Â© 2000 Florida Anthropological Society Inc.
The Florida Anthropological Society Inc. holds all rights to the
source text of the Florida Anthropologist and shall be
considered the copyright holder for the text 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 â€œfair useâ€
provisions of U.S. Code, Title 17, Section 107 are restricted.
Contact the Florida Anthropological Society for additional
information and permissions.
PUBLISHED BY THE
FLORIDA ANTHROPOLOGICAL SOCIETY
MMBN SMS E lG[a M
THE FLORIDA ANTHROPOLOGIST
Vol. V May, 1952 Nos. 1-2
An Archaeological Manifestation of a Natchez-Type Burial
Ceremony - - - -I- illiam H. Sears 1
J. Clarence Simpson: 1910-1952
------- ---------- --lJohn i. Griffin 8
The Manufacture of Fluted Points
---------- ------------ -Wilfred T. Neill 9
Some Problems in the Practical Application of Somatotyping
------------------- -A. K. Bullen 17
The Harbor Key Site, Manatee County, Florida
-- ------- - --- -Ripley P. Bullen 21
Graham R. Reeder, Bonnie Bell, and Blake Whisenant
Contributors to This Issue
Editorial Office at the Florida State University, Tallahassee
Printed at the University of Florida, Gainesville, for the
Florida Anthropological Society
AN ARCHAEOLOGICAL MANIFESTATION OF A NATCHEZ.
TYPE BURIAL CEREMONY
William H. Sears
This paper represents an attempt to indicate the existence of a generic
relationship between the mortuary ceremonialism as expressed archaeologi-
cally in a burial mound at Kolomoki and that described as having taken place
among the Natchez Indians and their neighbors in the seventeenth century.
The descriptive archaeological data will necessarily be kept to a rather
sketchy outline, and the Natchez data reduced to a similar outline. A prob-
able area through which this type of ceremonialism is found in archaeolog-
cal expression is also indicated.
I might first point out the salient features of the historic ceremonialism,
features culled from the several accounts covering a number of groups and
some depth in time, as given in Swanton's "Indian Tribes of the Lower
Mississippi Valley and Adjacent Coast of the Gulf of Mexico." (Swanton,
1911). The ceremonies in which we are interested, in fact almost the only
ones described in the seventeenth century accounts, are those which were
initiated at the death of an individual from the top stratum of a strongly clas-
stratified society. Such individuals were a chief, a chief's wife, or a war
chief. While minor ceremonies were being carried on around the death bed of
the individual concerned in his or her house, scaffolds were erected in the
public square and the important ceremonies to follow were organized and
Later, while the body was being carried to the place of burial on a
litter, numbers of wives, retainers, and relatives were ceremoniously stran-
gled on the scaffolds. Burial then seems to have followed a pattern of rigid
placement according to rank; the wife or wives in the grave with the chief
inside the temple, individuals next in importance in front of the temple, to
the left and right of the door, and so on.
Other features which may be of some significance in this discussion are
indications in the literature of later exhumation of these bones and storage
in the temple, retention of other bones or dried bodies in the temple, trophy
skulls used as temple decorations, and the occurrence of pottery figurines,
wooden figurines and stone idols in the temple (Swanton, 1911, pp. 138-181).
From one point of view, then, these funerals, with their retainer sacri-
fice and burial, elaborate processions, set positions for graves, and all of
the other elements noted in the historic accounts, may be regarded as mor-
tuary memorializations of the strongly stratified social system. They do not,
This paper was read at the seventeenth annual meeting of the Society for American
Archaeology in Columbus, Ohio, in May, 1952.
as far as I can see, bear any relationship to the Southern Cult, except that
they are elaborate ceremonies. As Waring and Holder (1945) have indicated,
and as Waring has demonstrated clearly in an unpublished manuscript, the
Cult, as we know it from the artifacts found at such sites as Etowah and
Spiro, must have been an early expression of the agricultural ceremonies
coming down to historic times in the green corn ceremonies of many peoples.
That these ceremonies may have been practiced in early times by people
having a rigidly stratified social system and its related mortuary practices
is irrelevant here. The important fact here is that there were two distinct
ceremonial complexes, mortuary and fertility rites, and that there is no
necessary relationship between them.
Mound D at Kolomoki seems to reflect the same sort of social structure
and associated mortuary ceremonialism as that described for the Natchez and
their neighbors. I will outline here the physical archaeological data in terms
of. mound construction by the natives, and not according to the excavation
procedure of the archaeologist.
The first step was cleaning the ground surface down to sterile soil over
an area about thirty feet in diameter, even scraping aside a few inches of
rather thick midden. In this area were buried five individuals in log-outlined
tombs. After the graves were filled in, six large logs, each about eight feet
high, were held upright on the ground surface, and earth piled around them.
This formed a small primary mound, which was covered loosely with rocks.
The tops of the logs protruding from it were connected with lighter poles,
forming a scaffolding about seven feet high. One trophy head with a copper
and pearl cymbal-shaped ornament on the forehead was placed on top of this
primary mound, under the center of the scaffold.
Three burials were made after the scaffold was completed and before
any more earth was added to the mound. One male, accompanied by many
large barrel-shaped conch shell beads, was placed in a grave just off the
south edge of the scaffold. The grave was lined and covered with rock slabs.
Two other individuals (the only two definitely female skeletons in the mound)
were placed in two more rock slab graves just in front (i.e., east) of the
scaffold. The male in the rock slab grave was, I am sure, the individual
whose death initiated the entire cycle of ceremonialism and mound building.
This is attested by the special grave type, grave construction only after
completion of the scaffold, and the orientation of later developments around
the location of his grave. Also of considerable importance at this point is
the fact that, after this grave was filled in, it was covered with a rectangular
pole framework which can only have been a litter.
