, the presence or
absence study did strongly suggest that the common denominator of all 10 sites is the Ar-
chaic period. Secondly, one of the spokeshaves had been fashioned on the broken distal end
of a Bolen Beveled point which is known to have an early Archaic provenience (Bullen 1968).
The other reworked distal fragment was bifacially worked as is undiagnostic of any cultural
An Attempt at Experimental Archaeology
To better understand the functions and wear patterns of the hated spokeshave with
graver spurs a simple experiment was performed. Following the lead of Wilmsen (1970:71)
who feels that flake uniface tools with tapered margins or stems can best be considered
socketed tools, a section of deer antler was hollowed out for socketing purposes and a freshly
made hafted spokeshave like the one under discussion was fitted into the hole. The hated tool
was then applied to a 2- by 2-inch board of soft pine. When pulling the instrument toward
me it was found to be an effective device for smoothing the wood' s surface but very little
shaving was accomplished. The instrument was then pushed away in the opposite direction
and this time it served as an excellent shaver or miniature wood plane since many large
shavings were produced. The graver spurs were also applied to the wood and were seen
to be quite useful for slotting and grooving the board (Goodyear 1971:105).
While this experiment is incomplete in many ways it at least demonstrated that this
kind of tool in a haft could readily work wood. Some wear chippage or nibbling resulted on
the reverse face of the experimental spokeshave from the pushing or shaving function. The
Brand site specimens, however, do not possess any more than slight wear chippage on the
reverse face and it may be that they were only pulled up the scraped or smoothed surface
rather than pushed down it. The battered and worn condition of some of the obverse face
(working concavity) tends to corroborate this.
DIMENSIONS OF HAFTED SPOKESHAVES WITH
GRAVER SPURS FROM FLORIDA AND ARKANSAS
Assemblage Number Milliameters Edge
Max. Min. Mean angle
Brand site, 4 Length 29 19 23 750 Max.
Arkansas Width 22 14 17.5 740 Min.
Thickness 7 4.5 5.5 750 Mean
Gulf coast 17 Length 31 -
of Florida Width 21 -
Thickness 6.5 -
Two assemblages, one from Arkansas and the other from Florida, containing what is
apparently the same type of tool have been compared. It has been seen that the tools from
two different states are essentially the same, although the Florida specimens were not ex-
amined with the microscope for duplicate wear patterns.
Data have been presented for both assemblages which indicate that the hafted spoke-
shave has a late Paleo-Indian or early Archaic temporal placement in the Southeast.
The morphological traits of the tools as well as the wear patterns on the working con-
cavities and spurs indicate the concavity or spokeshave edge of the implement was used to
scrape or smooth what must have been rounded or cylindrical surfaces in a movement where
the tool was probably drawn toward the craftsman. The lack of reverse face polishing and
nibbling indicates that the tool was not regularly pushed away from the user, or at least work
movement was not preceded with the obverse face first. The experiment with the hafted spoke-
shave suggests that these tools would be excellent for working bone and wood. The other tools
found in the working floors of the Brand site further suggest that a bone or wood material was
being rocessed since many of the associated tools such as burins (on Dalton points), pieces
esquil ees, gravers, cobble anvils and choppers, grooved ironstone abraders, all strongly
suggest a bone working industry.
The gravers spurs on the hafted spokeshaves are analogous to the heavy duty graver
spurs on end scrapers from the Brand site which revealed evidence of heavy usage in the
form of battering and polishing. MacDonald (1968:92-93) states that at the Paleo-Indian
Debert site in Nova Scotia the spurs on the end scrapers were also probably used to groove
or slot bone preparatory to splitting. Given the male oriented tasks of the stone tools from
the Brand site, the hafted spokeshave was probably a man' s tool involved with the manu-
facture of other tools and possibly tool hafts made of bone and wood. The hated spoke-
shave with graver spurs, an unheralded tool in the Southeast, should be given considera-
tion as a possible temporal marker of Archaic man in the Southeast.
I should like to thank Richard MacDonnel, Mark Brooks, Phil Jordan, Walter Askew,
and Lyman O. Warren for allowing me to inspect the tools from their collections from the
west coast of Florida. Frank Bushnell did the photography of the Florida specimens and
his cooperation is gratefully acknowledged. Dan Printup, photographer for the Arkansas
Archeological Survey, photographed the Brand site specimens. The Arkansas Archeological
Survey should also be cited for its sponsorship of the Brand site excavation, particularly
Dr. Dan F. Morse the Brand site field director, as the excavation and analysis provided
much of the supporting data for this paper.
Bullen, Ripley P.
1968 A Guide to the Identification of Florida Projectile Points. Florida State
Goodyear, Albert C.
1971 The Brand Site: The Dalton tool kit with an Intrasite Analysis. Unpublished
masters thesis on deposit with the University of Arkansas, Fayetteville.
MacDonald, George F.
1968 Debert: A Palaeo-Indian Site in Central Nova Scotia. Anthropology Papers,
No. 16, National Museums of Canada. Ottawa.
Semenov, S. A.
1964 Prehistoric Technology. Cory, Adams, & Mackay Ltd.
Wilmsen, Edwin N.
1970 Lithic Analysis and Cultural Inference: A Paleo-Indian Case. Anthro-
pological Papers of the University of Arizona, No. 16, Tucson.
Copies of THE FLORIDA ANTHROPOLOGIST may be purchased from
the Society at $6. 00 per volume ($1. 50 for single and $3. 00 for double numbers)
for Volumes I through XIX; for Volumes XX through XXV the price is $4. 00 per
volume ($1.00 for single and $2.00 for double numbers). Volumes 1:1-2; 6:1,
4; 8:2; 10:1-2; 13:1; 17:1; 18:1; 21:4; and 24:1 are out of print. An index,
covering volumes I-XXIV, is available at $1.00.
Copies of the FLORIDA ANTHROPOLOGICAL SOCIETY PUBLICATIONS,
a monograph series, are available at $2.00 each for Nos. 1, 2, 5, and 6.
No. 4 is $4.00 and No. 3 is out of print.
Orders with remittances should be sent to either the Treasurer or the Editor.
The Society has no resale discount for dealers or resale agents.
INFORMATION FOR AUTHORS
THE FLORIDA ANTHROPOLOGIST publishes original papers in all
subfields of anthropology with an emphasis on archaeology. Contributions
from allied disciplines are acceptable when concerned with anthropological
problems. The journal's geographical scope, in order of importance, in-
cludes Florida, Southeastern United States, Caribbean, and related regions.
Manuscripts should be double spaced and typed on one side only of an
8 1/2 by 11 inch sheet leaving at least a 1 inch margin on all sides. Contri-
butors are referred to American Antiquity and the Chicago Manual of Style
for matters of style and reference. Textual references should be similar to
"Smith 1964: 78-80; Jones 1969: P1. II, b) No footnotes. In the "Refer-
ence Cited" section, at the end of an article, do not abbreviate names of ser-
ies or journals. Books need full title, publisher's name, and place.
All illustrative material-line drawings (in India ink) ard photographs-
should be included in one numbered series of "Figures" suitable for a maxi-
mum reduction to 6 1/2 by 8 1/2 inches. Parts of figures are designated by
lower case letters. Figures must be accompanied by brief descriptive cap-
tions. All tabular material must be included in separately numbered "Tables. "
Twenty-five reprints without covers of each article are furnished free
to each author or group of authors. Additional copies, with or without covers,
may be purchased but the editor must be informed of the number desired when
the article is offered for publication.