Title: Certain sand mounds of the St. John's river Florida
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00074980/00002
 Material Information
Title: Certain sand mounds of the St. John's river Florida
Series Title: Certain sand mounds of the St. John's river Florida.
Physical Description: 2 v. : ill. maps, ; 36 x 28 cm.
Language: English
Creator: Moore, Clarence B ( Clarence Bloomfield ), 1852-1936
Holmes, William Henry, 1846-1933
Publisher: Levytype Co.
Place of Publication: Philadelphia
Publication Date: 1894
Subject: Mounds -- Florida -- Saint Johns River Region   ( lcsh )
Antiquities -- Saint Johns River Region (Fla.)   ( lcsh )
Antiquities -- Florida   ( lcsh )
Genre: non-fiction   ( marcgt )
General Note: Pt. 1 includes Earthenware of Florida, by W. H. Holmes.
 Record Information
Bibliographic ID: UF00074980
Volume ID: VID00002
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: ltuf - ADB7152
oclc - 01549841
alephbibnum - 000588401
lccn - 02012881

Full Text



Through t4 &"-z~


Mem. q-- 4z-




m^,,.-- 2-


i. I













mound have. now been laid bare, while the remaining portion has been dug into to a
depth of 7 feet. Subsequently, the mound was restored, that so great and historical
a land mark should not pass from sight. The results of the last investigation were
mainly cumulative, for though a goodly tribute was levied in objects of stone,
copper, and earthenware, no new feature of importance was added to the results of
the preceding season.
During the second investigation, our experience as to the locality of objects
was repeated, since on the base their occurrence was comparatively infrequent,
while between the base and the upper seven feet of the mound almost nothing was
met with, the superficial seven feet containing nearly the entire collection made by
us, though all classes of objects, whether of copper, of earthenware,.or of stone,
were represented on or near the base.
In connection with Mt. Royal, it must be borne in mind that the shape of the
mound gives evidence of a much greater height in former times, and the plough in
recent years has contributed its share to the diminution of the altitude. Nothing,
then, found in Mt. Royal may be considered as of other than original deposit.


Human remains, as before, were found in the last stage of decay, and, with
the exception of a few teeth, were not preserved.


Spade-shaped Imiplement.-Six feet from the surface in the S. E. slope of the
mound, at a point about four feet from the base, lay an adult human skull in the
last stage of decay, and apparently unassociated with the remainder of the skeleton.
A few inches distant lay a beautifully polished hatchet of stone, a spade-shaped
implement of stone, a tubular bead of sheet copper and an ornament of the same
material. With the copper, and dyed a bright green by contact with it, were a
number of deciduous molars and incisors plainly having no connection with the
adjacent cranium which, moreover,.had its dental complement.
The implement, a modification of the type known as spade-shaped, was six
inches in length and, unlike the two found by us in Mt. Royal the preceding season,
did not have the handle tapering to a blunt point, while the nicks, or tally marks,
were wanting (Fig. 1).
This type of implement, though found so far north as Ohio, is of more frequent
occurrence in the Southern States. Its discovery in Florida, so far as reported, has
been confined to Mt. Royal.
Polished hatchets.-During the investigation seventy polished hatchets and
chisels-" celts" so called-were taken from the mound, which, with the sixty-seven
found by us the previous year, make a total yield exceeding that from any one
mound of which we have knowledge.
18 JOURN. A. N. S. PHILA., VOL. X.


FIG. 1. Spade-shaped implement, Mt.
Royal. (Full size.)

None exceeded ten inches in length, while
some were not over three or four inches long.
Nearly all forms were represented, including one
much hollowed out on one side like a gouge. A
number were split and broken at the cutting edge,
showing rough usage. In one case marks of
encircling thongs or cords were plainly visible
(Fig. 2).
Arrow and Lance Points.-Nineteen arrow
and lance points, all of chert or of kindred mate-
rial, rewarded our search. No novel types nor
unusual sizes were represented.
Miscellaneous.-At various depths during the
investigation, were a fragment of a boat-shaped
ceremonial (Fig. 3), several sinkers," or pendent
ornaments, one gracefully fashioned from polished
trap rock (Fig. 4), and a number of small sheets
of mica; also the enamel covering of the tooth of
a man-eating shark. The bony portion, which
may have contained a perforation for suspension,
was missing. It is not unlikely that such teeth
served to tip projectile weapons.
In different portions of the mound were bits
of galena, sometimes several in association, the
largest being perhaps two-thirds the size of a
closed fist. On the base, at that point about
thirteen feet below the surface, were two lumps
of the lead sulphide and a cube of the same mate-
rial about "7 of an inch in diameter, perforated
for use as a bead-a bright and attractive orna-
ment when uncorroded. In no other mound of
the St. John's.have we seen galena utilized as an
A number of perforated pearls were met with
in association with small shell beads in various
portions of the mound.


NOTE.- The chemists who made analyses given on pages 34 and 39 of Part I of
this report now state that LEAD was present in the sulphuric acid used by them. These
analyses are, therefore, valueless and are withdrawn, as are remarks on copper on
page 35.


The yield of copper during this second investigation was somewhat disappoint-
ing, for while a considerable number of ornaments and implements rewarded our
search, but little was found differing to any extent from those met with during our
first visit.
Eleven and one-half feet from the surface was an oblong ornament of sheet
copper 2'8 by 3-7 inches. A central perforation was surrounded by a circle of

FIG. 3. Fragment of boat-shaped" ceremonial,
Mt. Royal. (Full size.)

FIG. 2. Celt" with marks of thongs,
Mt. Royal. (Full size.)

FIG. 4. Sinker" or pendent ornament,
Mt. Royal. (Full size.)

projections on one side, indentations on the other. A line of similar design ex-
tended from the circle to each corner. Upon one side were remains of a woven
vegetable material well represented in Fig. 5. With it lay a circular boss of cop-
per 1-6 inches in diameter (Fig. 6).
It is worthy of remark how widespread is the envelopment in bark or veget-
able fabrics of copper in the graves and mounds where its presence has been noted.
We see it of frequent occurrence in Florida; it has been described in Georgia, in



FIG. 5. Ornament of sheet copper with vegetable fabric adhering, Mt. Royal.
(About one and one-quarter size.)

FIG. 6. Ornament of sheet copper, Mt. Royal.
(Full size.)

FIG. 7. Ornament of sheet copper, Mt. Royal.
(Full size.)

FIG. 8. Copper effigy of serpent, Mt. Royal. (-Full size)


FIG. 10. Ornament of sheet copper,
Mt. Royal. (Full size.)

FIG. 9. Ornament of sheet copper, Mt. Royal. (Full size.)

FIG. 11. Ornament of sheet cop-
per, Mt. Royal. (Full size.)

FIG. 12. Bead of sheet copper, Mt. Royal.
(Full size.)

FIG. 14. Ornament of sheet cop-
per, Mt. Royal. (Full size.)

FIG. 13. Bead of sheet copper, Mt. Royal.
(Full size.)


Tennessee, in Ohio, and we read of copper implements sometimes found in Canada
wrapped in the skin of the beaver.1
Figure 7 represents an ornament of sheet copper with concavo-convex central
boss surrounded with beaded ornamentation.
On the base of the mound, apparently unassociated, was an object of copper,
in our opinion a serpent effigy. The curves are somewhat more graceful than
shown in Fig. 8.
An interesting specimen was an oblong sheet of copper 4-5 by 3-2 inches. A
centrally perforated boss about 1"75 inches in diameter was equidistant from the
corners, upon three of which were raised elliptical ornamentations. The fourth
corner was missing through breakage (Fig. 9).
Figure 10 represents an oval ornament of sheet copper with central oval boss
surrounded by double lines of beaded ornamentation.
About 4-5 feet from the surface with fragmentary human remains was an oval
copper boss about 1-7 by 2 inches (Fig. 11). Through the usual central perforation ran
two cords of vegetable material terminating in a knot on the outer, or convex, side
and with free ends on the inner, or concave portion. These cords doubtless served
as means of attachment. Adhering to the copper were remains of some vegetable
material, probably bark, in which it had been wrapped. In association were a few
small beads of shell.
About four feet from the surface, near fragmentary human remains, with a
polished hatchet of stone, was an elliptical bead of sheet copper about 2-7 inches in
length and 1-1 inches maximum diameter (Fig. 12). With it were several small
copper beads and a tubular bead of bone 1 inch in length and "2 of an inch in
This large elliptical bead and a smaller one of the same pattern from another
portion of the mound (Fig. 13) were made each of a single sheet of copper bent or
hammered until the edges overlapped, a more difficult
process than the method prevalent in Ohio, where,
Hi Professor Putnam informs us, similarly shaped beads
are made of two sections, one fitting into the other.
In Fig. 14 we have a small concavo-convex boss
of sheet copper resembling several found in Mt. Royal.
In caved sand and broken into many pieces by the
fall was an object of wood, possibly a pin, beautifully
FIG. 15. Fragmentary object of carved with a twisted design, copper-coated (Fig. 15).
carved wood, copper-coated, Unfortunately, no restoration was possible, little more
Mt. Royal. (Full size.) remaining than the fragments given and the metal.
Eleven feet from the surface was a piercing implement of copper pointed at
either end. Its length was 19 inches of which 9 inches have been taken by us
for purposes of analysis. Its maximum diameter is -2 of an inch. As is so com-

i Fourth Annual Report, Canadian Institute, page 62.


only the case with copper in Mt. Royal, the implement had been wrapped in bark
or some vegetable fabric, traces of which, in the last stage of decay, remained.
Results of analyses of portions of this implement will be found in our note on
About seven feet from the surface in a pocket of deep red sand were three
hatchets of polished stone, a bead of calcite with incomplete perforation, and a
copper piercing implement 11"2 inches in length, similar in shape to the one just
described. The usual decaying vegetable matter was present.
During the, investigation a number of small copper beads of various patterns
were met with, and a number of sheet copper ornaments of somewhat similar pat-
tern to those already described. Not any of the objects of copper found during our
investigation are exact duplicates in size and design as would surely be the case
had they been manufactured in numbers by Europeans for the purpose of sale or

During our second investigation a number of new and interesting types in
earthenware rewarded our labors, though no vessels of any considerable size were
represented. All but two had been perforated as to the base, either before or after
baking, and with but two or three exceptions, all were strictly of the mortuary
type so prevalent in Mt. Royal, though not of such fantastic designs as were noted
in other localities.
In the western slope of the mound, six feet from the surface, lying above two
polished hatchets, were nine vessels of earthenware of which eight were intact
and in actual contact. Above them were the fragments of a bowl of earthenware
with stamped decoration. The entire bottom and portions of the side of this bowl
were wanting. We shall describe these vessels in order.
1. A vessel resembling an inverted truncated cone with a height of 3'5 inches;
a maximum diameter of 2-5 inches; diameter of aperture, 2 inches. The rim is
beveled (Plate XVII, Fig. 1).
2. A vessel somewhat similar in shape, save as to the brim which extends
laterally. Height about 3 inches. Diameter at aperture, from which a rim extends
-3 of an inch, is 2"7 inches. The base diameter is 1-2 inches (Plate XVII, Fig. 2).
3. A dipper-shaped vessel, perforated for suspension near the extremity of
the handle. Length of bowl with handle, 5-5 inches. Depth of bowl about 1"5
inches. Maximum width, 3-2 inches (Plate XVII, Fig. 3).
4. A curious vessel consisting of two hemispherical cups united as shown in
Plate XVII, Fig. 4, each having a diameter at aperture of about 2-8 inches, and a
height of about 1"5 inches. Below the margin of the outer side of each is a perfora-
tion for suspension. This novel type, unfamiliar to experts to whom it has been sub-
mitted, to a certain extent suggests the double shell form of vessel figured by General
Thruston,' who states in a personal communication that, while several of that kind
1 Antiquities of Tennessee, page 312.


have come to his notice, the form figured by us, to his knowledge, is unrepresented
in the pottery of Tennessee.
We believe this specimen to be unique.
5. A graceful urn of classical type, with a height of 3-7 inches, a maximum
diameter of 2-7 inches, and a diameter at aperture of 1-5 inches. It is perforated
for suspension (Plate XVIII, Fig. 1).
6. A semi-ovoid vessel having a height of 2-3 inches, with a maximum
diameter of 2-5 inches. The margin of the mouth presents three equidistant pro-
jections which, continuing down the sides in the form of ridges with incised lines,
meet at the base. The mouth is surrounded by raised and incised decoration as
shown in Plate XVIII, Fig. 2.
7. A bowl with contracted rim, resembling in shape Figure 3, Plate VI, of
Part I, though somewhat larger. Its decoration of parallel lines as shown in that

Z., .M

FIG. 16. Curved tube of earthenware, FIG. 17. Wedge-shaped object of earthenware,
Mt. Royal. (Full size.) Mt. Royal. (Full size.)
figure has, in addition, parallel rows of punctate markings at intervals between
these lines. Double perforations for suspension are on either side of .the rim.
8. An earthenware pot, similar in type to Plate VII, Fig. 1, Part I, though
differing somewhat in size. Single perforations are on either side of the upper
All these vessels of earthenware show perforation as to the base, made previ-
ous to baking.
In different portions of the mound, respectively four and five feet from the
surface, were two tubes of earthenware of about equal dimensions, flaring from the
centre out and encircled at the middle by a raised band. Length, 5'1 inches;
diameter at extremity, 1"7 inches (Plate XVIII, Fig. 3).
Hour-glass shaped tubes of this pattern in stone are not uncommon. They are
figured by C. C. Jones as from Georgia, by Thruston as from Tennessee, and by
Holmes,1 who shows a strikingly similar specimen 5-5 inches in length.

1 Rep. Bur. Eth., 1881-1882, page 454.


We believe the occurrence of this type in earthenware to be hitherto unreported.
Fragments of another tube of earthenware, differing somewhat from the pre-
ceding, were discovered elsewhere in the mound. The restoration is figured, Plate
XVIII, Fig 4.
A curious curved tube of earthenware, having a nearly square section, was
discovered in caved sand (Fig. 16).
Figure 17 represents an unidentified wedge-shaped object of earthenware.
Five feet from the surface was a dish, with perforation made subsequent to
baking. Length, 6 inches; maximum breadth, 4 inches (Plate XIX, Fig. 1).
Figure 2, Plate XIX, represents a round imperforate bowl with the ordinary
stamped decoration. The material is of fairly good quality differing from the
flimsy structure of the mortuary pottery, to which class this vessel evidently does
not belong. Diameter, about 6 inches; height, about 2 6 inches. It came from a
depth of 14 feet.
Not far from the base was an imperforate vessel having a length of about 3-5
inches, and 2-25 inches in breadth. Its depth was 1-5 inches (Plate XIX, Fig. 3).
As shown in Plate XX, Fig. 1, we have a specimen of mortuary pottery, with
base perforation made previous to baking, having a height of 3-25 inches, and a
diameter across projecting rim of 3"5 inches. Its diameter at base is 2-2 inches.
Plate XX, Fig. 2, is a" rudely made pear-shaped vessel, with a height of about 4
inches, and a maximum diameter of 2-5 inches. Its base contains a perforation
contemporary with its manufacture.
The largest vessel found during our second investigation was an almost cylin-
drical jar. It is of dark material upon which considerable polish has been con-
ferred. At opposite sides below the rim are perforations for suspension. The
bottom is entirely wanting. Height, 7-5 inches; diameter at mouth, 4 inches;
diameter at base, 4-5 inches. It lay near the centre of the mound, 10 feet from the
surface (Plate XX, Fig. 3).
Plate XX, Fig. 4, represents a small mortuary vessel with base perforation made
previous to baking. Height, about 3 inches; maximum diameter, about 2 inches.
Fifteen feet from the surface was a bowl with small handle projecting laterally.
The base perforation had resulted from a blow subsequent to manufacture. Height,
about 2-75 inches; diameter, 3 inches (Plate XXI, Fig. 1).
Plate XXI, Fig. 2, represents a bowl with projecting handle, having base perfo-
ration made after completion. The dimensions of this vessel, which was found 12
feet from the surface, are 3-75 inches by 3-25 inches, with a depth of about 1-5 inches.
A number of additional vessels, presenting no points of especial interest, found
during our second investigation, will not be particularly described.


In Part I of this report, after a fairly thorough investigation of Mt. Royal, we
still hesitated to give our conclusions as to the period of its construction, though at
the time we were in a position to speak with tolerable certainty.
19 JOURN. A. N. S. PHILA., VOL. X.


It will be remembered that we have historical record of Mt. Royal nearly one
and one-quarter centuries ago, and at that period live oaks three feet in diameter
grew upon it. While we know that testimony as to age deduced from oaks is
less reliable than that derived from the family of firs, yet few will deny that live
oaks of this size point to a considerable number of years for development, possibly
to a time previous to the Spanish and Huguenot occupation.
We have, however, stronger evidence of antiquity than this. At the present
writing (1894) but a fractional portion of the mound remains untouched, while the
greater portion has been handled twice, and in part four times by men so thor-
oughly trained and stimulated to such vigilance that, in the rebuilding of the mound,
no objects were met with beyond two arrow heads and the handle of a vase.
hn all these weeks of work during which, as we have seen, hundreds of objects
were discovered, not one was brought to light in any way indicating a knowledge
of Europeans. It is simply out of the question to suppose that the rich tribe that
piled up Mt. Royal possessed the knives, the beads, the hawk-bells, and brass-ware
of Europe without placing them in the burial mound with the dead; and, without
hesitation, we positively assert that Mt. Royal and all its contents date from a
period prior to the landing of the Whites.