These three graves, and consequently the primary mound, were then
covered with a mantle of yellow clay, given the shape of a small, rectangular,
flat-topped mound with rather steep sides, a miniature of the temple mound
at this site. At the south end of this structure, over the grave of the para-
mount individual, the same clay was piled about a foot higher in the form of
a disc. The only function of this disc seems to have been to mark the area
of the important burial it covered. This area continued to be important in
further ceremonies and in the construction which was a part of the cere-
A mass deposit of pottery vessels, trophy heads, and conch shell dippers
and fragments was next placed in front (along the central east edge) of this
flat-topped mound. This deposit contained over one hundred vessels, large
parts of many others, several thousand sherds, and five trophy heads, two
of them with copper ornaments and beads still in place. Many of these ves-
sels were in very elaborate forms, animal, bird, and human effigies being
particularly prominent. While many of these vessels were made: specifically
for mortuary offering, as indicated by their pre-cut kill holes, others, lacking
this feature, seem to have been designed originally for non-mortuary functions.
Some of these latter, killed by having their heads knocked off, are specimens
of particularly good and elaborate workmanship. This leads us to. believe that
their original functions were socially important ones, and it seems at least
possible that they were used as temple furniture.
A number of other trophy heads were deposited around the edges of this
small platform mound, and several complete bodies were laid around its
edges in log-outlined tombs. More earth was then added over the entire area.
Over the- northern half, from which the remains of the scaffolding still pro-
truded, a specially selected brown earth was added. More bodies were placed
in a steady succession of interments over the southern half of the mound. In
order of interment, starting almost directly on top of the disc-shaped addition
to the core mound, we have a complete extended body in log-outlined tomb.
Following this, another extended body was placed in log-outlined tomb. A
second may also have been deposited at the same time, just to the south.
Two more bodies, in log tombs again, followed these. Next in this area, still
over the disc-shaped cap on the miniature platform mound, a cremation in
situ of bones from at least two individuals was undertaken. Finally, two
trophy skulls and a long bone bundle were deposited. All of the complete
bodies in the log-outlined tombs were partially cremated in place. Rocks
were placed over the ashes and partially cremated bone, and yellow clay was
added before another body was laid down.
Interlensed basketloads of yellow and brown earth, at the point of junc-
ture between the two types, indicate clearly that piling of brown earth over
the northern half of the platform mound went on simultaneously with the dep-
osition of bodies and addition of yellow earth over the southern half. To
complete this stage of building, a very thin layer of yellow clay was placed
on top of the brown earth.
On completion, this construction stage was roughly circular in ground
plan, but had a flat top. A mass cremation took place on this flat platform,
which was some fifteen feet high. The area as we found it was covered with
charcoal, decayed and partially calcined bone, and rocks over the charcoal
and bone. Several areas of concentration indicated that assortments of bones
which might have represented complete bodies had been burned, and in other
spots single skulls were present.
A rather confusing development took place at this point. Three large pits,
their bases lined with logs, were dug down into the yellow clay slopes at the
southwest corner of the mound and along the rear (west) edge. A single ex-
tended body and several skulls were at the base of each pit. In the fill of
each were several more single skulls, put in at different levels, most of
them resting in hands with the forearms attached. Almost all of the cymbal-
shaped meteoric iron ornaments we found in this mound were associated with
Finally, covering all of the preceding construction, burials, and pottery
deposit, a mantle of red clay was added, varying in thickness from four to
five feet on top of the mound to fifteen to twenty feet on the lower slopes.
On a point to point comparison, (it may be seen that) the implied cere-
monialism in Mound D and that described for the lower Mississippi valley
groups share many basic features.
In Mound D, as with the Natchez and Taensa, the death of an individual
initiates an elaborate and costly set of ceremonies. Since they are elaborate
and costly, and are in the archaeological context oriented around this one
individual, his paramount position in the social structure is clear. Following
this death, we have, in Mound D, the clearing of an area and the burial in it
of a set of adult individuals. Since they were almost certainly fleshed bodies
and were buried at the same time, retainer sacrifice seems certain. Over their
graves, with the graves in some measure used as a foundation, a scaffolding
was erected, a feature appearing in connection with retainer sacrifice among
Natchez. Here it could have been used for the sacrifice of the two females
buried directly in front of it.
The litter found over the grave of the paramount individual implies that
the body was carried upon it to the burial area by a number of people. This,
plus the presence of the retainers and the two females, would indicate a
ceremonial procession, certainly including as participants the litter bearers,
the females who possibly were wives of the deceased, and relatives, servants,
or other retainers. Since the scaffolding implies sacrifice of some of these
people on the spot, we should also expect in the procession individuals
who were to strangle the wives and servants.
The erection of a small replica of the temple mound over these graves and
the rather definite alignment and spatial interrelationship between the graves is
analogous to the Natchez burials in the temple and in front of it, also in a defi-
nite pattern. In Mound D, since the ceremonies oriented around the main burial
were to continue, a spot over the main burial was marked by the disc-shaped
The next step in Mound D, deposition of the mass pottery and trophy head
cache, has no parallels in the Natchez descriptions. The large number of effigy
vessels specially made for mortuary use seems, however, to be related to the
whole complex of retainer sacrifice, through intent for service in the afterworld.
This seems to have been carried even further in Mound D than might have been
expected, since additional vessels, trophy heads, long bone bundles, and bas-
kets of cremated human remains which must have been part of the temple furni-
ture were also placed in the mound. Perhaps we are at a point here at which the
sacred character of the chief, or priest-chief, had developed to the point of God-
hood, and his identification with the temple, so that its furnishings, too, must
accompany him to the hereafter. Something of the sort may also have happened
among the Natchez, which would account for the arguments as to what was in-
cluded among the temple furnishings at slightly different points in time. (Swan-
ton, 1911, p. 164.)
The further sacrifice of retainers, indicated by the extended bodies placed in
log-outlined tombs as the mound was added to over the main burial, is simply,
then, an extension of the earlier sacrifices, a continuation of a series of rites
with such sacrifice as a major feature at many points. Perhaps the burials noted
in the large pits related in this way too, although they certainly look like after-
The final mass cremation at this stage, including largely human odds and
ends, is probably indicative of a final housecleaning in the temple, a continua-
tion of the process initiated by placing the more important objects in front of
the miniature temple mound.
There are, of course, loose ends to this point-to-point comparison. However,
general features tend to be very similar both individually and in the organiza-
tion of the ceremonial cycle. Such features are the initiation of the ceremonies
at the death of an individual of great social importance, a procession with the
individual carried on a litter, sacrifice of retainers, and burial in a specific spa-
tial pattern in a sacred area.
It seems to the writer that the implied ceremonialism in mound D differs
from that of the Lower Mississippi Valley largely in the direction of greater
elaboration. Such features as those listed below appear to be in the Natchez-
Taensa tradition, but the implications is that the concepts were carried further
in Mound D than in the Lower Valley in the 17th century. These features are:
Ceremonial clearing of a grave area; retainer sacrifice and burial as part of the
preparation for a ceremonial area; sacrifice of retainers at a number of points in
the ceremony, apparently before, during, and after the interment of the paramount
individual; the apparent inclusion of trophy heads, bone bundles, and baskets of
cremated human remains which had been stored in the temple; the burial of other
temple furniture in the mound (as noted above, these last two may have Natchez
parallels), and the completion of the burial area as a large mound.
Probably the sort of mortuary ceremonialism, and, for that matter, social
organization, typical of Kolomoki at its peak period of development and use,
represent both the complex and the social system at their peak. The mortuary
complex and social system of the Lower Valley in the 17th century, then, may
be viewed as dilutions through time and through participation by a smaller
group than that at Kolomoki.
This mortuary ceremonialism seems to have been the major diagnostic of
what has been described as Late Weeden Island culture, a mortuary complex
which moves through space and time apart from Weeden Island culture per se
(Sears, 1950, pp. 71-97). At Kolomoki, by the time Mound D was built, the
Weeden Island element had dropped completely from ceramics used in the vil-
lage areas. This also seems to have been true at the large Hall, Tucker, Mound
Field and Bird Hammock sites described by C. B. Moore (1902, 282-303, 275-
265, 306-320; 1918, 561-564), whose mortuary ceramic complexes and village
ceramic complexes are related to those at Kolomoki. (See Sears, 1950, pp. 51-
97, for mortuary assemblages. Judgment of relationship of village assemblages
is based on inspection of relevant collections.) There are smaller sites on the
Florida Northwest Coast in which this separation of "church" and "state" in
ceramics probably applies.
The same sort of mortuary ceremonialism, and consequently the same social
organization, seems to have been the mode in the Tampa Bay area in this time
period. After working through Mound D, where the clear stratification made the
interrelationships of the various elements clear, Ifeel that such mounds asThomas
Weeden Island (Willey, 1949, pp. 331 and 335) and Prine (Bullen, 1952) were the
products of similar sets of continuous mortuary ceremonies. In them, as in Mound
0, there is stratification of burials, with flexed burials lowest, extended and
secondary burials in a primary mound, and secondary burials in later levels. This
has been interpreted as the result of re-use through time by genetically related
cultures, with burial practices having changed as the mound accumulated. It
would seem that interpretation of this stratification as reflecting different cere-
monial functions for the individuals involved in each burial type and stratum,
following the Mound D data, is more probable. There may have been, however,
some late re-use of some of these mounds, as indicated by intrusive burials.
Although the evidence cannot be presented in detail here, there is evidence
that this same strongly stratified social system is expressed in burial mounds
through the entire Gulf Coastal Plain, from the Gahagan Mound to the Tampa Bay
Bay group, in a time period which seems to just precede the appearance of
Southern Cult paraphernalia. Since this is also a distinctive ceramic province,
in this general time period, it would seem that the sort of social organization
possessed in historic times by the Natchez and their neighbors was once com-
mon to a culture area extending along the coastal plain from east Texas to
Tampa Bay, a culture area which later broke down into more distinctive local
variants, in the early part of the Temple Mound II or Early Mississippi period,
probably under the impace of adjacent inland cultures which had been diverse
for a longer time span.
Bullen, Ripley P.
1951. "The Terra Ceia Site, Manatee County, Florida," Florida Anthro-
pological Society Publications, no. 3. Gainesville
Moore, Clarence B.
1902. "Certain Aboriginal Remains of the Northwest Florida Coast,"
Journal of the Academy of Natural Sciences of Philadelphia, vol.
12, pt.. 2. Philadelphia.
1918. "The Northwestern Florida Coast Revisited,'.' Journal of the
Academy of Natural Sciences of Philadelphia, vol. 16.
Sears, William H.
1950. The Prehistoric Cultural Position in the Southeast of Kolomoki,
Early County, Georgia. Published in microfilm, University of
Michigan, Ann Arbor.
Swanton, John R.
1911. "Indian Tribes of the Lower Mississippi Valley and Adjacent
Coast of the Gulf of Mexico," Bureau of American Ethnology,
Bulletin 43. Washington.
Waring, Antonio J., Jr., and Preston Holder.
1945. "A Prehistoric Ceremonial Complex in the Southeastern United
States," American Anthropologist, vol. 47, no. 1.
Willey, Gordon R.
1949. "Archaeology of the Florida Gulf Coast," Smithsonian Miscel-
laneous Collections, vol. 113.
University of Georgia
J. CLARENCE SIMPSON: 1910-1952
J. Clarence Simpson died at his home in Marianna, Florida, during the night
of March 29, 1952. With his passing, Florida anthropology has lost a sincere
friend and tireless worker. Geologists, botanists, zoologists, historians, and
others also will miss Bruce, as his friends called him, for he was a mine of in-
formation on all aspects of Florida.
In 1930 he joined the staff of the Florida Geological Survey, and devoted
practically all of his subsequent years to that service. His interest in arch-
aeology stemmed, his mother has said, from the finding of his first arrowhead
at the age of ten. It was an interest that grew and became more mature through
the years. The Simpson collection, the result of his private activities over a
number of years, is an outstanding example of a cataloged and cross-indexed
He supervised the WPA archaeological investigations in Hillsborough
County from 1935 to 1938. The period following the close of field activities saw
the dispersal and loss of much of the recovered material and many of the notes.