This mound, two or three hundred yards back of Duval's Landing, on Blue
Creek, was partially investigated by us last year. It must not be confounded with
the mound in the pine woods two miles west, the demolition of which is described
in Part I.
With a considerable force of men the mound at Duva's was attacked on all
sides during portions of three days, and dug through at a level somewhat below
the base.
SAs before, numerous skulls and long bones filled with roots were met with, and
in addition, at a depth to insure originality of deposit, were two polished hatchets of
stone found separately, and in association with each other an arrow head and a drill.
The feature of this mound was the earthenware.
About two feet below the surface was a small undecorated bowl with base
perforation conferred by fracture subsequent to baking.
Three feet from the surface in the southern slope of the mound and at that
point not far distant from the base, near an isolated cranium, lay an imperforate
vessel 1-5 inches high, with a maximum diameter of 3-2 inches. From one side of
the margin projected an animal's head, to serve as a handle (Plate XXI, Fig. 3).
This type, so well known in Missouri and Arkansas, where the ware is supe-
rior to that of Florida, is also a familiar pattern in the stone graves of Tennessee,
where the material is still better. The type, though not unknown in Florida, is of
infrequent occurrence, and serves to illustrate the curious medley met with in the
Peninsula-this pocket of the United States.


In another portion of the mound was a small bowl in fragments, showing base
perforation made previous to completion of the vessel.
Two and one-half feet down, and undoubtedly of original deposit, a matter of
easy determination in this mound since, it will be remembered, a layer of sand
tinged pink with hematite ran through the mound at a point not far below the
surface, was a simply indescribable object of earthenware, for an idea of which we
must refer the reader to Plate XXII. Its height is about 5-25 inches; its maximum
diameter, excluding handles, 5 inches; its base diameter 1-1 inches. Two curious
claws that had projected from the margin were unfortunately severed by the blow
of a spade, and the portion intervening between them and the rim received so
comminuted a fracture that certain restoration was impossible. We are, therefore,
uncertain whether this curious claw-like handle projected vertically or laterally;
and whether, in its original position, it was everted from or turned over the vessel.
We have endeavored in the restoration, which is indicated by broken lines, to
suggest the most likely position.
Of all the mortuary vessels of earthenware of fantastic design, to which no
possible domestic use can be assigned, and to which we have given the name of
"freak" earthenware, this vessel is easily the most curious. It is almost needless
to say to those familiar with this type that a base perforation, made previous to
baking, is present, or rather that the funnel-shaped body is minus a base.
Beneath the layer of cherry-colored sand was a curious bird effigy of earthen-
ware, with spread tail. Length, 6-7 inches; height of body, 2-5 inches; diameter
of body, 2-7 inches (Plate XXIII, Fig. 1).
At a depth to indicate original deposit was a vase of globular shape with pro-
jecting rim. A flat base, with perforation made previous to baking, enabled the
vessel to maintain an upright position. Height, 3-8 inches; diameter across rim,
5-8 inches; diameter of body, 4-8 inches (Plate XXIII, Fig. 2).
From another portion of the mound came an extremely rude imperforate
vessel, thick and undecorated, with inverted rim. Height, 2-5 inches; diameter,
2-4 inches; diameter of aperture, 1-2 inches (Plate XXIV, Fig. 1).
Several vessels of earthenware, presenting no new features as to size, shape, or
decoration, were met with in various parts of the mound, and with the others
illustrated the rule that all customs have exceptions, since certain vessels from
Duval's showed base perforation made previous to, and others after, baking; while
others again were imperforate.


Although we looked vainly for copper in this mound, we think the effigies of
earthenware, the "freak" pottery, and the pink-tinged sand connect it with the
mound in the pine woods two miles distant. In neither of these mounds were any
objects discovered indicating a knowledge of Europeans. We are, therefore, of the
opinion that both were abandoned at a period prior to the Conquest.



This low mound, with disproportionately large base diameter, is referred to in
Part I.
Having acquired by purchase the small frame house that stood in the way the
previous season, we completely demolished the mound, a work of no small difficulty
since matted roots of the scrub palmetto impeded investigation.
The form of burial was of the bunched variety, and neither human remains
nor implements lay at a depth from the surface greater than two feet, or somewhat
over one foot above the base. As a rule, unassociated with, but occasionally accom-
panying human remains, were seven hatchets of polished stone; a handsomely
wrought "sinker" or pendent ornament of the same material, 2-75 inches in length,
rimmed for suspension on one end and having a central encirc-
ling grooved line (Fig. 18); a scraper, so called, of chipped
chert, 3 inches in length and a chisel wrought from the lip of
a marine univalve, probably Strombus, about 3-5 inches in
length, with a maximum breadth of 2-75 inches. This chisel,
with a median longitudinal groove on one side, is the only
ornamented implement of the kind we have met with on the
river, and with those from the Tick Island mound, to which
reference will be made later, is the only occurrence of this
implement in the sand mounds of the river so far as our in-
vestigations extend. We have occasionally met with scrapers
S wrought from Fulguzr in the mounds, but the discovery of
the shell chisel, in our experience, has been limited to the sur-
face of the shell-heaps, and to the body of the shell-heap at
Mulberry Mound.
The mound was almost destitute of pottery, a few small
FIG. 18. "Sinker" of fragments, plain and stamped, alone rewarded our search.
polished stne, e Hitch One cranium from this mound (Academy of Natural
Sciences' Catalogue, No. 1789) will be described by Dr. Allen
in his paper accompanying this Part of our report.


The large mound on Tick Island had been investigated by us upon so
many occasions, as detailed in Part I, that farther exploration was impossible with-
out disturbance to the score of large bearing orange trees, and to dozens of the
smaller sweet and sour varieties of the same species growing upon it. Finally, after
considerable negotiation, permission was obtained, and the mound, surrounded on all
sides by a party of twenty-two men to dig, with the usual overlookers and searchers,
was completely levelled during five days of March, 1894.
We have described in Part I the structure of this interesting mound. Upon


:several occasions small pockets of pinkish sand were found in association with pro-
-ducts of aboriginal art. In the sand mounds of the St. John's it is a rule without
exception, and the only one of which we know, that when in a mound pockets or
layers of sand artificially colored by the aid of red oxide of iron, are met with, in
that mound surely are burials and relics of the handiwork of the aborigines.


As in our previous investigations, human remains were met with in great
numbers in the upper or brown sand layer, on the shell base in the white sand and
in the base to a depth of one foot. Extended search showed none at a greater depth.
Crania.-Two calvarias from the base under unbroken strata where the mound
had almost its maximum height were recovered in fairly good condition. (Academy
of Natural Sciences' Catalogue, Nos. 1794 and 1796.)
Teeth.-In two molars from separate jaws coming from on or near the base
were traces of extensive decay. With the exception of a decayed tooth, discovered
by us in the Thursby mound at a depth to guarantee original deposit, we recall no
dental caries met with by us in the mounds of the St. John's, save in several cases
of superficial interment where the period of burial, and even the race of the de-
ceased were in question.
Pathological Specimens.-Throughout the mound were a number of pathological
specimens showing to a notable degree ostitis and periostitis, and in some cases a
marked curvature.
Humeri.-No effort was made looking to the preservation of humeri other
than those on or in proximity to the base, which, beyond all peradventure, were
,contemporary with the mound.
Of 186 humeri, 73 showed perforation, a percentage of 39-2.
In one well-marked case the opening in the septum between the fosse had a
,diameter of 13 mm., the greatest we have ever met with or seen reported.

Male. Female. Uncertain.
Perforate. Not. Perforate. Not. Perforate. Not.
Rights 17 30 9 6 15 23
Lefts 8 24 15 9 9 21
25 54 24 15 24 44

Measurements in mm.

Average Minimum Maximum Oscillation
Perforate. Diameter. Diameter. Diameter. Exponent.
Male 25 5-7 1-5 13" 2-6
Female 24 5-2 1- 9- 2-1
Uncertain 24 6-7 1-5 10.5 2-4


Of 73 perforations, 41 are from the right side and 32 from the left. The-
excess of percentage of perforation in female humeri, originally remarked by Broca,.
holds good in this case as in Thursby Mound, to which reference will be made later.
In that mound, however, humeri in the uncertain class show perforations exceeding
the imperforate in number, which is not the case here.


Arrow and Lance Points.-Five arrow and lance points were met with during
the last investigation. Of these, four were found separately on the base, and one 8
feet from the surface in the lower, or white sand layer.
Implements ofPolished Stone.-At different points on the shell base were seven
hatchets or chisels of stone, "celts" so called, from two to four inches in length, the-

FIG. 20. "Sinker" or pen-
dent ornament, Tick Is-
land. (Full size.)

FIG. 19. Stone implement, Tick Island. (Full size.)

smallest being roughened on three sides. In caved sand from the upper layer was-
a "celt" 5-5 inches in length. The majority of these implements were of a ruder
type, less smooth, with cutting edge less truly ground than specimens from many
mounds of the river.
On the shell base was found an implement of polished stone somewhat resem-
bling a hammer, rounded and roughened at either end (Fig. 19).
This style of implement, the common type of the pueblos, has not been met-
with elsewhere by us in the river mounds.
Near by lay two polished celts," one in a fragmentary condition.


On the base, in association with human remains, as in fact were all relics from
the bottom of the mound, was a highly polished and carefully made "sinker" or
pendent ornament, rimmed at either end for suspension, having a length of about
-25 inches and a maximum diameter of 1-3 inches (Fig 20).
About 15 inches from the surface, grooved on one side as by the sharpening of
pointed tools, was a mass of stone, green in color, weighing 5-25 pounds. A por-

FIG. 21.

FIG. 23.

FIG. 22.
"Sinkers of shell, Tick Island. (Full size.)

tion of this specimen submitted to Dr. E. Goldsmith, of the Academy of Natural
Sciences, was determined as follows: I have to report that it is a fragmental rock,
and has nothing in common with green-stone or serpentine. This green rock has
frequently been termed jasper, merely on account of its hardness and finely granu-
lar texture; but, in reality, it is a felsitic tuff cemented together by silica (when
in solution) which afterwards took, by metamorphic change, a crystalline structure."


In no mound of the St. John's have we found aboriginal art in shell so numer-
ously represented as in the sand mound at Tick Island. In immediate association,
ten feet from the surface, on the shell base, were four "celts" of shell ranging in
length from 2-3 to 3-8 inches. A portion of the largest was missing. With these


were three spindle-shaped "sinkers" rimmed at the end for suspension, and four-
beads of shell from 1 to 1-3 inches in length.
Separately on the base were five "celts" of shell from 2-5.to 6 inches long,.
five shell "sinkers" from 3 to 4-5 inches in length; one shell bead with a maximum
diameter of 1-5 inches, and the columella of a marine univalve worked at the ex-
tremities and smoothed at the junction of the whorls (Fig. 24).

FIG. 24. Columella of marine univalve, Tick Island. (Full size.)
In association on the base, in one of the rare pockets of red sand met with in
the mound, were five shell beads ranging in maximum diameter from about -3 to,
-75 of an inch.
Three feet from the surface was a gouge wrought from the shell of Fulgur.
Five drinking cups made from Fulgur perversum were met with during the
investigation. One, imperforate, lay five feet from the surface; the others, from or
near the base, showed the usual perforation below the shoulder opposite the aper-
ture. One of these cups lay within a larger specimen.
An elliptical ornament of shell 3-5 inches in length was found on the base,.
and a shell gouge was exhumed 3 feet from the surface.
For fuller details as to many of these interesting types, the reader is referred'
to Professor Holmes' exhaustive paper "Art in Shell," Report of Bureau of Eth-
nology, 1880-1881.

On the base of the mound was an implement of bone, pointed at one end,
having the articular portion at the other.


The discovery of copper in the Tick Island Mound was a matter of surprise,
and congratulation, after much repeated and unsuccessful search.
Six inches from the surface, with human remains, was a centrally perforated:
disc of copper 3-2 inches in diameter, encircled near the margin by a series of small
indentations forming projections on the other side..
This ornament is shown in Fig. 25 with section removed for analysis, the
result of which will be found included in the Note on copper in the latter part of
this Report.


In caved sand was a tubular copper bead with overlapping edge, 1-2 inches in
At the base of the eastern slope of the mound, at a point about four feet
from the surface, was a part of a lower jaw of an unidentified mammal, copper-
coated on one side, represented by the ramus and a portion of the body showing the

FIG. 25. Copper disc, Tick Island. (Full size.)

FIG. 26. Portion of jaw of mammal, copper- FIG. 27. Portion of jaw of mammal, side
coated side, Tick Island. (Full size.) showing bone, Tick Island. (Full size.)

sockets of two molars. The work was neatly done, and at one place showed
.piecing on of extra copper where the original sheet had failed to afford sufficient
material. A part of the copper sheathing extending beyond the bone, indicated a
loss of the enclosed material through the ravages of decay. Figs. 26 and 27 show
20 JOURN. A. N. S. PHILA., VOL. X.


both sides of this curious object, which will recall to the reader the copper-coated
jaw and portion of the skull of a gray fox found in Mt. Royal, in which case, how-
ever, the bone and teeth were entirely included by the metal.
Almost in the immediate center of the base of the mound, on the shell, 12 feet
from the surface, in association with a "celt" of shell and human remains, was an
object of copper, closely resembling, in shape, the carapace of a small turtle. Its

FIG. 28. Copper effigy of turtle shell, Tick Island. (Full size.)

length was 2-8 inches, its maximum diameter 2-2 inches, its height 1-2 inches (Fig.
28). Adherent were remains of some vegetable fabric. This ornament, for such it
undoubtedly was, had been constructed by placing, one above the other, two sheets

FIG. 29. Tobacco pipe of gritty ware, Tick Island.
(Full size.)

FIG. 30. Tobacco pipe of gritty ware, Tick
Island. (Full size,)

of copper of similar shape, which shape had probably been conferred by hammer-
ing or pressing them over the upper shell of a turtle. At all events, such is
believed to have been the method in Ohio, where the scales and minor features of


the shell are found in relief on the copper. In this case, excessive deposit of car-
bonate prevents exact determination. That these turtle shells of copper, in this
instance, together formed one ornament was shown by the fact that double perfora-
tions at the sides and ends were coincident, which would not be the case had the
piercing in each object been accomplished prior to their union. On the outer
surface of two of these perforations were knots of vegetable material, which had
doubtless served, in connection with cords, for purposes of
attachment or suspension.
When we recall the area covered by the Tick Island
lt mound in connection with its height, the discovery of but
kiJ three objects of copper other than superficially would indi-
4 1 c cate a high degree of rarity for this metal at the period of
Si construction of the mound.

,- .
FIG. 31. Pin of coral, Tick
Island. (Full size.)

Professor Putnam
come under his notice.


In caved sand from the upper stratum, at that point
many feet in thickness, from the same portion of the mound,
though not in immediate association, were two tobacco pipes
of gritty ware of high specific gravity (Fig. 29 and 30).
This type in earthenware is well known in the West. We
have seen an almost exact fac-simile from Indiana.
The great rarity of tobacco pipes in the mounds of the
St. John's has been commented upon before. Considering
the intimate relations between the smoker and his pipe, re-
lations which would undoubtedly be expected to continue in
another world, we are at a loss to account for the scarcity of
tobacco pipes in the river mounds or, for that matter, in the
State of Florida.