It is almost entirely through the personal efforts of Clarence Simpson that the
information available today was preserved, and for this alone Florida anthro-
pology owes him a sincere debt.
Illness had been his unfortunate lot for a number of years, but Bruce con-
tinued his work with the same cheer and vigor as always. His agile and pen-
etrating mind, with its vast store of knowledge, was always at the service of
his colleagues. The niche he left will not be filled.
John W. Griffin
THE MANUFACTURE OF FLUTED POINTS
Wilfred T. Neill
In 1926, near the town of Folsom, New Mexico, Dr. J. D. Figgins found
chipped stone projectile points in association with the remains of extinct ani-
mals. As time passed, artifacts more or less similar to these "Folsom points"
were found in many parts of the New World. The general term of "fluted point"
has been proposed for these artifacts (Shetrone, 1936). The name focuses at-
tention upon a remarkable feature of the points: a longitudinal flute or groove,
running down each face of the blade, from the base a varying distance toward
the tip. Several theories have been advanced as to why and how the flute was
made (Wormington, 1939). These theories will be considered briefly.
It has been suggested that fluting was done to facilitate hafting. This
seems unlikely, for the following reasons:
1. Some points are fluted obliquely, not longitudinally (Fig. 1, A). Occa-
sionally a point is fluted obliquely on each face, the flutes running in opposite
directions. Such fluting was scarcely done to simplify hafting.
2. Some points fluted on one side only (Fig. 1, B). Some are not fluted at
all, although otherwise similar to fluted points, with which they are associated
(Fig. 1, C). Still others may bear two or even three flutes of varying lengths on
one face of the blade, and a single flute on the other (Fig. 1, D). Many points
are fluted to the tip on one side and only near the base on the reverse side (Fig.
1, E). In some cases the flute is longitudinal but is not located along the mid-
line of the point (Fig. 1, F). Often the flute on one face of a point does not
correspond in position with that on the other face. A satisfactory theory should
account for all these variants; they are difficult to explain in terms of hafting.
3. Many points are fluted from base to tip (Fig. 1, G). It is unlikely that a
wooden shaft extended completely to the tip of the points, where its presence
would hinder penetration. If fluting were done merely to facilitate hafting, only
basal flutes would have been necessary.
4. A point is best hafted by insertion into a slot cut or worn in the end of a
wooden shaft. An unfluted point is just about as easy to haft in this manner as
a fluted one. Vast numbers of unfluted points apparently were hafted without
difficulty in past ages.
It has been suggested that the flute functioned like the so-called blood
groove of a bayonet, to increase bleeding, This theory is easily refuted:
1. A fluted point tipped a wooden shaft; the projectile, when throw, doubt-
less buried itself deep in the flesh of the target. Under such conditions it is
unlikely that the flute would have much effect on bleeding.
2. Many points were fluted only near the base, where the flute was mostly
or entirely covered by the wooden shaft and the binding.
3. The flute is longest and most pronounced on certain long, narrow points
with very straight edges (Fig. 1, E, F, G). These points are less likely to in-
crease bleeding than almost any other style of point. A wide blade with jagged
edges would be most likely to produce copious bleeding.
4. Many artifacts other than projectile points are neatly fluted on both sides.
These include knives, scrapers, drills, and other tools (Fig. 1, H). Such fluting
was not done accidentally, and the implements are not fluted points reworked by
later peoples. The existence of fluted scrapers, drills, etc., rules out the blood
It has been argued that the fluting was done to make the points lighter. This
is questionable, for the following reasons:
1. Probably a little more weight, not less, toward the head of the projectile
would make for greater accuracy and penetration.
2. Granting the desirability of lightening a projectile toward the head, why
not chip a little from the base of the blade, or the tip, or the edges? Simpler yet,
why not scrape a few slivers of wood from the anterior end of the shaft? The re-
moval of a long spall from the face of the blade is a remarkable procedure to go
through merely to lighten a projectile toward the head.
3. In most fluted points a very light spall was removed. The difference in
weight occasioned by this removal would scarcely be perceptible when the point
was mounted in a shaft several feet long, even granting that primitive people are
very sensitive to the "feel" of their weapons.
4. As mentioned previously, many fluted artifacts are not projectile points.
It is improbable that fluting was done to lighten scrapers, drills, awls, etc.
5. Fluted points are very variable in weight. The variation is brought about
by differences in size and material. The length ranges from less than an inch to
nearly seven inches. Materials include several forms of chalcedony, jasper,
flint, chert, quartzite, and rock crystal. Points of almost identical size may
differ considerably in weight.
Lastly, it has been suggested (verbally if not in print) that the flute was
packed with poison, which hastened the demise of the game animals hunted by
the Paleo-Indian. Several facts militate against this idea:
1. Any poison viscous enough to be packed into the flute probably could as
well be smeared onto any point or about the binding of the point.
2. Wide, serrated points would introduce poison into the blood stream much
better than the narrow, straight-edged fluted points.
3. Suitable blood poisons are probably lacking in many areas where fluted
points occur. Plant poisons are rarely hemotoxic, and animal venoms rapidly
lose their toxicity upon exposure to air.
4. In any event, fluted drills, awls, and scrapers were not poisoned.
Current theories do not adequately account for fluted points. A new explan-
ation of these artifacts is offered herein. This explanation is not termed a theory,
for it is not a surmise but an account of a method whereby the author has made
dozens of fluted points which are indistinguishable from those of the Palen- In-
dian so far as workmanship goes.
In general, chipped stone implements can be made easily. Four things are
needed by a would-be chipper:
1. A thin, flat piece of stone. The material should be flint, chert, or some
other substance that breaks with a conchoidal fracture.
2. Something in which the piece of stone may be held rigidly. Several primi-
tive vises have been figured by Moorehead (1910), and others are readily con-
ceived. Lacking any of these, the human hands will suffice.