During previous investigations several masses of coral
were met with in the mound. One piece found by us evi-
dently formed part of a pin, used, doubtless, for ornamental
purposes. The head, about 1-9 inches in length and 1-25
inches in breadth, remained. Attached to it was a small
portion of the shank (Fig. 31).
informs us that this is the first example of worked coral to


As during previous excavations, few vessels of earthenware were encountered.
As before, fragments of earthenware, to many of which the outline of projec-
tile points had been given, lay with the bodies. It is worthy of note that arrow


heads described as of sheet copper (probably brass) are found in the graves of post-
Columbian Canadian Indians.1 This thin and flexible material could have served no
practical purpose, and. was doubtless put in as a makeshift, as were the earthen-
ware arrow heads of some of the river mounds.
It is our belief, based upon careful supervision of ourself and a number of ex-
perienced assistants, that during the demolition of the mound no sherds of stamped
decoration were met with below the immediately superficial portion.
In the upper stratum were several bowls, perforated as to the base subsequent
to completion, presenting no features of interest.
On the base were various sherds, two showing decoration with crimson pigment.
The bottoms of two small vessels were met with, also on the shell, each with four
small knobs to serve as a means of support, enabling the vessel to maintain an
upright position. This unusual addition to mound earthenware was met with
during our earlier investigations at Tick Island, and will again be referred to in
connection with the mound at Racey Point.
In the white sand layer, 8 feet from the surface, was a boatshaped imperforate
vessel. Unlike all pottery found upon the base, it was apparently unassociated
with human remains. Approximate measurements are,-length 5-2 inches, maximum
width 2-2 inches, depth 1-3 inches (Plate XXIV, Fig. 2).
On the base of the northern slope of the mound, 6 feet from the surface, with
human remains, was a rudely made but spirited earthenware effigy of a duck-
probably a male wood-duck, a variety found in the neighborhood. The length of
this curious effigy, from end of bill to tip of tail, is 9 inches, its breadth 3-5 inches
(Plate XXIV, Fig. 3).
Imbedded in the base of shell, 10-5 feet from the surface, was a vase with six
upright flutings around the body. The bottom, which had been knocked out, lay
with it, and is included in the restoration given in Plate XXV, Fig. 1. Height 4-5
inches; diameter at mouth, including flaring rim, 4-4 inches.
Three and one-half feet down was an imperforate vessel resembling the bowl
of a spoon, with perpendicularly projecting handle, terminating in an effigy of an
animal head. Length about 3 inches; breadth about 2 inches; height, exclusive
of handle, -8 of an inch (Plate XXV, Fig. 2).
Two globular vessels lay in association 10 feet from the surface. Both showed
base perforation subsequent to completion. They were undecorated. Approximate
measurements of the larger are, maximum diameter 7-5 inches; diameter of aper-

S" These arrow-heads are rudely and carelessly cut from portions of European copper kettles. They
must have been quite useless for shooting, and seem to have been made simply as substitutes for flints
for burial purposes. The large number of European articles found in some of the graves renders it toler-
ably clear that although those who made the graves retained the old custom of depositing objects of vari-
ous kinds with the bodies, they had lost the art of making chipped arrow points. Here we seem to have
an overlapping of the old and the new order of things, as in many other places where specimens of white
manufacture are buried in accordance with the traditional customs of the Indians. In this case arrows
were probably regarded as a necessity, and the ghosts of these thin copper specimens were thought suffi-
cient for spiritual uses."-Report Canadian Institute, Session 1891, page 51.


ture 5-5 inches; height 4-2 inches; of the smaller, maximum diameter 5-5 inches;
diameter of aperture 4-5 inches; height 2-7 inches.
Near the base was a portion of a bird effigy of earthenware representing the
forepart of the body with the head and neck extending vertically. The intended
likeness is not apparent.
It will be noted that no "freak" pottery was met with in the Tick Island
mound, nor, with one exception-a small fragmentary vessel found superficially-
was any perforation discovered made previous to baking.
One point of great interest in connection with the earthenware of the Tick
Island mound must be noticed here. As we have stated, the mound of sand is
constructed upon a base of shell. Whether this shell base was a shell-heap pre-
existing at this place, utilized for the construction of the mound, as was the case at
Thursby Mound, at Ginn's Grove, at Thornhill Lake, and at the Indian Fields, we
are unable to decide, though indications point to a negative conclusion, based upon
the facts that the shell does not extend beyond the mound, as is the case at other
localities, and that a long causeway of shell running through swampy ground con-
nects the shell base with higher territory farther back. In any event, this mass of
shell, which at the centre of the mound attained a height of five feet, whether a
pre-existing shell-heap on the spot or a mass of shell-heap material brought from
elsewhere, preceded the period of the construction of the sand mound. In the
shell was fibre-tempered ware, totally dissimilar from sherds found in the sandy
portion of the mound, or in any sand mound of the river, though its presence is
noted in the neighboring shell ridges. We shall allude to this again in Notes and

The researches of Professor Wyman failed to discover canine remains in the
shell heaps of the St. John's. The jaw of a dog found by us in a shell heap of the
upper river is figured and described by Professor Cope in our paper in the Naturalist
for July 1893. Subsequently, another canine jaw of undetermined variety was
exhumed by us from another shell-heap.
In the shell base of the Tick Island mound were skeletal remains as to which
Professor Cope writes: "The bones you send are those of a dog, but of what species
I- am not sure. It is not wolf nor coyote, but differs from ordinary breeds of
domestic dogs. Nevertheless, it may be- some form domesticated by the Indians
with which I am not familiar."

On the central portion of the base, in immediate asso-
ciation, were two flat pieces of coquina, a pebble hammer, a
portion of a columella of a large marine univalve, rounded at
FIG. 32. Shark's tooth either end, and the tooth of a shark, perforated at the base as
with perforation, Tick
Island. (Full size.) for suspension (Fig. 32).



Of all the important mounds bordering the St. John's or its tributary streams,
that on Tick Island was most surely free from all previous investigation, inspired
either by scientific motives or by hope of treasure. In swampy ground, hidden by
towering palmettos, on an island usually uninhabited and miles from every line of
travel, the existence of the mound was known to few. With trivial exceptions, in
all probability, the entire contents of this tumulus are in the cases of the Peabody
Museum and of the Academy of Natural Sciences. Among all the objects on dis-
play there is not one in any way indicating or suggesting European influence.
We know Florida to have been the first portion of the Uiited States to expe-
rience European contact, and that Cape Canaveral and other points of its eastern
coast saw many wrecks of vessels bound from the New World to Spain. Indeed,
we are told by the Hugienot LaudonniBre that, in his time (1564), the Indians
acquired various possessions from vessels cast away, and even at this early period,
doubtless, many pro cts of European art had a fairly wide distribution in the
Peninsula. Tick Island isarcely ten leagues from the sea and to a certainty, in
our opinion, had the mound been in process of construction in post-Columbian
times, such objects would, at least to a limited extent, have found their way into
the common burial place of the island.
It is, therefore, our belief that the mound at Tick Island was no longer in use
as a place of interment when white influence was first felt in Florida.


Wishing to learn the entire contents of a mound which, superficially, had
yielded such rich returns on former visits, and having again obtained the cordial
permission of Mrs. L. P. Thursby, of Blue Spring, the owner, we devoted nine days
of January, 1894, with a:force of fourteen men to handle the spades and our usual
complement of five to direct and oversee the work.
The mound was completely demolished.
Thursby Mound, it will be remembered, was situate on an arm of land between
Lake Beresford and the river. Extensive shell deposits on the eastern side of the
river where the mound lay, and the great shell bluff, mounds and ridges of Huntoon
Island immediately opposite, testify to a numerous Indian population in former times.
The mound, as our previous work indicated, was largely composed of white
sand with thin local layers of shell of considerable extent and occasional pockets
of shell, sometimes from one to two feet in thickness. But three or four pockets
of sand, artificially colored red, were met with, thus calling to mind the mound at
Tick Island, which Thursby Mound greatly resembled in nearly every respect.


In the south, southeast, and southwest portions of the mound, beginning at the


SHELL BLUFF NEAR THURSBY MOUND. (From photograph by Author.)


margin of the base and continuing through the central portion, human remains were
constantly met with on and in the shell base. In certain cases, determination
of the form of burial was impossible, owing to the ravages of decay and to the
intermingled condition of the bones, but in the case of the greater number the form
of burial was shown to be in anatomical order, though in various degrees of flexion.
In the white sand, somewhat above the base, but at depths to indicate original
burial, bunched interments were met with. Superficial burials were all in ana-
tomical order.
Crania.-Though the bones in Thursby mound, owing to assimilation of lime
salts from the shell, were in a better condition than those from the down river
mounds, but two crania were saved in a comparatively unbroken condition.
(Academy of Natural Sciences Catalogue, Nos. 1781 and 1782.)
Humeri.-Of 264 humeri from original burials on or near the base, 132, or
exactly 50 per cent., showed perforation.


Male. Female. Uncertain.
Perforated. Not. Perforated. Not. Perforated. Not.
Rights. 19 40 23 7 22 16
Lefts. 22 38 14 8 32 23
41 78 37 15 54 39

Measurements in mm.
Average Minimum Maximum
Perforated. Diameter. Diameter. Diameter. Oscillation Exponent.
Male 41 4-97 -5 9-5 2-19
Female . 37 67 2-5 11-5 1-32
Uncertain 54 6-4 2-" 10*5 2-1

From the above tables it will be seen that of 132 perforations, 64 are from the
right side' and 68 from the left. The excessive percentage of perforations in the
case of female humeri and in those of uncertain sex, which are doubtless mainly of
females of masculine build, is worthy of note. In these two classes also, the size
of the perforation considerably exceeds that noted in male humeri.
Seven superficial humeri contained 3 perforations.
In caved sand, and therefore of undetermined depth, 15 humeri were found.
Of these all but 3 showed perforation, a percentage of 80.
Of 14 humeri of children between one and twelve years of age, not included
in foregoing data, 3 showing perforation were noted; one from a child of about
twelve years of age, with an opening 5 inm. in diameter; and 2 from children about
six years of age, with perforation respectively 5 mm. and 1 mm. in diameter.



Ten "celts" of polished stone were met with, the majority on the base in
the vicinity of human remains. One rude
chipped implement of chert also was found,
while a number of arrow heads of chert,
chert breccia and of chalcedony were singly,
.j as a rule, on the base of the mound.
A :A fragment that had formed part of a
large vessel of steatite was obtained, at a
h,-- 4,e depth denoting original deposit. No vessel,
or fragment of a vessel, of stone has before
t / -' been found by us in the river mounds,
though a piece of considerable size was taken
from the river, not far from Thursby mound.
Near the base, in the southern slope of
the mound, was a lance head of chert 5-35
S""inches in length, with a maximum breadth
"4" of 2"5 inches (Fig. 33).
y'. i. f The unusual scarcity of lance heads in the
river mounds is worthy of remark. During
' all our mound investigations but five have
been met with exceeding 4 inches in length.
In caved sand, and therefore of unde-
Stermined depth, was a highly polished bead
/ '- of red jasper with lateral perforation. Its
length was 1"1 inches, its width 1 inch, its
maximum thickness -65 of an inch. To
those familiar with the character of jasper,
upon whose surface steel fails to make an
V-) impression, the even perforation of this bead
will be a matter of surprise. Its accomplish-
FIG. 33. Lance head of chert, Thursby i t o jse
Moun. 33. Lac e head of chert, Thrs meant was probably rendered possible by the
aid of a revolving reed with sand and water.


Two and one-half feet from the surface, with human bones, was a drinking cup
wrought from Fulgur perversum. A similar one was met with in caved sand.
Both had suffered the usual mutilation by intentional perforation of the base.
Curiously enough, and in marked contrast with other large river mounds, no
shell beads were met with in Thursby mound, except with superficial interments.



At the margin of that portion of the mound where the remarkable deposit of
earthenware effigies and pots was found by us the preceding season, a number of
similar objects-ten or a dozen-were met with, presenting, however, no new types.
About 18 inches from the surface, in the eastern slope of the mound, not far
from the margin of the base, with human remains, was a small animal effigy, similar
in type to those previously found, which, it will be remembered, were on the southern
slope. With this exception, none of these curious effigies were met with in any
other portion of the mound.
Sherds were infrequent, and all, with possibly one or two exceptions, were
undecorated, save in some cases by the use of red pigment.
Under the summit plateau, from 7 to 10 feet from the surface, were a number
of earthenware bowls and vessels of large size, one having a diameter of 21 inches.
They were not in immediate association, and none, even excluding a portion
knocked from the bottom, which was common to all, was complete. The
missing portions were not found in association, and it was evident that the canny
builders of the mound had utilized broken vessels for mortuary purposes. Certain
ones, in fact, were so fragmentary that, presenting no decoration, their retention
was deemed needless.
Among the earthenware vessels from this portion of the mound was a curious
fragment, having a height of about 5 inches and an approximate diameter of 8-25
inches (Plate XXV, Fig. 3).
A circular flaring bowl, having traces of red pigment inside and out, had a
diameter at mouth of 15 inches (Plate XXVI, Fig. 1), and a globular vessel deco-
rated around the margin, shown half size in Plate XXVII, Fig. 3, lay not far distant.
Still another vessel, bell-shaped, had major and minor diameters of 10 and 4
inches, respectively. As both margins exhibit broken surfaces, its original shape
cannot be determined.
In the eastern slope, 2 feet from the surface, with human remains, was some-
what over half of a vessel entirely new in our experience on the St. John's (Plate
XXVI, Fig. 2). Length of fragment, 8 inches; maximum breadth, 8 inches;
height at aperture, 4-7 inches. The material is of superior quality, thin and
thoroughly baked; the punctate decoration carefully stamped. The shape of the
entire vessel must have been ellipsoidal, though with pointed extremities. Through
the upper portion of the remaining end are two perforations, made previous to
baking. While the one below the margin is easily explainable as for purposes of
suspension, it is hard to account for the other situated not far above the base
towards the end of the slope. The care exhibited in the manufacture of this vessel,
its superior quality and conscientious decoration, separate it from the class of mort-
uary "freaks," though, on the other hand, the lower perforation made the retention
of liquids an impossibility, and would seem to unfit it for domestic purposes.
Ten feet from the surface, with human remains, was a curious oblong article
21 JOURN. A. N. S. PHILA., VOL. X.


of earthenware of undetermined attribution. Four legs, two of which are somewhat
imperfect, furnish means of support. The upper surface, having a decoration of
incised lines, is 5 by 3-5 inches. An oval depression, about 2 by 2-5 inches, sur-
rounded by a raised margin, was intended, apparently, as a receptacle of some sort.
One end of this puzzle for archaeologists had, apparently, been trimmed down as for
the removal of a broken portion, possibly a handle (Plate XXVII, Figs. 1 and 2).
When discovered, this object was covered with a fine white powder resembling
wood ashes, wholly unlike the sand of the mound. At places the depressed circle
on the upper surface seems to show traces of fire. We have seen in Japan, articles
suggesting this one, supporting a live ember, around which sat a company of smokers
making use of the fire from time to time to ignite the contents of their pipes. This
suggestion may be taken for what it is worth.
In different portions of the mound, at depths to guarantee original deposit,
were the headless body of a duck of solid earthenware, 10 inches in length, and two
heads, one of which is shown in Fig. 34, evidently belonging to some member of

FIG. 34. Head of duck in earthenware, Thursby Mound. (Full size.)
the duck family. Unfortunately, neither head corresponds to the trunk. They are
of coarse manufacture, and recall to a certain extent the earthenware heads figured
by Mr. S. T. Walker (Smithsonian Report, 1883, page 67) as coming from north-
western Florida.
In caved sand was a tube of earthenware 2-75 inches in length.
Several pots and bowls, undecorated, of medium size were found superficially.

Several "celts" of iron or of steel were found superficially.

9 ?'

BURIAL MOUND AT GINN'S GROVE. (From photograph by Author.)

I: * .: r~-
;~. F.

...~. .

JA "v9~

I ---- ; --




While implements of iron, which certainly, and an ornament of gold and one
of silver,' which possibly, show European contact were found in the vicinity of the
surface, nothing at any depth indicated a knowledge of the arts of the white man.
In our opinion, Thursby Mound belonged to a period prior to the Discovery.


This mound has twice before been examined by us. It was again visited with
a force of eight men to dig, and a careful investigation made, with the kind per-
mission of J. N. Whitner, Esq., the owner. As before, superficial burials were found
and abundance of human remains along the base. Two crania (Academy Cata-
logue, Nos. 1784 and 1788) partly imbedded in the shell base were saved in fairly
good condition.
As on former occasions, relics were conspicuous by their absence, and our hopes
to find additional earthenware of the superior quality formerly yielded by the
mound, were doomed to disappointment.
On the base, with the skeleton of a child, was a bead of shell about one inch
in diameter.
Beneath the western slope of the mound, on the base, at that point about 2*5 feet
from the surface, was a portion of a molar of a mammoth (Elephas primigenius
Americanus). This fossil tooth was probably prized by the Indians as a stone.
Fossil remains have been found by us in Florida, bordering the river and in the
neighboring clay. We have seen remains said to have been brought up in seines.
Teeth of the mastodon and of the mammoth have been found in the mounds of
Two feet down was an undecorated bowl about 6 inches in diameter, in frag-
The mound at Ginn's Grove is the one to which Dr. Brinton assigns a compara-
tively modern origin. During the many days of work included in our three visits,
with many trained assistants in addition to those handling the spades, we have dis-
covered nothing in any way indicating contact with the whites, nor have we noticed
marks of disturbance on the base, and we are therefore of the opinion that Dr.
Brinton's conclusions may be dismissed as, based upon superficial and insufficient

Reference is made in Part I to two mounds at this point. Eight days of Janu-
ary, 1894, were devoted, with a large party of men, to the total demolition of the
larger mound and the excavation of the central portion of the smaller one, in which
part alone burials were met with in that mound.
See Part I.



As during our first visit, human remains were found in anatomical order in the
large mound, no case of the bunched burial being discovered. Including the skele-
tons found during our previous visit, somewhat over fifty were noted throughout
the mound, lying from a depth of 10 feet to within a short distance of the surface.
No crania were saved. Indeed, so great was the pressure of the sand and of the
material from the shell-heaps, which lay between and above the sand strata, that
even long bones were crushed, and one cranium had suffered so greatly that the
opposite sides were in contact, giving it somewhat the shape of a tureen.