3. A chipping tool, roughly awl-shaped. Bone or horn tools are excellent, al-
though the modern chipper may prefer an iron punch, which does not need resharp-
ening so frequently.
4. Some sort of hammer or mallet. Lacking this, a stone may be held in the
The above items will be referred to hereafter as the blank, the vise, the chip-
ping tool, and the mallet.
Given these four items, chipping is simple. The blank is held in the vise.
The point of the chipping tool is placed on or nearly on the edge of the blank,
and the butt of the tool is dealt a smart rap with the mallet. A chip is thrown off
the blank, leaving a small depression on that side of the blank away from the
chipper. The size and shape of the chip thrown off will vary with the angle at
which the chipping tool is held, the strength and sharpness of the blow dealt
with the mallet, and the point on which the chipping tool was placed (whether
on or near the edge of the blank). The blank is held in various positions, and
the chipper shapes it by the removal of flakes, working always from the edge.
With a little practice, a twelve-year old child can turn out points superior in
in workmanship to most of those found on field sites.
Let us consider one of the items needed for successful chipping the thin,
flat piece of stone. Where does one obtain such a blank? Most modern chippers
find them on old Indian sites, along with reworkable fragments of broken arti-
facts. But suppose one must procure blanks from the rocks of the field. Usually
one can pound a boulder into pebbles without obtaining more than one or two
flat, thin flakes in the process. Mostly one obtains worthless chunks, and a few
large, heavy flakes which can be chipped only into rude-looking, thick projectile
Many prehistoric peoples made just such artifacts, working the original or
"primary" flake into an implement. Other peoples, among them many of the later
Fig. 1. A-G, projectile points described in the text. H, fluted implements other than
projectile points. Scale as shown.
p ^" ^
Fig. 2. A-F, stages in the manufacture of a fluted point (somewhat diagrammatic).
G-&, fluted points made by author. I, fluted points unlike those of the Paleo-
Indian. Scale as shown.
Indian tribes, used a more specialized technique. They took a large chunk of
stone, and, by skilfully placed blows, removed from it a number of flat, thin
flakes. These "secondary" flakes were easily chipped into fine implements.
Cores from which secondary flakes were struck are commonly found. Some cores
are astonishingly regular, indicating a high degree of skill in the art of second-
ary flaking. The manufacture of secondary flakes is a specialized, advanced
technique generally associated with later Indian cultures.
It is possible to make a neat, thin point without recourse to secondary flak-
ing. First a primary flake is knocked from a boulder. Usually such a flake is
roughly wedge-shaped (Fig. 2, A). This blank is held firmly in the vise. The
chipping tool is held almost vertically, and sharp raps of the mallet cause a
series of large, "deep" chips to be thrown off (Fig. 2, B). The process is re-
peated from the other edge (not face) of the blade. It is seldom possible to make
the chipping from opposite sides meet, and so a ridge is left down the midline
of the point (Fig. 2, C). The entire process is now repeated on the other face
of the blade. At this stage the projectile point is ridged in cross-section. Next
the point is turned up on end, and the central part of the base-to-be is chipped
down a bit, so that the point of the chipping tool can be seated on the end of the
ridge (Fig. 2, D). The tool is held almost vertically, and a single sharp blow of
the mallet usually causes the entire ridge to flake off. The ridge is then removed
from the other face of the blade in similar fashion. The projectile point is now
satisfactorily thinned, it is fluted, and incidentally it has a concave base (Fig.
2, E). Some fine retouching here and there completes the point (Fig. 2, F).
Occasionaly the blow intended to remove the ridge actually splits off an edge
of it. In such a case, one or two additional blows are necessary to remove the
entire ridge. The resultant point will then bear two or three adjacent flutes, pos-
sibly all of unequal length, as in (Fig. 1, D).
As mentioned previously, most usable blanks are roughly wedge-shaped.
Some are thick only toward one end, and from these short-fluted points are made.
Others are thick along most of their length, and would be worked into long-fluted
Some blanks are flat on one face, and therefore need fluting only on the
other face. An occasional primary flake is quite thin and flat, and consequently
needs no fluting at all.
The chipping from one edge of the blank can sometimes be made to meet
that from the other edge, especially if the blank is narrow and of a material such
as chalcedony, which chips easily. In this case the blank becomes somewhat
diamond-shaped in cross-section. Such a blank can readilybe fluted even though
there is no well-defined ridge present on either side. The long spalls struck off
will show transverse chipping.
The author examined an interesting fluted point from Kentucky. It had nearly
been completed. The ridge had been struck from one side, but the resultant flute
had bitten too deeply and obliquely into the edge of the blade, rendering it unfit
for use. Probably the point had then been thrown away. The other face of the
blade still bore a ridge, extending from the base about half way to the tip. The
existence of such artifacts suggests that the Paleo-Indian used the same fluting
technique as that developed by the author. Unfinished points should be watched
for carefully on Paleo-Indian sites. In this connection it is interesting to note
that McCary et al. (1949), reporting on a "Folsom workshop" in Virginia, found
implements which they characterized as follows: "The base is not concave Jut
'wavy,' thus proving the previous existence of a platform from which the longi-
tudinal flakes were struck."
Fluting is a simple, logical method of thinning a projectile point or other
artifact. It is a technique involved in making, not in using, chipped stone imple-
ments. It would be of great importance to a primitive people who lacked the
much more specialized art of producing secondary flakes, but who desired thinner,
neater, more symmetric blades than could ever be chipped from primary flakes.
The technique of fluting is not at all difficult to one has had a little prac-
tice in the chipping of stone. It is so simple, and so readily discovered, that
it should have been hit on, or resorted to at least casually, by peoples other
than the Paleo-Indian. There is some indication that this was indeed the case.
Fluted projectile points unlike those of the Paleo-Indian are occasionally found
(Fig. 2, I); and, as noted previously, fluted drills, awls, knives, and scrapers
Quite a few unscrupulous individuals today are manufacturing and selling
fluted points. This fact is perhaps better known to relic dealers and collectors
than to professional archeologists. An anonymous article in the Tennessee
Archeologist (1950) gave an account of a man in Colorado who makes great
numbers of fine fluted points, "ages" them chemically or mechanically, and
sells them at high prices. A few years ago, the present author examined about
200 perfect, long-fluted points which were offered for sale by various dealers.
Fully 90 per cent of these points could be recognized as fakes.
In 1947 the author began experiments in flint chipping, with the express in-
tention of discovering how and why fluted points were made. After a day's prac-
tice, the explanation of fluting was evident. After two days' practice the author
attempted to make a fluted point. The result (Fig. 2, G) was not inferior to many
western Folsom points. Since then he has made a great number of chipped stone
implements, among them about 150 fluted points. One of these, made in 1949, is
illustrated (Fig. 2, H).
The author has made fluted points using only primitive tools. A simple wood-
en vise (two sticks bound together and then wedged apart) is actually superior to
a modern iron one, which may break the implement. Bone or horn tools chip stone
as readily as an iron punch, but need resharpening frequently. Chalcedony and
jasper are easiest to work into fluted points, and flint is nearly as satisfactory.
Chert is a bit harder to work, but is usable. Rock crystal and quartzite can be
thinned fairly well by fluting, but milky quartz can not. Obsidian is surprisingly
unsatisfactory. Fine points can be made from the thick glass bottoms of reagent
bottles. A fluted point can be made in 45 minutes or less. Contrary to popular
belief, it is most unusual to break a fluted point during the process of manu-
The accompanying drawings were made by the author from artifacts in his
collection. Unless specifically stated in the text, these artifacts are genuine
ones, definitely known to have come from field sites.
Previously advanced theories do not adequately account for fluted points.
Fluting is a technique involved in making, not using, artifacts. It is a simple
method of thinning a blank which otherwise could have been worked only into a
rude, thick point. Fluting would have been an important technique to peoples
who lacked the more specialized art of striking secondary flakes from a core.
1950. "Folsom Fakes," Tennessee Archeologist, vol. 5, no. 1, p. 6
Moorehead, Warren K.
1910. The Stone Age in North America. 2 vols. Boston.
McCary, B. C., J. C. Smith, and C. E. Gilliam
1949. "A Folsom Workshop Site on the Williamson Farm, Dinwiddie
County, Virginia," Quarterly Bulletin, Archeological Society
of Virginia, vol. 4, no. 2, unpaged.
Shetrone, Henry C.
1936. "The Folsom Phenomena as Seen from Ohio," Ohio State Arche-
ological and Historical Quarterly, vol. 45, no. 3.
Wormington, H. M.
1939. "Ancient Man in North America," Colorado Museum of Natural
History, Popular Series No. 4.
Ross Allen's Reptile Institute
Silver Springs, Florida
SOME PROBLEMS IN THE PRACTICAL APPLICATION
A. K. Bullen
That there are differences in body build no one will deny; how finely we
can distinguish between "types" and just where these lines should be drawn,
however, seems to be a moot question. The initial three types expounded since
Hippocrates seems too few; Sheldon's seventy-six types may be too many. It
would appear that the fineness of description and the features emphasized should
be determined by the ultimate use of the observations. (See Sheldon, 1940, pp.
64-66.) If detail is required, by all means it should be collected. If broad trends
are all that is needed for analysis or correlation, a mass of detail is superfluous.
Studies by the author over the past years have included somatotypes obtain-
ed by various methods. Initial investigations used somatotypes already assigned
by Sheldon and Stevens on a series of adolescent boys (Bullen, 1945). A dif-
ferent approach was used in the analysis of a group of college women (Bullen
and Hardy, 1946). A revised version of this method was used in later studies
at the Fatigue Laboratory, Harvard University. Male and female laboratory sub-
jects were available for extensive research carried on over several years. After
experience with somatotyping on the living in the laboratory, less detailed ap-
praisals were made of office and factory groups of men and women.
Methods used for the different studies emphasized advantages and disad-
vantages of various ways of arriving at a final somatotype rating. The adoles-
cent group was appraised by Sheldon, Stevens, and an assistant who used the
method described in Sheldon's book except that an allowance of two spaces was
made for the assumed difference of the index of linearity in the adolescent as
compared with the college group from which the tables in Sheldon's book were
compiled. These ratings were subjective choices within the range of types
listed on the "appropriate" lines of a table (Sheldon, 1940, p. 267). Several
difficulties appear with this method. However, the main problem still remains
one of standards and the relation of the adolescent to other groups.
In situations like this, either a relative or an absolute somatotype might
be assigned. By relative, we mean here the comparison of adolescents with
other adolescents. By absolute, we mean the amount of development of a
component as compared with the total possible range for all ages and both
Although many ways occur to the investigator which might sharpen the
accuracy of somatotyping or make it less open to misinterpretation, neverthe-
less, the general findings of the adolescent study remain essentially valid for
the following reasons. First, the study was one of comparisons of adolescents
with each other. By the definition used in this paper, it was a relative, not an
absolute, study. Secondly, except for individual study of a particular boy, find-
ings of necessity had to be grouped in some way. For correlation purposes of a
moderate or even fair-sized group, seventy-six categories are far too many for
practical purposes. Therefore, while the original appraisals may have "strained
at gnats," the final correlation tossed many borderline cases back into a com-
In study of college women (Bullen and Hardy, 1946), we aimed at getting
away from the controversies of a rating system. Also we found Sheldon's tables,
which are based on the index of linearity, unreliable for women. The index of
linearity trends were not identical for men and women although, of course, ex-
tremes were not found interchanged in type. The problem of standards was even
more difficult than in the case of the adolescent boys. For these reasons, we
chose specific observable points which we checked as present, absent, or in
recognizable combination of two trends.