Superficially, was a beautifully wrought arrow head. Two fragments of projec-
tile points were met with during the excavation.
Occasionally, with the bodies were small beads of shell, and with several a few
tubular beads of the same material, one inch in length and over.
Five feet from the surface, unassociated, was a portion of a ceremonial object
or gorget. The material was limestone;
S ------......... length about 2-5 inches; breadth 2
inches. On one side it presented a flat
surface, on the other a convex one. It
Swas perforated longitudinally (Fig. 35).
I n the northern slope, in the shell-
i^ heap debris surrounding the sandy por-
I /-'' tion of the mound, 4 feet from the sur-
face, lay the skeleton of a man, pre-
sumably a person of rank. Great quan-
tities of small shell beads lay along the
forearms from wrist to elbow. With them
FIG. 35. Gorget of limestone, Thornhill Lake.
(Full size.) were occasional beads of a polished rock,
which Dr. E. Goldsmith has pronounced
Catlinite. At either wrist was a miniature example of the double-bladed axe in
stone, variously termed gorgets, ceremonials or banner stones. Between the blades
was the usual perforation for suspension, These little gems were each 1-3 inches
in length, with respective maximum breadth of blade of '8 and -9 of an inch (Figs.
36 and 37).
On the breast of the same skeleton with many beads of Catlinite and of
shell, including some of the latter material tubular in form and over one inch in
length, and discoidal beads of shell having a diameter of -7 of an inch, was a
double-bladed axe of stone shown in Fig. 38.
Near the neck of the skeleton, with beads in great profusion, were the dis-
integrated remains of a small stone ceremonial. With them lay a pendent orna-


FIG. 36. Ornament of stone with sections, Thornhill Lake. (Full size.)

FIG. 37. Ornament of stone with sections, Thornhill Lake. (Full size.)

FIG. 38. Ceremonial axe of stone, Thornhill Lake. (Full size.)

\ I l1/u i / FIG. 40. Tooth of shark, Thornhill
I a Lake. (Full size.)

FIG. 39. Pendent ornament of stone, Thornhill Lake.
(Full size.)


ment of polished stone, perforated at either corner of one side, while the opposite
margin curved slightly upward and inward, as seen in Fig. 39.
Andrew E. Douglass, Esq., has called our attention to the fact that this curi-
ous ornament is made from the blade of a double-bladed ceremonial, doubtless ren-
dered useless through fracture. He informs us that a similar ornament was found
by him in a mound on Tomoka Creek, near the east coast, about 36 miles in a
straight line north of Thornhill Lake.
Among the beads on this skeleton was the tooth of a shark with double perfora-
tion at the base, for suspension (Fig. 40).
With the bones were fragments of charcoal.
In the light sand layer, 6 feet from the surface, lay a skeleton badly crushed.
Upon the sternum, with a few small shell beads, was a pendent ornament of stone,
similar to the one described, though somewhat
smaller. As the type is so unusual, we repre-
Ssent this one also in Fig. 41.
Tubular beads of Catlinite, never exceeding
five in number, lay with several skeletons.
The discovery of this stone, whose source
of supply is Minnesota,1 in a Florida, mound
showing absolutely no contact with the Whites,
is of considerable interest, since this rock
is believed by some to have come into use
FIG. 41. Pendent ornament of stone, with only in recent times. We believe the larger
section, Thornhill Lake. (Full size.)
mound at Thornhill Lake, in common with all
the more important mounds of the St. John's, to have been completed in pre-
Columbian times, and are, therefore, of the opinion that Catlinite was in use at a
period earlier than has been supposed.
In the central portion of the smaller mound were seven bodies in anatomical
order. Upon the breasts of three, with shell beads, lay gorgets representing a form
of the double-bladed axe, differing from that found in the larger mound. Two of
these, about the same size, were of phosphate rock, much the worse for age; the
other, probably a soft serpentine, also was in a crumbling condition from the lapse
of time (Figs. 42 and 43).

In certain respects, the mounds at Thornhill Lake stand alone. In no other
of the river mounds has the double-bladed ceremonial been met with by us, and in
no mound of the State, we believe, has it been reported accompanying human
remains. Andrew E. Douglass, Esq., had the good fortune to find in the mound

It is found also in Wisconsin and in Dakota. The reader is referred to an interesting article,
SCatlinite," by Edwin A. Barber, American Naturalist, Vol. XVII, page 754.

170 '



/,X I-

LARGER MOUND, THORNHILL LAKE. (Fr~om photograph by Author.)


near Tomoka Creek, to which reference has been made, eight beautiful ceremonial
implements of this type, some of which he has figured and described.1
Another feature, worthy of remark, is the absence in these mounds of other
implements of stone. One would at least expect hatchets and projectile points.
The most noteworthy point of the entire investigation was the total absence of
vessels of earthenware, whole or fragmentary, and the virtual absence of sherds.
During our first investigation of the larger mound at Thornhill Lake, though
each spadeful of material was closely watched, not one sherd was encountered,
save superficially. During the demolition of the mound the question of presence

FIG. 42. Ceremonial of phosphate rock,
Thornhill Lake. (Full size.)

FIG. 43. Ceremonial of serpentine, Thorn i i
Lake. (Full size.)

or absence of earthenware was carefully kept in mind. About four feet from the
surface were two sherds with the ordinary stamped decoration. We were present
at their discovery, and are convinced as to their position as stated. They were
doubtless of accidental introduction, since the entire mound yielded no other
In the smaller mound no pottery was met with.
This almost total absence of sherds and earthenware, fragmentary or otherwise,
is entirely novel in our investigation of Indian mounds devoted to purposes of
sepulture, and it is evident that, with the makers of the mounds at Thornhill Lake,
the custom of interring earthenware with the dead did not obtain-an unlocked
for departure.

SProceedings of the American Association for the Advancement of Science, Vol. XXXI, Montreal
Meeting, Aug. 1882.



In thick undergrowth, about one-half mile from the river, and about five miles
south of Palatka, is a mound of yellowish sand without stratification. Its height
is 6 feet, its base diameter 47 feet. It had undergone considerable previous inves-
tigation. About three-quarters of the mound were demolished by us with entirely
negative results.

In an arm of land formed by the union of Dunn's Creek with the river, is a
wooded territory, thick with trailing vines and scrub palmetto, forming a portion of
the Acosta grant, dating from Spanish occupation.
The mounds, several hundred yards apart, about equidistant, are possibly one
mile from St. John's Landing on the river and Westonia on Dunn's Creek.


The measured height of this mound, which was totally destroyed, was 9 feet
7 inches, but as it formed the centre of a depression from which its material proba-
bly came, its actual height was about two feet less. At a level to justify this sup-
position, a layer of sand 3 to 6 inches in thickness, showing the effects of fire and
mingled with charcoal, extended through the mound.
The base diameter was 62 feet.
A previous investigator had dug through the summit plateau to the base.
With the exception of a few fragmentary superficial bones, human remains,
represented by two isolated crania, were met with in but two places, at a depth of
5 and 6 feet respectively.
Three arrow heads and a small sheet of mica represented the total yield of
objects of stone.
Sherds were numerous, and of good quality, near the margin of the mound,
some having the complicated stamped decoration of Georgia and Carolina. This is
the farthest point south on the river where its occurrence has been noted.
Near the eastern margin of the mound, on the base, which at this point was
2-5 feet from the surface, unassociated, was a globular bowl of earthenware, bearing
traces of red pigment, with the usual base mutilation. Height, 6 inches; maximum
diameter, 9-25 inches.
Three undecorated bowls of ordinary pattern and of poor material were found
separately in pockets of sand whitened by fire and containing fragments of charcoal.


This mound, with a height of 5"5 feet, and a base diameter of 45 feet, was
virtually demolished. With the exception of a few fragments of human bones and


a small number of sherds, plain or decorated with pigment, our search was unre-

This mound, somewhat oblong in shape, had a base diameter of 35 by 45 feet.
Its height was 5 feet. The mound was entirely leveled. With the exception of
the usual charcoal and two or three sherds, absolutely nothing was encountered.


The shape of this mound, and the fact that it had undergone no previous in-
vestigation, inspired us with hope, especially as it was evident that the roots of
trees had protected its summit against the ravages of time. Its height was 7-5
feet, its base diameter 95 feet. The diameter of its summit plateau was 60 feet.
Twenty-one men working five days leveled it to the base. There was no stratifica-
tion. Charcoal was found in places, usually on the base, and occasionally bones of
deer in association. In the body of the mound were no human remains, though
two human skeletons were met with superficially. Sherds were very infrequent.
On the base, separately, were five rude arrow heads and an unfinished imple-
ment of chert. In caved sand was a polished hatchet of stone, and two feet from

FIG. 44. Object of Steatite, Mound 4. (Full size.)

the surface was an object of Steatite, 3 inches in length, with cross-hatched decora-
tion on upper surface (Fig. 44).
The disproportionate area of the summit plateau gave to the mound the ap-
pearance of a platform rather than of the usual cone, truncated near the apex. In
view of the possible accidental introduction of the arrow heads and the absence of
human remains and relics other than superficial, we may consider this mound to
have been erected possibly for domiciliary purposes.


On the mainland near Bear Island on the Ocklawaha, about 8 miles from the
mouth, was a small mound of white sand demolished by us with entirely negative

This interesting little mound, on a bluff of the Ocklawaha at Davenport, about
12 miles by water from the St. John's, was composed of yellow sand with consider-
22 JOURN. A. N. S. PHILA., VOL. X.


able quantity of pinkish sand towards the margin. Its height was 3-5 feet, its base
diameter 35 feet. It was razed to the ground.
Human remains were few and very fragmentary, the bunched form of burial
being indicated. In all, 6 polished hatchets of stone were found, one at a depth of
5 feet, a point considerably below the level of the surrounding territory. At depths
to indicate original deposit were a pebble hammer 1-75 inches in length; three
rudely worked or incomplete implements of chert and one fragment showing
cleavage. In addition were shell beads in small quantities, a columella wrought
from the axis of a marine univalve and a'piercing implement of shell.
The feature of the mound was the earthenware which was of the "freak"
variety, and met with at depths to indicate original deposit.
A curious object to a certain extent resembling a spool in form, recalls a
somewhat similar piece figured in Part I, from the mound in the pine woods near
Duval's. In that case, however, portions from either end are missing, rendering
impossible a determination as to original shape. This spool-shaped object from
Davenport, with longitudinal central perforation, has a height of about 4 inches.
One end has a diameter of 3 inches, that of the other is somewhat in excess (Plate
XXVIII, Fig. 1).
In Plate XXVIII, Fig. 2, we have a fragment of an undetermined object of
earthenware, probably belonging to the "freak" variety.
A vessel about 3 inches in height, with laterally projecting rim, having with
mouth aperture a diameter of about 6-5 inches, was found in a somewhat fragmentary
condition. On either side, below the rim, are double perforations for suspension
(Plate XXVIII, Fig. 3).
Two curious objects of earthenware, resembling in shape the ear-plugs known
to have been worn by the aborigines, were discovered in different portions of the
mound. The maximum diameter of the smaller, which is shown Plate XXVIII,
Fig. 4, is 2-6 inches; its height 1-8 inches. The height of the larger is 2 inches;
its maximum diameter 2-8 inches. A perforation runs longitudinally through each.
Several additional vessels of the usual mound type were discovered during the
Nothing in the mound at Davenport indicated a knowledge of the Whites by
its makers.


On the property of Mr. John Purdy, who kindly granted permission to inves-
tigate, about one mile north of Spring Run, was a mound about 2 feet in height,
with a base diameter of 35 feet. It was totally demolished. Charcoal and Palu-
dine were found at various points. A few sherds were met with, but no other
evidence of man's handiwork. No trace of human remains was apparent.
On the property of Mr. Greenleaf, at Silver Spring, was a mound a trifle larger
than the preceding one. It was virtually dug through. Three fragmentary crania,


and several long bones were exhumed, but no sherds other than superficially, and
no other relics, were met with.

Juniper Creek enters Lake George about 2-5 miles west of Volusia Bar. About
1-5 miles above the mouth of the Creek, on the left hand side going up, is a small
shell deposit described by Professor Wyman. A southwestern course for a quarter
of a mile was followed to reach the mound which was concealed by a thick growth
of scrub and trees. Its height was 5-5 feet, its base diameter 40 feet. It was com-
pletely demolished. There was no stratification. Fragmentary human remains
were met with in one place superficially. In another.portion of the mound were
two coarse earthenware bowls, just beneath the surface, showing the usual base per-
foration. With the exception of several fragments of earthenware, lying together
at a depth to indicate original deposit, and a number of bits of charcoal throughout
the mound, nothing showing the work of man was discovered.
The results obtained by the excavation of one mound, frequently overturn
conclusions derived from the demolition of another. Within 8 miles of the rich
pottery bearing mounds of Volusia, and hardly 5 miles distant from Duval's and its
interesting neighbor in the pine woods, symmetrical in shape and absolutely intact,
the mound near Juniper Creek held out every reason to suppose that a rich reward
awaited the archeologist.

About one-quarter of a mile back of Astor was a small symmetrical mound of
white sand. Total demolition yielded nothing beyond a lance point on the base, 4
inches in length.
The division at Palatka in this report of the sand mounds of the St. John's is
not an arbitrary one, since a little north of that place, comes to an abrupt conclu-
sion the great chain of fresh water shell-heaps extending in close succession to the
river's source.1
It will be observed that, with the exception of the two great mounds not far
from the river's mouth, the mounds of the St. John's north of Palatka are of a
height considerably inferior to many of those of the upper river. It must be borne
in mind, however, that the territory north of Palatka, often early Spanish grants,
has long been under cultivation, and that the continuous use of the plow with the
aid of the elements has to a certain extent rounded off the summits. Nevertheless,
judging from the base diameter also, the mounds of the upper river must have
originally exceeded in size most of those to the north of Palatka.
In this connection it is well to remember that a numerously represented class
among the smaller mounds on the St. John's consisted of those where, to a large

I American Naturalist, Nov. 1892, page 919.


extent, irrespective of association with skeletal remains, a comparatively central
deposit was made of objects of aboriginal art, often placed in comparative proximity
to the surface. It is evident, then, that mounds of this class on the lower river,
long under cultivation, have lost a great majority of their mortuary deposits, a
point to be kept in mind in considering their comparative poverty.
Another point to be noticed is the almost total absence of incised decoration in
the earthenware of the lower river, which in nearly every case was plain or deco-
rated through the medium of the stamp, and that certain stamped ornamentation of
complicated design, unknown on the upper river,' was not infrequent in the sand
mounds within a dozen leagues of the river's mouth.
The absence of skeletal remains in condition for preservation is a marked
feature of the mounds of the lower river. Indeed, with the exception of one skull
from the mound near Julington Creek, no human remains other than fragmentary
were encountered by us.
In many instances, it will be remembered, tumuli of the upper river were built
upon shell-heaps, while a certain percentage of shell is often found scattered
throughout the sand. It is from such mounds, where lime salts from the shells act
as a preservative agent, that human remains may be looked for in comparatively
good condition. From the down-river mounds, then, in a region where no shell-
heaps are found, but little in the way of skeletal remains was to be expected.


This mound, on the river's edge in a bearing orange grove, has long been under
cultivation. Its present height is 5 feet 7 inches; its present base diameter, which
probably considerably exceeds the original, is 61 feet.
The original mound was dug through by permission of Mr. Hanna, superin-
tendent of the grove, and subsequently refilled.
The mound was not distinctly stratified, though along the base ran a layer of
sand with numerous bits of charcoal, having a darker color than the remainder of
the mound, which is whitish sand sprinkled throughout with minute pieces of char-
coal. Toward the centre of the mound was a dark layer of sand near the base,
about one foot in thickness, surmounted by a somewhat similar layer not directly
in contact.

Human remains, too fragmentary through decay to determine form of burial,
were found in six or seven instances.

Two feet from the surface, with human remains, was an unsymmetrical imple-
ment with rounded edge, about six inches in length, probably used as a hammer.
Two arrow heads were found separately on the base and four superficially.
1 We have found one specimen of this pottery, superficially, in a mound about ten miles south of


Two and one-half feet from the surface was a discoidal bead of Steatite, -8 of
an inch in diameter. Human remains were in associa-
Comparatively superficial, though doubtless origi-
nally at considerable depth from the surface, were three
small "celts" of stone, separate and unassociated.
One foot from the surface was a pendent ornament
of stone, about 3 inches in length, triangular in shape,
grooved for suspension (Fig. 45).