These points were taken from Sheldon's original criteria. Checks were made
on the theory of possible dominance of one or two of the three trends, endo,
meso, and ecto. The predominance or combination of tendencies for 105 specific
points represented the major emphasis of the individual.'It appeared that a check
of observable points, with no problem of a rating continuum, could be used for
comparisons of incidences between any age or sex groups. For this reason,
while all these data were not at hand, the method lent itself to more universal
application than those based on relative ratings for individuals within a given
age or sex group. Also it could be applied before all groups had been evaluated
on a rating scale.
The theory that more traits would be observable when the component was
more dominant worked out in checking. The check list of points is presented in
the report of the study (Bullen and Hardy, 1946). Where a feature was not clearly
present or where two tendencies did not result in a clear combination of the
two, no check was made. A large number of "average" totals resulted.
As with the adolescent findings, the final results were far more trustworthy
than the minute deviation of "one" between somatotypes. Study of ten extremes
for each tendency endo, meso, ecto, and average provided interesting com-
parisons in the incidence of criteria. Analysis of these incidences clarified
some controversial points which had been used as masculinity-femininity in-
dicators in some studies of gynandromorphy.
In the Fatigue Laboratory studies on the living, research was done on both
men and women and people of different ages. Findings provided data on specific
points which are of interest as to different sex and age groups. Presence of com-
bined tendencies was easier to check on the living than from pictures. Here
again for practical application and theorizing, we used a rough mesh in the
last analysis, thereby minimizing the importance of minute borderline decisions.
Industrial applications of somatotyping and study of vocational incidence
of certain body builds brought us experience of somatotypin g in the field. Here
experience with the general descriptive characteristics of different areas of the
body was invaluable as all clues had to be used. It was not feasible to expect
civilian occupational groups to disrobe at the wish of the investigator. In a
medical, military, or educational setting, this might be possible; then any rough-
er appraisals would be considered slip-shod practice. However, recent observ-
ations on big industrial groups, and individual appraisals of groups of 150 work-
ers suggest that trends can be obtained by less detailed methods which show in-
teresting and "sensible" distributions.
Of course, this presupposes a trained observer of wide experience and famil-
iarity with the observational criteria. For this work, the initial refinement of the
76 types was entirely omitted and people immediately were classed in one Jf
seven broad categories. While Sheldon might not call anything that rough "so-
matotyping," we cannot ignore the fact that in the final analysis much reported
data is "put back" into these or even fewer inclusive groups. The problem
resolves itself, as mentioned above, into what use is to be made of the data.
If it is a detailed study of one individual, the rough method is obviously unde-
sirable. If, on the other hand, the study is one of group tendencies, over-refine-
ment in the initial observations may be uncalled-for; and from a practical point
of view very often impossible.
The following seven trends could be estimated with some degree of confi-
dence. We tend to "describe" people to this extent even in conversational
descriptions or literature. The seven types were: Predominant Endo, Predom-
inant Meso, Predominant Ecto, Meso-Endo, Meso-Ecto, Ecto-Endo, and Moderate
Mixtures. Decisions were not difficult, as predominant types were immediately
assigned to endo, meso, or ecto categories. Where decision of dominance was
difficult, but two trends were clearly present with the third virtually lacking, the
individual was assigned to the appropriate meso-endo, meso-ecto, or ecto-endo
groups. Where no clear dominance or combination of two trends was present, the
person was considered a moderate mixture, that is, a middle range combination of
the three major tendencies. Where dysplasia or gynandromorphy was obviously
present, the fact was noted.
The seven categories listed above appeared sufficient to reveal trends of
correlation with occupation or temperamental tendencies. Specific findings as to
distribution and correlations with the seven types will be presented in two forth-
coming publications. Significant areas for more detailed research will emerge
from this sort of widespread inclusion of a body build estimate in conjunction
with other physiological, psychological, or sociological studies. We should not
hesitate to use the tools now at hand. When better ones are devised, we shall
then be ready to use them to the best advantage in the most fruitful areas of
Bullen, A. K.
1945. "A Cross-Cultural Approach to the Problem of Stuttering,"
Child Development, vol. 16, nos. 1-2, pp. 1-88. Washington.
Bullen, A. K. and H. L. Hardy
1946. "Analysis of Body Build Photographs of 175 College Women,"
American Journal of Physical Anthropology, new series, vol. 4,
no. 1, pp. 37-68. Philadelphia.
Sheldon, W. H., S. S. Stevens, and W. B. Tacker
1940. The Varieties of Human Physique. New York.
THE HARBOR KEY SITE, MANATEE COUNTY, FLORIDA
Ripley P. Bullen, Graham R. Reeder, Bonnie Bell, and Blake Whisenant
In Manatee County, between Terra Ceia and Piney Point, lies Bishop Harbor,
separated from Tampa Bay by Harbor Key. Hearing rumors of "scientific" archae-
ological work at Harbor Key during the fall of 1951, the junior authors visited the
key. They did not locate the "diggings" but discovered a flat-topped temple
mound on the eastern side of the key. Their investigations uncovered pottery and
two burials in the lower part of theramp.
This work was called to the attention of the senior author by T. Ralph Robin-
son of Terra Ceia, and he visited the site with Reeder and Mr. Robinson on
January 3, 1952. This article has been written to record the site, previously
unreported, and what information is available about it. It is hoped the other in-
vestigators will also record their data to make the record complete.
The temple mound, built of shells and other village debris, is about twenty
feet high, with steep sides and a flat top, which measures approximately sixty
by twenty feet. A ramp leads downward from the middle of the western side,
pointing a little north of west. Estimates of basal dimensions were not made
because of heavy ground cover.
Remains of what appears to have been an extensive ditch, cut into the
mound many years ago, may be seen beside its southern end. Otherwise the
temple mound appears undamaged, except for small superficial tests dug by
Reeder, Bell, and Whisenant.
These tests uncovered an occasional sherd but a much greater concentra-
tion of pottery was found in holes dug to remove skeletons. It is doubtful the
latter sherds represented burial offerings. The skeletal material came from two
holes about five feet apart, each hole about two by four feet, located at or very
near the lowest part of the ramp. Both burials were, apparently, bundle inter-
ments of middle aged females.