On the base was a pendent ornament from the colu-
mella of a marine univalve.
Unassociated, near the surface of the mound, was a
drinking cup wrought from Fulgur perversum. -
With human remains, near present surface of
mound, in association, were a pin of shell about 2-5 inches
in length with flat, circular head (Fig. 46); a spherical
headed pin of shell about 4-5 inches in length (Fig. 47); FIG. 45. Pendent ornament of
another, 5-5 inches long, with two circular cavities in store, East Palatka. (Full
head and two below on shank, which, however, are omit-
ted in the cut (Fig. 48). With these were two portions of the columelle of marine
univalves, each about one inch in length, resembling beads, though lacking the per-
foration (Figs. 49 and 50). Shell pins considerably longer than the ones discovered
here, are found in the stone graves of Tennessee.
A curious object of shell (Fig. 51) from this mound was probably worn
through the lobe of the ear. We are indebted to Professor Putnam for details as
to the discovery of similar ones in other localities, lying in close proximity to crania.


Two feet down, unassociated, was a bowl stamped in places, about 2-5 inches
in height; the bottom was wanting.
At the same depth, though not in association, was an unornamented vessel of
earthenware, about 5-5 inches in height and 4 inches in diameter at aperture. The
bottom was missing through intentional fracture. At different points on the sides,
as though still farther to kill" the vessel, were seven perforations (Plate XXIX,
Fig. 1). This is the sole instance of side perforation of earthenware, made subse-
quent to baking, met with by us on the St. John's.


Scattered throughout the mound were another bead of stone, one of clay and a
number of small beads of shell; also pebbles, chippings of chert, and bits of Hematite.



We are informed that a windmill formerly occupied the summit of this
A previous visitor had dug a hole near the summit, about 3 feet in depth,
traces of which were plainly apparent. At the bottom were carefully deposited an

FIG. 49. Imperforate bead of
shell, East Palatka. (Full

FIG. 47.

FIG. 50.

Imperforate bead of
East Palatka. (Full

Pins of shell, East Palatka.

FIG. 48.
(Full size.)

iron hoop, half a glass bottle, and the base of a brass candlestick. Near them was
a "celt" of stone, which the investigator had overlooked or ignored.
Nothing of original deposit in this mound indicated a knowledge of the arts of

FIG. 46.



Rice Creek enters the St. John's from the west 5 miles north of Palatka.
About 3 miles up the creek, on the left hand side going up, nearly opposite the

FIG. 51. Ear ornament of shell, East Palatka. (Full size.)
brick works, is a road about one-quarter of a mile in length, passing through the
swamp to the high pine woods. About 40 yards from the edge of the swamp was
an unstratified mound of yellow sand, 2 feet 10 inches in height, with a base diam-
ter of 44 feet. It was virtually demolished. Two or three mouldering fragments
of human remains were met with, and eight unassociated sherds, mostly of gritty

Deep Creek enters the St. John's from the east, about two miles north of Federal
Point. Following the course of the creek for one mile, and tying to the left bank
going up, we found a mound about 50 yards in the swamp. The service of a guide
was necessary. The mound, which was entirely demolished, was. unstratified, of
whitish sand with admixture of some foreign substance, probably clay, rendering
it extremely tenacious.
Burials, 28 in number, were of the bunched variety, the entire skeleton upon
no occasion being fully represented.
Scattered through the mound, as a rule unassociated, were 8 "celts" of stone.
one having a length of 9 inches. In addition, were a wrought pebble, a slab of
stone 7"5 by 5-5 inches and one inch in thickness, and one arrow head.
One drinking cup from Fulgur perversum and two others somewhat fragmen-
tary were met with. Two showed base perforation.
A number of sherds scattered throughout the mound were, with one exception,
Nothing in this mound gave evidence of contact with the Whites.


This symmetrical mound, in view of the road at Racey Point, had a height of
6 feet, a base diameter of 60 feet. It was totally destroyed. It was composed of
brown sand, through which were scattered particles of charcoal. The base was not
distinctly marked.


Almost in the centre of the mound, about 4 feet from the surface, was a layer
of brownish material, 2 feet long by 1"5 feet broad, and 3 inches in thickness.
This material was soluble in bisulphide of carbon, and when burnt upon platinum
foil, left a small whitish ash. It was bitumen.
We are indebted to the United States Geological Survey for a paper1 from which
we learn that bitumen has a wide distribution, being found, among many other
States, in North Carolina, Georgia, and Alabama. It is also present in Cuba. We
are informed by chemists that analysis will not determine the locality from which
a particular asphalt is derived.
On, and in some cases in, this layer of bitumen were four bits of quartz; one
piece of chert; a pebble about 3 inches in length; an oval shell bead with lateral
perforation, 1-4 inches in length. This perforation was filled with the material in
which the bead lay.
The use of bitumen as a medium of adhesion is not met with on the St. John's
as in California, where vessels were repaired and im-
plements hafted through its agency.
Human remains were met with in but four places,
and in very fragmentary condition. The bunched burial
( was indicated.
'' Six feet from the surface, near fragmentary human
.0 remains unaffected by fire, was a pocket containing a
S" considerable number of bits of calcined human bones.
S"T* l A previous excavation in the summit plateau of
this mound may have disturbed certain interments.
S .In caved sand was a polished stone implement with
cutting edge, about two inches square. It had proba-
C bly seen service in a handle as a chisel or scraper.
\ Two feet from the surface, entirely unassociated,
was a lance head of chert, 5 inches in length, of a pat-
tern unknown on the St. John's, or,'so far as we can
Learn, elsewhere. Above the tang, extending '75 of an
inch, is a straight edge on either side. Above, well-
S. worked serration extends 1-5 inches to where the
weapon begins to taper sharply to a point (Fig. 52).
,1. Four feet from the surface was a small sheet of mica.
S2. The sherds of this mound, while not numerous,
FIG. 52. Lance point of chert,
Racey Point. (Full size.) were interesting. Few were stamped. Some of good
SAsphaltum in 1893, by Clifford Richardson and E. W. Parker.


material were colored with red pigment, while several bore new designs, one appa-
rently representing the stem of a plant with leaves, the whole surrounded by
numerous indentations.
In this mound was remarked for the first time in comparatively many sherds

^ -r; "

o; *_ _:. .^ 1
/l~t~r --- ,,, *,tl 'V
^" ^^C
/*= f ~li~iiieF' ^

FIG. 53.

and vessels, a curious custom hitherto unnoticed by us on the river (and it is our
opinion that its occurrence would have been noted) and unreported elsewhere.

FIG. 5.5.

FIG. 54.

Fragmentary ware, instead of separation by a clean fracture, showed around the
margin the work of a pointed tool, through the agency of which the division had
23 JOURN. A. N. S. PHILA., VOL. X.


been effected.
chipped away.
Figures 53, 54,

Vessels bore evidence of similar treatment where portions had been
This novel and unaccountable proceeding is well illustrated in
55, 56, and 57.

FIG. 57.
FIG. 56.
Sherds showing marks of pointed tools. (Full size.)

Its occurrence was noted in a number of mounds north of Racey point so far
as and including the tumuli at Beauclerc, a distance of 30 miles in a straight line.

FIG. 59.

FIG. 58.
Vessels of earthenware, Racey Point. (Full size.)

Another point of interest in this mound was the absence of base perforation of
any description in the earthenware.


Four and one-half feet from the surface was a vessel with flaring rim, without
decoration. Four small legs on the curved base had enabled it to maintain an
upright position. Part of the upper portion, including about one-half of the margin,
was wanting. Approximate height, 4-25 inches; diameter at aperture, about 5
inches (Plate XXX, Fig. 1).
About five feet from the surface were five earthenware vessels practically
entire, and two in a somewhat fragmentary condition. The imperfect ones pre-
sent no features of interest. The largest of the remaining five is a jar with
contracted neck and flaring rim, a small portion of which has been removed by a
pointed tool in the manner to which reference has been made. Height, about 6-5
inches; maximum diameter, 5-8 inches; diameter at aperture, 3-2 inches (Plate
XXIX, Fig. 2).
The other four are undecorated, save in one instance, where a species of fluting
runs at intervals about half-way up the sides. All have base supports or legs, and
as this feature is of extreme rarity, not on the St. John's alone but in other sections
of the United States, these vessels are all shown full size (Figures 58 and 59 ; Plate
XXX, Figs 2 and 3).
In caved sand beneath the summit plateau was a bowl in fragments in associa-
tion with a vessel of earthenware, 3 inches in height, having a diameter at aperture
of 3-1 inches. This vessel had a novel orna-
mentation conferred by pressure, as shown
Plate XXX, Fig. 4.
Two feet down, unassociated, was a por-
tion of an object of earthenware, apparently
the bowl of a large tobacco pipe (Fig. 60).

No object in the mound at Racey Point
showed evidence of White contact, and we
are, therefore, of the opinion that the mound '. '
was completed in pre-Columbian times. ."


This mound, which was totally demolished
by us, lay one-half mile south of Picolata on
the river's edge. It was the property of
M. P. Usina, Esq., to whom we are indebted
for permission to investigate. Its height was _,
3 feet 7 inches; its base diameter 38 feet.
Fl. 60. Bowl of tobacco pipe, Racey Point.
Previous investigators had rendered almost 60. Bo of tobacco pie, R Point.
(Full size.)
useless the work of exploration. With the
exception of a few arrow points, nothing of interest rewarded our search.



This mound lay a short distance north of the St. Augustine road, about 6-5
miles from Picolata. It presented a somewhat unsymmetrical appearance through
previous superficial examination. Its height was about 5 feet; its base diameter
37 feet. With the exception of the marginal portion, believed to have been largely
washed or thrown from the summit and sides, the mound was totally destroyed. It
was unstratified and contained the usual admixture of charcoal. The form of
burial was the bunched variety, the bones being in the last stages of decay. Two
polished hatchets were found, while 4-5 feet from the surface, in pinkish sand, with
human remains, was a thin sheet of copper, about 3-5 inches by -5 of an inch, with
a thick backing of wood greatly decayed. Sherds were infrequent. Nothing indi-
cating a knowledge of the Whites was discovered.


This mound, near the southern bank of Clark's Creek, about one mile from the
St. John's, had an approximate height of 6 feet, with a base diameter of 70 feet.
It was unsymmetrical in shape, showing considerable wash from heavy rains. It
was virtually demolished. Its composition was yellowish sand, containing occasional
particles of charcoal.
Total results were three or four small fragments of human bone at various
points; several arrow heads; one hammer stone; one small polished hatchet, and
a few sherds, showing separation by pointed implements.
Nothing to indicate a post-Columbian origin for the mound was discovered.


This mound, in the midst of a grove of bearing orange trees, on the river's
edge, about three miles north of Picolata, was the property of Dr. L. H. Harris, of
Pittsburgh, Pa., to which gentleman we are indebted for active assistance, and for
permission to investigate various sites bordering the St. John's, belonging to him.
The mound, the usual truncated cone, had a height of 6-5 feet, with a base
diameter of 64 feet. Upon it grew forest trees, hickory, water-oak, and the slow
growing live-oak, one of the latter having a circumference of 10 feet measured 5
feet from the ground. The mound was destroyed.
Stratification was marked. Extending from the surface to a depth of 2 feet
was a layer of sand pinkish through admixture of Hematite. Below the upper layer
was the yellow sand of the surrounding territory reaching to the base, the line of
which was marked by a dark band about 2 inches in width, sprinkled with charcoal
and showing at places marks of fire.
Upon the base of the mound, including the central portion, was a flooring of
split plank in the last stages of decay, about 13 feet square. Its thickness was 2
inches. A portion of this wood submitted to Professor J. T. Rothrock was kindly


identified by him as red cedar. No evidence in the mound tended to show the
former existence of a house or pen, as is sometimes seen in Ohio. Upon the
planking lay a long stone "celt" in three fragments. Whether this breakage
occurred in process of splitting the wood we can but surmise. Flooring of this
character is unique in our experience of Florida mounds.


In all, thirty-four bundles of bones were met with in separate localities, though
in no case did a bundle represent an entire skeleton, the smaller bones of which
were never included, and even the larger at times were absent. Upon one occasion
portions of three skeletons lay under one cranium.
The bones were mainly within three feet of the surface, and as a rule occupied
a central position, the marginal portion of the mound being almost destitute of
human remains. The bones were in the last stages of decay.
In the eastern slope of the mound, about one foot below the present level of
the surrounding territory, was a bunched burial. One yard south of this was a
species of pocket containing charcoal in small bits, mingled with numbers of small
fragments of human bones, including the head of a radius, a small portion of a
jaw, a bit of a cranium, and a part of a vertebra. With these were fragments of
bones of lower animals.
Near the surface, with disturbed strata above, lay a skeleton at length in
complete anatomical order. The bones of this secondary interment were in good

On the base of the mound was an arrow head of chert showing marks of fire.
Beyond this and the fragmentary "celt," to which reference has been made, no
implements or ornaments lay along the base.
Never deeper than 3-5 feet from the surface, sometimes with human remains,
were twelve hatchets of polished stone, varying from 3-5 to 9 inches in length.
In the eastern slope of the mound, unassociated with human remains, 4-5 feet
from the surface, were two pockets containing 29 and 51 chippings of chert,
Entirely unassociated, 3 feet 8 inches from the surface, not far from the center

FIG. 61. Ornament of polished stone, Harris Mound. (Full size.)

of the mound, was an object, probably an ornament, of highly polished greenish
stone, possibly Serpentine. Its length is 2-7 inches; its maximum width -7 of an


inch. It consists of five hemispheres ranged in a line in contact, much like certain
bar-pins worn by women at the present time. The central and largest hemisphere
has one on either side equalling each other in size, included by two smaller terminal
hemispheres, each -4 of an inch in diameter (Fig. 61). This beautiful ornament has
been submitted to various high authorities, who in every case have pronounced it
probably unique.
Somewhat less than 4 feet down, in mutual contact, but unassociated with
human remains, were a lance point of chert; a pendent ornament of stone, 3-5
inches in length, with double perforation, slightly convex on one side, flat on the
other (Fig. 62); a pebble, almost spherical, having a
diameter of one inch. Through the centre of this peb-
ble, a perforation had been attempted from either side
and for some reason abandoned. To one side of this
a hole had been drilled, resulting in a fracture of the
outer portion of the stone, while on the other side of
the attempted perforation the endeavor had been safely
carried to completion.
With two bunched burials were respectively 28 and
30 large beads wrought from the columella of marine
univalves. The maximum size of any bead was 1-4
inches and 1-2 inches in its major and minor axes.
With the bones of a child lay a shell drinking cup
with perforated base.
No whole pottery was encountered. Sherds were
SThroughout the mound were fragments of red
In the western slope of the mound, above the base,
FIG. 62. Pendent ornament of in actual contact, were about a dozen cockle shells
stone, Harris Mound. (Full (Cardizum) and salt water mussels (Modiola plicatula).
They were unassociated with human remains.


There is no reason to suppose that this mound was constructed subsequent to
the Discovery.

Bayard Point, nearly opposite Picolata, on the western side of the St. John's,
is about seven miles south of Green Cove Springs. On the point is a fortification
of considerable size, with moat, forming three sides of a square. The river side is
unprotected, possibly through erosion by the current. This fortification is said to
date from the Spanish occupation, veterans of the Seminole war positively affirming
its pre-existence to that war.


About one-half mile in a northerly direction inland was a tumulus 4 feet 9
inches in height, with a base diameter of 45 feet. It was leveled to the ground by
the-kind permission of John P. Pratt, Esq., manager of the Borden Estate.
The mound was composed of whitish sand with occasional pockets of charcoal.
It was unstratified and without clearly marked base.
In the extreme southern margin of the mound, where it was almost level with
the surrounding territory, two feet from the surface, lay the skeleton of a man buried
at length in anatomical order. The head was toward the west. Parallel to the
body was a flint-lock musket or rifle, the muzzle toward the feet. In association
were 14 spherical bullets. The bones were in fairly good condition.
In the northern portion of the mound where its full height was attained, 3 feet
from the surface, lay the skeleton of a woman, with a layer of superimposed bark.
Near the cranium were silver earrings of European pattern and the cervical verte-
bras had in association a great quantity of oval white glass beads, and five brass
(not copper) finger rings. On the chest were many small round beads of the same
color and material as the others. Additional ones were present around the wrist.
With the left hand were four bits of glass, the largest, oblong, about 1 by 2 inches.
In the bones of the right hand, resting against the upper portion of the femur, was a
mass of bright red pigment, dissimilar to the red ochre of other mounds. Chemical
tests show this coloring matter to be mercuric sulphide-the cinnabar of commerce.
Somewhat south of the centre of the mound was a male skeleton at length,
with head pointed northwest. Alongside lay a flint-lock rifle or musket, muzzle
toward the feet. With it were a bone-handled awl and probably the remnants of a
powder horn, the base partially studded with brass-headed tacks. In addition, was
an object of iron with a gun flint in contact, probably a flint and steel.
Four or five sherds lay in different portions of the mound, none in association
with the bodies.

The results of this investigation are of deep interest in connection with the
river mounds, and from them we may draw the following conclusions:-
1. That at least one low tumulus1 was built by the Indians of the St. John's
in post-Columbian times.
2. That their most valued possessions were buried with them, including brass,
not copper, and that articles of European make, when in possession of the Indians,
were freely interred, even the easily obtainable red oxide being supplanted by the
commercial cinnabar.