Specimens from the temple mound, combining those collected on January 3
with those found previously by the junior authors, include a slightly flattened,
ball-shaped object, seven-eights by one and one-eighth inches, formed of sandy
limestone by pecking; a piece of sandstone possibly a fragment of a grinding
stone; a detached columella, a Strombus shell hammer, a Venus shell anvil,
three worked Busycon shells, one possibly part of a pick or hammer and another
possibly part of a dipper, and fragments of pottery.
The latter may be divided into an eroded check-stamped, St. Johns Plain
(five specimens), and undecorated sand-tempered sherds (102 specimens).
Temper of sand-tempered sherds is fine but abundant, and some have a laminated
structure. All available rims are pinched inward to a greater or lesser degree.
This pottery, due to an absence of diagnostic specimens, cannot be placed
chronologically with any exactness. It would appear, however, to have closer
affinities with pottery from the relatively late Saftey Harbor period of the Tampa
Bay area than with that from earlier times. Such an affinity would agree with the
association with a temple mound, as such structures are typical features of the
Safety Harbor period.
The lowest part of the ramp continues as a low ridge for the short distance,
but does not join a village midden. After examining the temple mound on January
3, we returned to the boat and went about a half mile further southwest along
Harbor Key. Upon landing again and penetrating the mangroves, we came upon
the southern end of a shell ridge composed of midden debris. This ridge became
wider and higher as we proceeded nowthward, but in no place was more than a
few feet in elevation.
Adjacent to the shell ridge, at a point which we guessed was somewhere
near the place pointed toward by the ramp of the temple mound we found a low
burial mound, built of sand, about forty feet in diameter, This burial mound had
recently been badly pitted. It was nearly completely covered by holes and spoil
piles. Long bones lay on the surface beside some of the pits and fragments of
skulls in others.
No pottery was noted by us at the burial mound but several undecorated,
sand-tempered sherds were found on the shell ridge. These were similar to some
found at the temple'mound.
At Harbor Key we have an Indian village arrangement or plan similar to
others in the Tampa Bay area. The pattern consists of a long shell ridge or
midden, upon which Indians lived, bordering shell fish producing water. Close
by, upon a convenient spot, was the burial mound. At a greater distance was
built the temple mound, with a ramp pointing towards the village area. Usually,
the space between the temple mound and the village area or shell midden is
devoid of cultural remains, suggesting it may have functioned as a plaza.
A possibly important aspect of the Harbor Key site is the lack of any
cultivatable land. At present the temple mound and the shell ridge village
area are surrounded by mangroves, which in turn are surrounded by water. If
the inhabitants practiced agriculture they must have done it elsewhere.
Harbor Key represents the thirteenth site with a temple mound reported
between the Anclote River and Sarasota. The Indian population around Tampa
Bay, supported by a combination of agriculture and shell fish collecting, must
have been rather heavy during the times these mounds were built, the Safety
Harbor Period, circa 1400-1700 A.D.
Gainesville, Palmetto, and Terra Ceia
CONTRIBUTORS TO THIS ISSUE
William H. Sears is an assistant professor in the Department of Anthropology
and Archaeology of the University of Georgia, Athens. Dr. Sears's second major
report on the Kolomoki excavations was reviewed in The Florida Anthropologist
in the November, 1951, issue.
Wilfred T. Neill is Director of the Research Division, Ross Allen's Reptile
Institute, Silver Springs. Although herpetologist by profession, Dr. Neill has
long been interested in American Indians and their cultures, and is the author of
Florida's Seminole Indians, recently published at Silver Springs.
A. K. Bullen, an anthropologist and member of our society, has been a re-
search fellow in the Fatigue Laboratory, Harvard University, and civilian con-
sultant to the U. S. Army.
Ripley P. Bullen is assistant archaeologist of the Florida Board of Parks
and Historic Memorials.
John W. Griffin, who wrote the tribute to J. Clarence Simpson, is archae-
ologist of the Florida Board of Parks and Historic Memorials.
FLORIDA ANTHROPOLOGICAL SOCIETY
President: Frederick W. Sleight, Mount Dora
First Vice President: H. James Gut, Sanford
Second vice President: A. T. Ewell, Tallahassee
Secretary: John W. Griffin, Gainesville
Treasurer: Ripley P. Bullen, Gainesville
Editor: Robert Anderson, Tallahassee
Executive Committeemen: John M. Goggin, Gainesville
Albert C. Holt, Jacksonville
Gilbert L. Voss,.Coral Gables
Membership in the Florida Anthropological Society is open to all who are
interested in its aims. Dues are $3.00 per year. Members receive the Florida
Anthropologist and other publications of the Society. Application should be
made to the Treasurer, who should be addressed also concerning receipt of
General inquiries concerning the Society should be addressed to the
The Secretary and the Treasurer may be addressed at 103 Seagle Building,
Manuscripts should be sent to the Editor, at the Department of Anthropology
and Archaeology, Florida State University, Tallahassee.
Address items for the Newsletter to the President, Box 94, Mount Dora.
FLORIDA ANTHROPOLOGICAL SOCIETY
The Florida Anthropologist
Vol. I, Nos. 1-2, 3-4
Vol. II, Nos. 1-2, 3-4
Vol. III, Nos. 1-2, 3-4
Vol. IV, Nos. 1-2, 3-4
Vol. V, Nos. 1-2
Price of each double number is $1.00 to Members, $1.50 to non-
Nos. 1-16 . . .. . . . 15 cents each
No. 1 "Two Archeological Sites in Brevard County, Florida,"
by Hale G. Smith. 32 pages, 4 plates . . 50 cents
No. 2 "The Safety Harbor Site, Pinellas County, Florida," by
John W. Griffin and Ripley P. Bullen. 42 pages, 4 plates
.. . . . . . . . 50 cents
No. 3 "The Terra Ceia Site, Manatee County, Florida," by Ripley
P. Bullen. 48 pages, 7 plates . . .... 50 cents
All publications may be ordered from the Treasurer:
Ripley P. Bullen
103 Seagle Building