A symmetrical little mound stood about 200 yards from the river, within sight
of the road at Orangedale. Its height was 6 feet, its base diameter 36 feet. It
SThe mound was of irregular shape, and not the usual truncated cone.


was entirely dug through. It was composed of light yellow sand, with a slight
intermingling of particles of charcoal. Sand of lighter shade indicated the base.


The form of interment was the usual bunched variety, and skeletal remains
were comparatively numerous for so small a mound. The usual disregard as to the
selection of bones of the same skeleton was well emphasized in this mound, one
bundle containing fragments of crania, four femurs, and three tibiae.
In the central portion of the base was a layer of human remains, indiscrimi-
nately mixed, a number of square feet in extent, while almost in the exact center
of the base was a deposit of fragmentary calcined human bones in sand whitened
by the action of fire. In the sand beneath this deposit lay unbroken human re-
mains, unaffected by fire, indicating that the interment took place subsequent to
the extinction of the flames.
Several interesting pathological specimens were encountered.


Nine "celts" were met with, from 3 to 8-5 inches in length, as a rule unasso-
ciated with human remains and somewhat superficial, though one was from a depth
of 5 feet.

Sherds were very infrequent. Plain or stamped alone were met with.


Nothing indicating White contact was met with in this mound, whose origin
may therefore be considered as prior to the Discovery.


This mound, 3-5 miles south of Green Cove Springs, was leveled to the base
with the consent of Mr. J. F. Geiger, the owner. Its height was 3-5 feet; its base
diameter 65 feet. It had apparently been under cultivation and had suffered by
the wash of rains. A number of fragmentary human remains were found, and in
one case a bit of Galena within a lower jaw.


This mound, 5 miles northwest of Green Cove, had been considerably leveled
by previous investigation. Its height was 4 feet, its base diameter 60 feet. It was
leveled to the base. No stratification was noticed. In occasional pockets of pink
sand were many shell beads with human remains. Two "celts" of graceful pattern


were found, one roughened for hafting (Fig. 63). Sherds were mainly of the
stamped form of decoration.

Within a few yards of the hotel at this place is at present a very symmetrical
mound, 7 feet high and 60 feet through the base. During the past thirty years it
has been opened a number of times and restored to a more symmetrical shape. Some
years ago, we are informed, two scientific
men from the North, unfamiliar with mound
work and the appearance of disturbed ma-
terial, found towards the centre a tomato
can and a sardine box- a powerful argu-
ment in favor of those advocating the the-
ory of a post-Columbian origin for the


This mound on Fleming's Island lies
within a short distance of the road leading
from the ferry at Black Creek. Previous
investigation has removed so large a portion
that no work was attempted.


~--~~~-I 4._ -___-~

-.-i. --.~

About one mile south of Switzerland, 'I- I. lmpiement oI stone, Muouna near
Peter's Creek. (Full Size.)
75 yards from the river's bank, were three
mounds in a line running north and south, their margins in contact. Beginning
with the northernmost their respective heights were 2 feet 7 inches, 2 feet 7 inches,
2 feet 4 inches; their base diameters 47 feet, 32 feet, and 42 feet. They were com-
pletely dug through, as was a small mound about 150 yards farther south. Nothing
of particular interest was encountered.


One mound, 2 feet 8 inches in height, with base diameter of 66 feet, was
demolished with the consent of Mr. Joseph Lennig, the owner. It had been under
constant cultivation. Bunched burials, charcoal, and sherds were met with much
below the level of the surrounding territory. So often has this occurred in our
experience that we are of the opinion that an excavation was made in many cases
for the first interments over which the mound was subsequently built. Nothing of
particular interest was encountered.
The second mound, about 400 yards east of the one on Mr. Lennig's property,
24 JOURN. A. N. S. PHILA., VOL. X.


was leveled to the base by the kind permission of its owner, J. C. Greely, Esq. Its
height was 4 feet 9 inches. Its shape was somewhat elongated and, exclusive of
material washed down after cultivation, its length was 75 feet. A few fragmentary
human remains were found at different depths, and a great variety of sherds, the
majority showing separation by pointed tools, as described in the case of the mound
at Racey Point. No features of especial interest were noted.


About 3 miles up Julington Creek on the south side, one mile west of Tar
Landing, on the edge of the swamp was a mound 6 feet in height, with base diameter
of 46 feet. At its southern margin was a trench several feet in depth, from which
at least a portion of the material for the mound had been obtained. The summit
plateau had been superficially examined previous to our visit.
With a force of ten men to dig, the mound was completely demolished within
four days.
It was composed of gray sand with an upper stratum of irregular thickness,
somewhat the color of terra-cotta, which analysis showed to be from oxide of iron
modified in shade by vegetable matter.


The usual bunched method of burial prevailed exclusively, or rather interment
after exposure, since in some cases the usual heaps were somewhat modified, the
bones being deposited at length, though not in anatomical order.
Five feet from the surface were human remains with long bones end to end for
a space of several feet; the cranium lay upon a tibia.
In all, 27 separate deposits of bones were observed (though some escaped us in
caved sand), consisting sometimes of but a single cranium unassociated; of a few
long bones without the skull; of the usual bunched burial; of the long bones at
length, of which mention has been made; and again of masses of bones of many
A little east of the center, about 4 feet from the surface, was a large bundle of
bones with four crania, while an isolated skull was one foot above.
Nearly at the center of the base of the mound was a matted mass of bones,
having a maximum thickness of 16 inches. The length of this deposit was 5-5 feet,
its greatest breadth 4 feet. Bones of all ages and sexes were piled together indis-
The almost universal use of fire in connection with the construction of mounds
of the St. John's was well illustrated in the case of the mound on Julington Creek.
In addition to the usual sprinkling of charcoal throughout the sand, there were at
least four fireplaces immediately over deposits of human remains, and these remains
gave no evidence of contact with the flames, save in one instance where a tibia
slanted upward through the sand into the fireplace.


A number of human bones lay 6 feet from the surface, near the center of the
base, with two crania in contact. All were unaffected by heat, though the remains
of a fire, 4 feet in length, were plainly apparent in the sand about one foot above.
Among the extinct embers lay considerable portions of a human tibia, a fragment of
lower jaw, and other human remains thoroughly charred.
In various portions of the mound human remains similarly treated were met
Four and one-half feet down, in the western slope, was a pocket of fragments
of human bones, unaffected by fire. With them were small pieces of calcined bone,
some of undetermined identity, others of lower animals.


I j ii

FiG. 65.
FIa. 64.
Implements of polished stone, mound near Julington Creek. (Full size.)

The reader is referred, in connection with the treatment of human remains in
this mound, to "Aboriginal Remains of the Piedmont and Valley Region of Vir-
ginia" by Mr. Gerard Fowke.1
One cranium (Academy Catalogue, No. 1783) was secured from near the base
in fairly good condition.

A few bits of Hematite were scattered throughout the mound. On the base,
unassociated, were two "celts" of stone, some distance apart.
SAmerican Anthropologist, October, 1893.


Three feet from the surface, unassociated, was an implement of stone, about 4
inches in length, departing somewhat in shape from the ordinary "celt," and more
resembling the modern idea of the hatchet (Fig. 64).
Of undetermined depth, in caved sand, was a flat implement of polished stone,
with cutting edge at one extremity. Its length is 3-5 inches; breadth 2-4 inches;
maximum thickness about -6 of an inch. We have not met with this type elsewhere
on the river (Fig. 65).
Upon two occasions, small shell beads were found with crania.
A number of sherds were met with at all depths, plain, stamped, and in two
instances, otherwise decorated. On the base was a portion of an undecorated bowl,
represented by hardly more than a few inches below the margin of the aperture.
Its diameter is 12 inches.

There is no reason to assign to the mound on Julington Creek an origin other
than pre-Columbian.


Mandarin Point makes into the St. John's, just north of Julington Creek. It
is now covered with a heavy growth of oak and pine, though the tract, an old
Spanish grant, plainly shows marks of previous cultivation.
The northernmost mound had lost considerably in height during civilized times.
Its altitude was 4 feet, its base diameter 55 feet. The work of a former investigator
was apparent. It was levelled to the base by us. Little of interest was discovered
beyond very fragmentary human remains, a few arrow heads, and many sherds on
which the work 'of pointed implements, to which we have already made reference,
was noticeable.
About one-quarter of a mile in an easterly direction from this mound was
another 3 feet 7 inches in height and 58 feet through the base. This dispropor-
tionate base diameter is a probable indication of a diminution in height through
cultivation in civilized times. The mound was totally demolished. Absolutely no
human remains were met with.
Three feet from the surface, near a deposit of charcoal, was a vessel 7 inches in
height and 7-5 inches in diameter. Its condition was somewhat fragmentary though
allowing restoration. Its ornamentation consists of a series of short intersecting
lines, formed by pressure of a twisted cord into the wet clay (Plate XXXI, Fig. 1).
The southernmost mound on Mandarin Point lay about 50 yards from the river's
bank, though hidden from view. Its height was 5 feet, its base diameter 47 feet.
A large central excavation had previously been made by the owner of the mound,
and we are of the opinion, judging from results obtained by us, that a report as to
the discovery of numerous articles during the first investigation is correct. The
remaining portion of the mound was virtually demolished.


The line of the base was considerably below the level of the surrounding ter-
ritory. Upon it was a layer of sand from one to two feet in thickness, blackened
by fire and filled with charcoal. In this layer lay many oyster shells and several
bones of the lower animals, among which were remarked part of the shell of a
turtle, the lower portion of the femur of a deer, etc. A slight sprinkling of char-
coal was apparent throughout the mound. Above the dark layer was sand of a
yellowish hue to the thickness of about one foot, and this was surmounted by a

0%_ _

W H A!3'

Fr(. 66. Tobacco pipe of coquina, Mound on Mandarin Point. (Full size.)

stratum 4 feet in thickness at the center of the mound, pink in color, and even at
times a bright cherry, through admixture of HIematite,
Human remains were of infrequent occurrence.
Five celts were met with separately, apparently unassociated with human
remains, from one to two feet from the surface. One was buried in a pocket of
scarlet sand; another, very symmetrical, has a length of 10-75 inches, the longest
from Florida to come under our notice, though implements of this character 13 and
14 inches in length are reported from other sections.
Unassociated, about one foot from the surface, was a tobacco pipe of coquina
with original fracture of the bowl. It was additionally injured by the blow of a
spade (Fig. 66).
Nothing in this mound in any way pointed to White contact.


Near the northwestern extremity of Doctor's Lake, at a settlement named
Peoria, is a low mound of yellow sand, the property of a native called Silcox, whose


charge for the privilege of excavation is disproportionately great considering the
unimportance of the mound.


In the town of Orange Park are two mounds which have been dug into to such
an extent that farther investigation was deemed inadvisable.


This mound, on property controlled by W. D. White, Esq., of Mandarin, was
courteously placed at our disposition by him.
The mound, about one-half mile east of Beauclerc, has been under continuous
cultivation and its height, 5 feet 7 inches, was no indication of its altitude in former
times. Its base diameter was 68 feet.
The mound was completely demolished.
It was composed of brownish sand. On one portion of the summit was sand
of a terra-cotta shade similar to the upper stratum of the mound at Julington
Creek. Probably the remainder of this layer was wanting through the agency of
the plough.
The base of the mound, somewhat below the surrounding level, had been one
vast fireplace. On it at one point, covering a number of square feet, were scattered
fragments of human bones and of bones of lower animals charred and calcined by
fire. Immediately above these, occupying a central position on the base, was a
solid mass of human bones in the last stage of decay. These bones, owing to ad-
vanced state of decomposition and pressure from the sand above, had virtually lost
their distinctive shapes, and lying in a compact mass, closely resembled a rotten
plank. They were unaccompanied by relics of any sort and showed no marks of
fire, proving that, previous to burial, the flames that charred the bones below, had
been allowed to cool. Local fireplaces were in various portions of the mound.
Rising from the base in the form of a cone beneath the yellow sand was a mass of
sand darkened by admixture of charcoal. Its exterior was marked by a black line
indicating the base of another fireplace. Toward the center of the mound the
apex of this cone was within one foot of the surface. In various portions of the
mound were fragmentary burials, representing portions of the skeleton, while upon
two or three occasions, not far from the surface, were skeletons in anatomical order.
One of these, the bones of which fully retained their organic matter, presented a
rather gruesome appearance, and doubtless belonged to a colored man who, we are
informed, had been interred there some ten years previously.
An occasional Fulgur caricaa) lay on the base, where also, at different points,
were found a small arrow head and two oval pebble-hammers.
One lance point and three "celts" were found separately from two to three
feet from the surface.
Sherds plain or stamped were abundant.



This mound, about one-quarter of a mile east of the smaller one, was of un-
usual shape as shown by diagram (Fig. 67). Its height, at the time of our visit, was


FIG. 67. Plan and elevation of larger mound, Beauclerc.

somewhat over 6 feet, though continuous cultivation had doubtless transferred
masses of sand from the upper portion to the margin of the mound. A small cir-
cular elevation, E, having a height of about 18 inches, contained nothing beyond
scattered sherds.
We were indebted, as in the case of the smaller mound one-quarter mile dis-
tant, to W. D. White, Esq., for permission to investigate.
Seventeen men to dig and four to direct the work, levelled the mound to the
base within six days.
It was composed of yellowish sand with the usual sprinkling of charcoal,
though no fireplace. marked the base which was of yellow sand of brighter shade.
At the highest portion of the mound, a little over six feet above the surrounding


level, charcoal was mingled with the sand to a depth of 8 feet. Locally through-
out the mound were fireplaces occasionally containing small fragments of bones,
human and of the lower animals.
Skeletal remains were found at every depth, though very widely separated.


FIG. 68. Implement of chert, larger mound, Beauclerc.
(Full size.)

FIG. 69. Section of same.

The form of burial was the bunched variety, and the bones were in the last stage
of decay, at times represented by fragments only.



Eighteen inches from the surface was a chipped implement of chert, about 6-5
inches in length, pointed at one end, with a cutting edge at the other. We have
found nothing of this type previously in the river mounds (Fig. 68).
A number of pebbles and eight arrow points were found separately and unas-
sociated at various depths.
Five feet from the surface were twelve small chippings of chert with no other
object in association.
Two small masses found together in this mound were chemically determined,
through the kindness of Dr. E. Goldsmith, to be an intimate mixture of carbonate
and phosphate of calcium and some aluminous matter"-of common occurrence in

In caved sand, and therefore of undetermined depth, was a heart-shaped
earthenware vessel. Exteriorly and interiorly are traces of red coloring matter.
A portion of the margin shows chipping from pointed tools, as does the vessel next
to be described, and the great majority of sherds in this mound which, the reader
will recall, is the northernmost limit of the prevalence of this curious custom so
far as our observation extended. The height of this heart-shaped vessel is 1-5
inches; its maximum diameter about 2-5 inches. It was apparently unassociated
(Plate XXXI, Fig. 2).
One and one-half feet from the surface was a small pot marked with red pig-
ment inside and out. Small handles or indications of them are on either side of
the aperture, the diameter of which is 2 by 2-5 inches. Height about 2-5 inches.
This vessel (Plate XXXI, Fig. 3) is of high specific gravity, and like all the whole
vessels, and most of the sherds, seems to be of gritty ware.
Two and one-half feet down, lying bottom up, unassociated, was an undeco-
rated bowl, 4-5 inches in height and 10-5 inches maximum diameter. It was of
superior manufacture with carefully bevelled edge. An accidental blow from an
axe, causing a certain base mutilation, showed the material to contain small frag-
ments of chert (Plate XXXI, Fig. 4).
In another portion of the mound was a small bowl undecorated and of poor
In caved sand was a neatly made vessel with six encircling parallel rows of
punctate decoration, extending laterally about two-thirds down the body from the
aperture, which is contracted to a diameter of about 1-75 inches; maximum diam-
ter about 3 inches; height about 2-5 inches (Plate XXXIII, Fig. 2).
Four and one-half feet from the surface were fragments of a large vessel with
stamped decoration of complex pattern resembling that of Georgia and of Caro-
lina. The rim is doubled over. The material is filled with fragments of chert, at
times the size of a pea. Fragments of this vessel were many feet apart, as though
strewn upon the mound in course of construction (Plate XXXIII, Fig. 1).
25 JOURN. A. N. S. PHILA., VOL. X.


Many other sherds of types of decoration uncommon on the St. John's were
found at various depths (Plate XXXII, Figs. 1 and 2).
Attention is called to the fact that no earthenware in this mound showed base
perforation of any sort, as was the case at Racey Point.


So continued had been the cultivation of the two mounds at Beauclerc that
objects superficial when found, had unquestionably originally been placed at a much
greater distance from the surface. It is our opinion that nothing in these mounds
was of secondary introduction, and that the mounds, through absence of all evidence
tending to connect them with White occupation, were abandoned prior to the coming
of Europeans.

The Grant Mound, on a high bluff overlooking the southern bank of the river,
at a point about 2 miles west of Mill Cove, has a base diameter of 216 feet. Its
height above the level of the bluff is 26 feet 8 inches; its shape, the usual trunca-
ted cone, so symmetrical in this case that the diameter of the summit plateau is but
24 feet.
The owner of the mound, James B. Grant, Esq., to whom with his family we
are indebted for many courtesies, gave cordial permission to investigate, though a
natural desire on the part of the owner to preserve so notable a landmark prevented
an attempt to demolish the mound. Unfortunately, unless a change of channel
occurs, this great aboriginal work will fall a prey to the river into which fully one-
third of its material has already gone, furnishing to neighboring residents a rich
harvest of aboriginal relics.
Six days, with a force of 16 men in all were devoted to the mound. Its com-
position, judging from the parts exposed by us and section laid bare by the water, is of
yellow sand with a base blackened by fire, containing a heavy percentage of char-
coal. Charcoal to a much less extent is sprinkled throughout the mound, while small
local layers and pockets of gray sand and of white sand, both of natural hue, and
of red, of pink, and of cherry sand colored by Hematite, abound, usually in connec-
tion with human remains and aboriginal relics. Near the surface beneath the sum-
mit plateau were masses of oyster shells, while single shells are scattered throughout
the mound.

The form of burial was that of bundles of bones piled without reference to
anatomical order. In no case was the entire skeleton represented, and frequently
but small portions were found. One bundle of bones was made up of six femurs
and two humeri belonging to at least four different individuals. All bones were in
the last stage of decay, making impossible any successful attempt at preservation.



In all, nine polished hatchets, or "celts,' were met with during the investiga-
tion, with one exception lying within a few feet of one another as to area, though
some were in caved sand, and others from 4 to 7 feet in depth. As a rule no bones
were found in association.
Three and one-half feet from the surface, with human remains, was a flattened
tube of sandstone 2-6 inches in length, with a maximum diameter of 1-5 inches.

FIG. 70. Bead of sheet
copper, Grant Mound.
(Full size.)

FIG. 71. Ornament of sheet copper, Grant Mound. (About full size.)

Another tube of about the same proportions, rough on the outside, has somewhat
the appearance of a natural formation.


A piece of Galena, somewhat over a cubic inch in size, lay unassociated at a
depth of 1-5 feet beneath the summit plateau.


Small beads of shell were found in great abundance, always in association with
human remains.
Several unimportant objects of shell were met with during the investigation,
including a drinking cup wrought from Fulgur perversum.


The presence of copper of aboriginal design, at depths to denote original intro-
duction, was a feature of peculiar interest in the Grant Mound.
In caved sand was a nearly spherical bead of sheet copper, -7 of an inch in
diameter (Fig. 70).


Five feet from the surface, in sand rendered bright cherry by admixture of
Hematite, with fragmentary human remains and large numbers of small shell beads,
were three tubular beads of thin sheet copper, with overlapping edges. Of these,
the two largest were each one inch in length with a diameter of "3 of an inch; the
smallest, about one-half these dimensions.
In another portion of the mound, 5 feet from the surface, was one tubular bead
of copper, similar to the largest above described.
In caved sand was an oblong sheet of copper 3'7 by 2-6 inches, with a central
oval boss having the usual perforation for attachment or suspension. At the three
remaining corners was repoussi decoration as shown in Fig. 71, which, by the way,
correct in other respects, is too long by -3 of an inch.
Seven feet below the surface, apparently unassociated, was a large oval bead
of sheet copper about 22 inches in length, with a maximum diameter of 1'2 inches
(Fig. 72). This bead, almost similar in shape and size to that described as coming

FIG. 72. Bead of sheet copper, Grant Mound. (Full size.)

from Mt. Royal, differed in that the edges met without overlapping. Like the Mt.
Royal specimen it was made of a single sheet of copper hammered over, differing
from Ohio specimens, which, as we have stated, are made of two sections fitted

Sherds were comparatively of infrequent occurrence and were almost univer-
sally undecorated or stamped in squares.
Five feet from the surface, in a pocket of gray sand and charcoal, was a tureen-
shaped vessel of earthenware, 5 inches long, 3 inches broad, and about 1-3 inches
in height. Through the bottom a small hole had been knocked subsequent to
At a depth of five feet, in another portion of the mound, was a curious object.
of earthenware. The body, from which the base had been intentionally omitted
in manufacture, resembles in shape an inverted truncated cone. The laterally
projecting rim, including aperture, has a maximum diameter of 4-7 inches. The
maximum diameter of the body is about 2-5 inches, its base diameter about one inch
less (Plate XXXIII, Fig. 3). We found in Mt. Royal an object of earthenware
similar to this in size and shape.


In the talus at the foot of the bluff was found a small semiovoid vessel with
base perforation made previous to baking.


The reader has probably remarked by this time that hearsay testimony as to
mounds and objects discovered in them has not occupied a prominent position in
this report. This has not been through lack of abundant supply, but while much
of this information is probably correct, much more is confused or not based upon
During the score of years that the mound has gradually succumbed to the wash
of the river, according to those living in the vicinity many curious objects have been
gathered by those keeping a watchful eye on the talus, especially after storms. We
have seen in the possession of reliable persons living near the mound, masses of
Galena, the largest about the size of a closed fist; various vessels of earthenware;
a discoidal stone, carefully dressed; several polished hatchets of stone; a crystal of
quartz about 3 inches in length and 1-5 inches in diameter, grooved for suspension
at the base. Laudonniere refers to "deux pierres de fin christal" as gifts from
southern Indians.
Mrs. Martha A. Millspaugh, residing not far distant from the foot of the Grant
Mound, has an interesting collection of objects, all of which, we are informed, were
obtained from the sand on the exposed side of the mound. We are indebted to her
for a piercing implement of copper, and a portion of another of great interest, since
the section at the point of fracture shows it to be formed of numerous thicknesses
of thin sheet copper beaten into close contact. These specimens, used for analysis,
will be particularly referred to in our Note on Copper.


Although we are told by the Huguenot chronicle1 that various paracoussys, or
kings, were in control along the borders of the St. John's, Satourioua, near Fort
Caroline (several miles below the Grant Mound), and Outina at a point up the river,
which archaeologists have considered to be at or near Mt. Royal, there can be no
question that tribes having identical mortuary customs, built the two great mounds
to which reference has just been made.
The unusual size of the mounds, the layers and pockets of brightly colored
The graphic story of Laudonniere detailing the ill-fated Huguenot expedition to the new world
should be in the hands of all making a specialty of the archeology of Florida, including as it does minu-
test particulars of the manners and customs of the aborigines.
While the Brevis Narratio, containing the story in Latin, with plates by Le M3oyne de Morgues, a
survivor, published by de Bry (Frankfort a M, 1591), or the first French edition of this "L Tistoire
Notable de la Floride" (Paris, 1586), or Hakluyt's quaint rendering into English of the same, may not be
within easy access to all, the Paris edition of 1853 is readily obtainable, as is the English translation in
"Historical Collections of Louisiana and Florida," New York, 1869. We are informed that an English
translation with photographic reproductions of Le Moyne's plates was published in Boston some years ago.
We have been unable to obtain a copy.


sand, the presence of copper, the occurrence of identical forms of "freak" earthen-
ware, taken with a similar form of burial establish this beyond reasonable doubt.
Unfortunately, the investigation of the Grant Mound has not been so extensive
as that of Mt. Royal, but nevertheless no mean amount of work was accomplished.
In this mound we have found no more traces of the iron distributed by the
Huguenots than were apparent in Mt. Royal, to whose king, we are distinctly in-
formed, such objects were sent, and it is our opinion that the Grant Mound, like
Mt. Royal, was a monument of the past when the Whites first set foot in Florida.


About one mile west of Mill Cove, some 150 yards from the river, on high
rolling ground sweeping the St. John's for miles, is an Indian earthwork of great
size, and of a form entirely novel for. mounds of the St. John's. In shape, it
resembles a circular platform, having a base diameter of 214 feet, with a circular
summit plateau 108 feet across. On the side facing the river, an approach carefully
graded, 126 feet in length and 88 feet in breadth, leads to the upper level. Such
approaches are not found elsewhere on the St. John's, though somewhat irregular
causeways of shell were noted, as the reader will recall, at Tick Island and at
Thursby Mound.
So regular and symmetrical in every way is this great earthwork that its origin
was universally attributed to the Huguenot French by those living in the neighbor-
hood. It is not likely, however, that the ill-fated founders of Fort Caroline (prob-
ably at St. John's Bluff, a few miles below) had time or disposition to pile up great
masses of sand, while works of defence are not approached by broad avenues of
easy grade. Moreover, Le Moyne, an eye-witness, has left us in Plate X1 a repre-
sentation of Fort Caroline showing no earthworks in connection with it.2
There can be no doubt, as the reader will presently see, that the mound near
Mill Cove is of aboriginal origin.
The owner, a foreigner, alleging superstitious terrors on the part of his family,
would consent to no extended investigation, though influence and inducements were
brought to bear.
A number of trenches were dug from the surface of the summit plateau, none
exceeding 6 feet in depth. These excavations were subsequently carefully refilled.
So far as our observation extended, the mound is of yellow sand with the usual
sprinkling of charcoal. Local layers of red sand near the surface, some several
feet in thickness, were encountered. In but one excavation, which happened to be
on the edge of the summit plateau, were human remains. These remains, the
1 Brevis Narratio.
2 While the drawings made by Le Moyne, or from his descriptions, are not exact in detail, as, for in-
stance, the use of bows of Chinese type by the Indians, and the obliging manner in which alligators allow
great stakes, propelled by half a dozen men, to be forced down their throats, yet we know bows to have been
in use, and doubtless heavy spears were thrust into the jaws of alligators which were by no means timid
on the St. John's at even a much later period, as we are told by the traveller Bartram. It is unlikely,
therefore, that features of importance, such as the appearance of Fort Caroline, would be misrepresented.


usual bunched form of burial, lay beneath a layer of sand colored red by admixture
of Hematite, and showed no evidence of previous disturbance.
The only objects of aboriginal workmanship met with in the mound were 5
arrow points and a tureen-shaped pendent ornament or toy of earthenware, with
perforation at end for suspension. Length 2 inches; breadth, 1"3 inches (Fig. 73).
A house, now destroyed by fire, at one time occupied
the summit of the mound. It is not probable that the
mound was in any way levelled for its reception, since
examination of the steep and symmetrical sides showed
no appearance of deposit from above.
It is to be hoped that at some future time a change of
ownership may permit farther investigation.


About one-half mile southwest of Fulton, a settle-
ment just south of St. John's Bluff, is a mound in a light
grove of oak and pine. The mound and surrounding FIG. 73. Pendant of earthen-
ware, mound near Mill Cove.
territory give evidence of previous cultivation. The (Full size.)
height, which has evidently suffered considerable diminu-
tion, is at present 7 feet, the base diameter 90 feet. About one-third of the area of
the mound was levelled by kind permission of H. J. Hole, Esq.
Locally were layers of oyster shells from one to several feet in thickness, with
occasionally in association fragments of bones of lower animals and vertebra of fish.
Sherds were infrequent.
Superficially was a fragment of human radius, but beyond this no signs of
human remains were met with, nor any implements whatever.


St. John's Bluff, by some supposed to be the site of the ill-fated Fort Caroline,
towers high above the water on the south bank of the river, about six miles from
the sea. At the point of union of the St. John's and a small creek, the bluff forms
a right angle, having a side parallel to each, the northern side overlooking the river,
the eastern rising above the creek and adjacent marsh.
Along the edge of this eastern side is a line of curious elevations, some oblong,
some conical, all very symmetrical. A few were utilized as batteries during the
late war by the removal of central portions. -
The most symmetrical of these eminences, almost a perfect cone, had a height
of 9 feet and a base diameter of 47 feet. About one-half of this mound was dug
away by us, removing a number of feet from the summit, and exposing a portion of
the base. Absolutely nothing indicating human agency was met with, even the
heretofore, ever-present, charcoal being wanting. We are in doubt as to the origin
of these mounds.




That the reader may be placed on an equal footing with ourselves in consider-
ing the probable age of the sand mounds of the St. John's River, we will give a
brief resume of the results of all our investigations. During the years devoted by
us to mound exploration on the river, we have seriously investigated about seventy-
five sand mounds, the majority of which we have levelled to the base, while the
remainder have received far from superficial attention.
In a number of these mounds, possibly half a dozen, we have found, super-
ficially, articles of iron, glass, etc., buried with later Indians, presumably in pre-
existing mounds, since, with other burials in the body of these mounds, articles of a
like nature were not encountered. Thursby Mound was a good example of this,
where, it will be remembered, iron implements lay near the surface, while at
greater depth none but objects of aboriginal manufacture were met with, though the
mound was completely demolished.
At Raulerson's, burials had been made at a late period on a pre-existing shell
heap, and covered with a foot or two of sand.
But one mound on the St. John's has shown human remains, other than super-
ficial, associated with objects of European manufacture, namely, the low mound at
Bayard Point.
The reader, turning to our description of this mound, will find that muskets,
tools of iron, bullets, bits of glass, and beads of the same material, with a red paint
of commerce, cinnabar, lay in association with the dead, and it is our opinion that
the discovery of this small mound, with its profusion of articles obtained from the
Whites, serves clearly to illustrate the tendency of the Indians to inter such pos-
sessions with the departed, and emphasizes the character of the mound.
In considering the question of the period of the construction of the mounds of
the St. John's, having no data as to their existence at the time of the Discovery,
though others are reported in regions not far removed, we are compelled to rely
mainly on negative evidence.
Now negative evidence, while not absolutely final, is generally accepted where
a sufficient amount is confronted by none of a positive character, and many things
universally admitted as facts are established by negative evidence alone. In this
dearth of positive information the story of the mounds must be told by the char-
acter of their contents, and inferred from what they do not contain.
We know it to have been a common custom to inter with the dead, possessions
in general use. With superficial burials we find a great variety of articles furnished
by the Whites, and highly prized by the aborigines.
We think then, it must be admitted that on the St. John's, as elsewhere, what
the Indians had they interred with the departed, and that when in an entire mound
no articles of European manufacture are met with, an explanation must be sought


under the hypothesis that such articles were wanting among them, and that the
makers of the mounds lived at a period when such objects were unobtainable, a
period prior to or bordering on the Conquest.
To one who has carefully read the reports of the two investigations of the
great mound at Mt. Royal, and noted the hundreds of objects discovered therein,
seemingly the product of aboriginal art, and the absence of so much as one bead
of glass or one implement of iron in the great mass of sand moved by us, the
conviction must be assured that a people so comparatively rich, and so devoted to
its dead, would surely have possessed and taken pride in interring with the departed
the many objects of gift and of barter, which Indians always acquire in intercourse
with Europeans, had the mound been in process of construction in post-Columbian
In this connection it may be well to notice a suggestion made by Professor
Holmes in his able and interesting paper on the earthenware of the St. John's (page
109), published with the first part of this report, that the practice of perforating
the base of mortuary pottery arose in post-Columbian times, through a desire to
render less attractive to the Spaniards the contents of the mounds.
In the preparation of his paper, Professor Holmes labored under a certain dis-
advantage in that, through our absence in Florida, he was unable to have clearly
set before him certain points in relation to depth and association of objects discov-
ered in the mounds, and it was with no small wonder, upon reading his paper, we
became aware how clearly in nearly every respect, from our somewhat imperfect
communications, he had grasped the subject.
However, such a hypothesis relating to the base perforation of mortuary pot-
tery, involving as it does so late an origin for the mounds, since this pottery is found
upon the base of many of them, seems to us entirely untenable. In the first place,
no European objects have been found in association with pottery of this class. And,
secondly, it is hardly to be supposed that aborigines interring only rude and com-
paratively useless earthenware through fear of subsequent loss by pillage should
place in immediate association unbroken ornaments of copper, entire implements of
polished stone, pearls, and other possessions dearest to the native heart.
We are then of the opinion that the manufacture of flimsy forms and mutila-
tion prior or subsequent to baking was practised in pre-Columbian times in deference
to some religious rite or established custom, and not at a later period through fear
of plunder by the Whites.
While we are not in a position to make a positive statement, it is our firm
conviction that at least all the larger mounds of Florida are of pre-Columbian origin.
It is but recently that thorough mound investigation has been practised, and it is
to be regretted that too frequently conclusions have been drawn by strangers to the
mounds, whence the material was derived,-and who were, therefore, not in a position
to consider the situation from every point of view.
To box the contents of a mound, the superficial with the base deposits, and to
label the whole "From a mound, Florida," is worse than useless-it is misleading-
26 JOURN. A. N. S. PHILA., VOL. X.


since the iron axe, the beads of glass, the pewter, and the lead from the superficial
burial are not distinguished from the deposits from the base, and too frequently are
factors in the attribution of the age of the mound.
The mounds of the west coast of the lake country, and of the prairie lands of
the northernmost portions of the State have received but superficial attention.
The Bureau of Ethnology, Report 1883-1884, page xxi, contains the statement
that but few of the Florida mounds were built for burial purposes, from which
statement the thoroughness of the work may be inferred. On the other hand, Mr.
Andrew E. Douglass, who has given serious attention to forty mounds of the east
coast, informs us that upon not one single occasion has he found in these mounds
other than superficially, objects connecting them with the civilization of the Whites.
In conclusion, we are firmly of the opinion that all the larger mounds of the
St. John's, and, with trifling exceptions, the smaller ones, date from a period prior to
the coming of Europeans.


During the four seasons covering our researches on the St. John's River, we
have given special attention to the question of the contemporaneity of the sand
mounds and of the shell-heaps of the St. John's, carefully weighing reasons for and
against, and especially directing our researches with a view to elucidate certain
points previously clothed in uncertainty, questions which the lamented Wyman,
pioneer investigator of the archeology of the river, doubtless would have taken up.'
Unfortunately, the results of our researches have not put us in a position to
make a positive statement. But it is our opinion that what we believe to be the
earlier of the sand mounds, as for instance that at Tick Island and the Thursby
Mound, were contemporary with what we consider the later shell-heaps; but
whether sand mounds were constructed contemporaneously with presumably the
older shell-heaps, and whether the period of the shell-heaps had not come to a close
when what are probably the comparatively later sand mounds, as typified by Mt.
Royal, were constructed, we are unable to say.
Before presenting in detail the reasons upon which this opinion is based, we
shall give a brief summary of the principal features of the shell-heaps of the St.
John's, founded upon our work of two seasons with steam-motive power and an
adequate force of men, virtually covering every known fresh-water shell-heap on
the river.2
SProfessor Wyman says: We leave out of consideration the burial mounds, which may possibly be
as old as the shell-heaps, because they have not thus far been satisfactorily examined and proved to be so.
Nearly all the explorations of them have been confined to the superficial portions where there are mixed
burials of an earlier, though perhaps not of the earliest, and the later inhabitants. The only mound
which.we have had an opportunity to examine quite to its base had only recent burials on the top, but
none whatever lower down. In its general appearance it exactly resembled the others."--Fresh Water
Shell Mounds of the St. John's River, Florida, page 47 (foot-note).
2 Partial results of these investigations are given in the American Naturalist, November, 1892, January,
February, July, August, 1893, January, 1894.


Bordering the St. John's and its tributary streams, beginning at a point a little
north of Palatka, and extending to the river's source, are heaps mainly composed
of fresh-water shells, with a small percentage of bones of lower animals, and occa-
sionally of man. Many of these shell-heaps, including some of the largest, have
no burial mounds of sand in association; at all events, above the surface. At all
depths in every shell-heap, without exception, are the remains of ancient fireplaces,
bearing certain testimony to the artificial origin of these heaps of shell which,
formed from kitchen refuse, gradually increased in size during the lapse of years.
In certain of these shell-heaps persistent search on our part has failed to reveal
a single fragment of earthenware which we know to be so abundant in other of the
river shell-heaps, and this negative result was also reported by Professor Wyman.
This absence of earthenware was attributed by him to the lack of its possession by
the makers of such heaps, and we see no reason to differ from that conclusion.
It has been asserted that the absence of fragmentary earthenware can be
accounted for under the hypothesis that aborigines residing on the shell-heaps made
their pottery elsewhere; or, even admitting its introduction, that the rude earthen-
ware of these people was subject to decay. To these suggestions we would reply
that no people habitually using earthenware, though made elsewhere, could form
and occupy great masses of shell without leaving behind numerous fragments of
their fragile vessels, as they have done in the other heaps; and that as we have
never in Florida met with aboriginal earthenware showing partial decay, we are
not inclined to the belief that the absence of pottery can be accounted for by this
agency. Moreover, in shell-heaps where pottery is discovered to a certain depth,
the lowest sherd-bearing stratum shows its earthenware to be, so far as decay is
concerned, in a condition equally as good as that of fragments from the upper por-
tions of the heap. There is absolutely no transition from strata devoid of earthen-
ware through others containing it in a crumbling condition, to layers of shell holding
the solid sherds, and this we say after seasons of work conducted under our personal
supervision, in our immediate presence at the excavation, with a corps of trained
assistants to examine every object brought to light, and to submit the same to us,
and in not one single instance have our conclusions been arrived at from the reports
of others as to discoveries made during our absence.
In comparatively recent times races have been met with to whom the manu-
facture of pottery was an unknown art, and we see no inherent reason why to the
early savages of the Peninsula we should not attribute an equal ignorance.
We do not assert that the art of manufacture of earthenware was self-taught
to the makers of the shell-heaps, acquired by the slow process of evolution, since
we are in possession of no data for or against; but we do say, and we are fully
convinced, that there was a time when the men of the earlier shell-heaps did not
include earthenware among their possessions. It seems not unlikely that the art of
pottery making was brought by a more advanced people, and that the abundant
sherds of certain shell-heaps prove such heaps of a later date.
In refuse heaps fragmentary objects alone can be expected, though from time


to time articles of value, doubtless through accident, found their way into the heaps
of debris, and it is fair to suppose that long-continued and persistent investigation,
involving the careful removal of great masses of shell, should yield a representative
collection of objects, many, of course, fragmentary, formerly possessed by the
makers of the shell-heaps. Beyond a very occasional arrow head, shell-heaps
devoid of pottery are virtually barren of relics of aboriginal art, nor in the heaps
containing earthenware is the search of the investigator as a rule much more
richly rewarded. Possibly an awl of bone or a gouge of shell may be brought to
light, but articles of luxury and of adornment are conspicuous by their absence.
To this there is one marked exception, Mulberry Mound, an island shell-heap
near Lake Poinsett, on the southern border of Orange County, rising sixteen feet
from the level of the river. This heap, which since our report in the "Naturalist,"
has again been thoroughly investigated by us during a number of days of April,
1894, when an excavation .16 by 24 by 16-5 feet deep was made, furnished a
bewildering list of objects hitherto undiscovered in the river heaps, including
gracefully shaped arrow heads; sherds decorated with crimson pigment; imple-
ments of bone, including long pins with head decoration; a fragment of a gorget of
shell; shell chisels and drinking cups; a rude hatchet of polished stone; the
human figure scratched upon earthenware;1 a graceful tobacco pipe,2 and other
objects occasionally found in the sand mounds and on the surface of the shell-heaps.
We believe, therefore, that a considerable divergence of time marks the period
of the construction of the shell-heaps. As we have stated, we are unable to deter-
mine whether the makers of the earliest shell-heaps interred their dead in mounds
of sand.
It may be suggested by a close reader of this report that certain sand mounds
have contained no pottery; and that these mounds may have been the work of the
men of presumably the earliest shell-heaps. To this we would reply that, save with
burials, no incentive existed for the inhumation of pottery, whole or in sherds, in
the sand mounds, and that even in the burial mounds it was simply a matter of
custom, almost universally followed, it is true, but still not without exceptions.
We have seen how, in the large mound near Thornhill Lake, but two sherds were
encountered, evidently of accidental introduction, but undoubtedly of original
deposit. The absence of earthenware in a sand mound is no proof that the makers
did not possess it.
It is true we have discovered in Persimmon Mound, a shell-heap, burials
on sand in anatomical order, and in Orange Mound, another shell-heap, beneath
three feet of shell, a small stratified sand mound containing skeletons, and in
neither of these cases were sherds or other objects in association, though in Orange
Mound fragmentary pottery is found with the shell to a certain depth below the
surface. We cannot, however, determine these shell-heaps as positively belonging
to the oldest shell-heaps of the river.

I This unique object is figured and described in the Naturalist, August, 1893.
2 American Naturalist, July, 1894.


That certain shell-heaps were in process of construction at a time when some
of the river mounds were built, we believe to be undoubtedly the case. Certain
varieties of earthenware are common to various mounds and shell-heaps, while
objects similar to all those discovered in Mulberry Mound, with the exception
of the unique incised effigy on pottery, are present in sand mounds. And further-
more, immediately adjacent to Mulberry Mound was a small burial mound of sand
which, as the nearest land is almost two miles distant, we think must be attributed
to the makers of the shell-heap, especially as the earthenware, including fragments
of tobacco pipes of similar pattern, was common to the mound and to the shell-
It must not, however, be assumed from this that neighboring sand mounds and
shell-heaps on the mainland are of necessity of the same period. We have seen
how in the shell base beneath the Tick Island Mound, was discovered a piece of
porous fiber-tempered pottery of a kind not present in the mound itself and never

FIG. 78. Fiber-tempered ware with fret decoration. Shell ridges, Tick Island. (Full size.)

found by us in any sand mound of the river. Yet, after a certain depth, in the
shell ridges of Tick Island this pottery is abundant.
Thus, having shown the connection between the shell-heap at Mulberry Mound
and some of the sand mounds, through objects common to them, we believe our-
selves in a position to trace relationship between Mulberry Mound and a class of
presumably older shell-heaps by means of this species of earthenware.
This porous pottery, which Professor Holmes calls fiber-tempered ware, consists
of clay originally reinforced by vegetable fiber, the destruction of which by expo-
sure to fire in course of manufacture has left small canals. This variety of ware, as
we have stated, is never present in the sand mounds, nor is it found by any means
in all the sherd-bearing shell deposits.of the river. We have met with it notably
at Huntington's, near Cook's Ferry; in Orange Mound, where it begins about one


foot below the surface, continuing from three to four feet down, after which, in the
central portion of the mound, no pottery has been discovered during many careful
excavations; and in the great shell ridges of Tick Island where, beginning at three
feet below the surface, its continuance has been observed through six intervening
feet to the base. This fiber-tempered ware is frequently marked by archaic deco-
ration not found on the pottery of the sand mounds, nor seen on the earthenware
of many of the shell-heaps. Professor Holmes in his interesting paper has pointed
out that this rude ware may have been constructed for shell-heap use alone, and has
suggested that curved decorations frequently found upon it are not characteristic of
earliest types. While admitting the probability of this, we are still inclined to the
belief that the presence of this pottery marks the earlier shell-heaps, though not
the earliest which are characterized by an absence of pottery, the latest class holding
sherds similar to the commoner varieties met with in the sand mounds.
At a depth of about thirteen feet from the surface in Mulberry Mound frag-
ments of this fiber-tempered ware are met with at intervals through the three
remaining feet to the base, indicating, we believe, the abandonment of one class of
ware for another of considerably better material; and thus, as we started to show,
connecting the inception of Mulberry Mound with, let us say, the middle period
of the shell-heaps, a point to which additional probability is lent by the paucity of
relics other than sherds characterizing the lower six feet of Mulberry Mound.
In the shell-heaps of the river we have discovered nothing of necessity con-
necting them with the sand mounds of which Mt. Royal is a type. The beautifully
polished and tapering hatchets, the pottery of erratic design, ornaments and cere-
monial objects of stone have in no instance rewarded our search in the shell-heaps,
and we are compelled, therefore, to consider the question of the contemporaneity of
this class of mounds with any of the shell-heaps an open one, possibly to be settled
by the results of future investigation.
In this connection we would point out, since Mt. Royal and kindred sand
mounds are in close proximity to shell-heaps, that the proximity of sand mounds to
shell-heaps on that portion of the river where shell-heaps are met with may arise
from the fact that these heaps of shell mark the choicest sites which a later people
would settle upon in selecting places of abode long after the abandonment of the
It will be well to bear in mind, moreover, in considering the contemporary
origin of sand mounds and shell-heaps, that the presence of layers and pockets
of shell in many mounds is not conclusive as to the use of shell-fish as food by the
makers of the mounds, since adjacent shell-heap material might readily be, and
doubtless sometimes was, used for purposes of stratification at a period subsequent
to the abandonment of the heaps.
Taking all these facts into consideration, we have arrived at the conclusions
given at the outset of this note, which, for emphasis, we repeat here:-
1. That no evidence so far discovered connects the oldest shell-heaps with the
sand mounds.


2. That while we found no fiber-tempered ware in the sand mounds, the pres-
ence of burials in sand in Orange Mound beneath strata containing this ware indi-
cates at least a beginning of the mound form of burial.
3. That certain shell-heaps probably, and Mulberry Mound certainly, were
contemporary with some of the sand mounds.
4. That absolutely nothing so far discovered indicates a connection between
any shell-heaps and the class of mounds characterized by layers of artificially col-
ored sand, abundance of finely wrought polished hatchets and mortuary earthen-
ware of fanciful pattern.


NOTE.- The chemists, who made analyses given on pages 34 and 39 of Part 1,
now state that LEAD was present in the sulphuric acid used by them. These analyses
are therefore valueless, and are withdrawn, as are remarks on copper on page 35 of
Part I.
Before proceeding to discuss the copper found in various mounds of the St.
John's, we wish clearly to define certain terms as they will be used by us during the
course of this inquiry.
Native Copper.-Native copper is metallic copper found in nature, often con-
taining other elements such as silver, iron, etc.
Copper Ores.-When copper is found in chemical combination with other ele-
ments it forms an ore, as for instance, the oxide, the carbonate, the sulphide, etc.
Melting and smelting.-To melt is to reduce to a liquid state through the
agency of heat.
To smelt is to recover the metal from the ore by the aid of heat, at times the
employment of certain fluxes and the use of carbon in the case of copper. Thus,
it is clear that to aborigines conversant with the melting of copper, the art of
smelting might be unknown.
For the sake of clearness, we shall take up, under different headings, the cop-
per of the river mounds, its workmanship, its composition, and its origin.


With one possible exception, a hawk-bell, a favorite gift and medium of barter
from the time of Columbus, found with iron and glass near the surface of the mound
1 Sincere thanks for valuable assistance are returned to Alexander Agassiz, Esq.; to James R. Cooper
Esq., Superintendent of the Lake Superior Smelting Company; to Professor Frank Hamilton Cushing
to J. B. Eckfeldt, Esq., Chief Assayer U. S. Mint, Philadelphia; to Dr. Persifor Frazer; to Dr. Harry F
Keller; to A. R. Ledoux, M.S., Ph.D.; to G. W. Lehmann, Ph.D.; to Dr. Edward D. Peters, Jr.; to
Professor J. W. Spencer, State Geologist of Georgia; to A. L. Walker, Esq., Consulting Engineer, Old
Dominion Copper Co., Baltimore.
To Professor James Douglas, President of the Copper Queen Mining Company, of Arizona, we are
especially indebted for continuous advice and assistance in the preparation of this paper, and for the loan
of many books and pamphlets otherwise inaccessible.
It is only fair to state that no one of the above-named gentlemen is responsible for all our statements
and conclusions as to copper.


at Dunn's Creek, no article of copper distinctly of European workmanship has been
met with by us in the mounds of the St. John's. Articles of brass, of whose origin
no doubt can exist, are of course not under consideration.


Before proceeding to discuss objects of copper discovered by us in the mounds
of the St. John's, which, it is strongly our belief, are of aboriginal design, we wish
earnestly to call the reader's attention to the admirable paper1 by Professor Cush-
ing, which we regret our space forbids us to quote at length.
In this paper it is clearly shown how the most complicated designs in sheet
copper hitherto brought to the attention of archaeologists can be reproduced with
purely aboriginal tools, and how the sheets can be beaten from native copper with
the aid of annealing. As to annealing we shall speak in another portion of this
The reader of the two parts of our report will recall that from five mounds2 of
the St. John's, whose contents gave no evidence of necessity connecting them with
a period subsequent to the Conquest, numerous objects of copper were taken by us,
including ornaments of sheet copper with various repoussA designs, beads of sheet
copper and beads of wood, shell and limestone copper-coated, jaws of mammals
encased in sheet copper, copper effigies of the serpent and of the turtle, and piercing
implements of hammered sheet copper. These piercing implements, of which the
longest measured 19 inches, seemed upon superficial examination to be wrought of
solid copper. A careful inspection, however, showed them to be made in at least
two different ways; some from strips of thick sheet-copper hammered over on itself
and rounded by hammering, others of many thicknesses of copper in thin encircling
sheets beaten into very close contact.
Now this class of objects, deriving its material from sheet copper of various
thicknesses, represents about all the work in copper so far met with on the St.
John's. We shall, therefore, confine ourselves to it, and not include in this inquiry
articles found in other sections, wrought from masses of solid copper, such as "celts,"
bracelets, heavy breast-pieces, and the like, whose material and manufacture by
cold hammering, so far as we know, is believed by none to be other than aboriginal.
In fact, Professor Putnam informs us that such implements and ornaments have
been reproduced under his direction without the intervention of heat.
The copper work of the St. John's is characterized by'the following features:-
1. The invariable lack of uniformity in size. If the reader will examine the
representations of objects from Mt. Royal and other mounds, it will become appa-
rent that no two articles coincide exactly as to dimensions.
2. The great diversity of shape and ornamentation. In all our mound work
we have discovered no two objects exactly alike. It is quite evident that had a
Primitive Copper Working, American Anthropologist, January, 1894.
2 Grant Mound, Mound on St. Augustine Road, Mound on Tick Island, Mt. Royal, and Mound in
pine woods west of Duval's.


FIG. 75. Ornament of sheet copper, repouss. decoration,
Mt. Royal. (Full size.)

FIG. 76. Piercing implement show-
ing manufacture from sheet cop-
per, Mt. Royal. (Full size.)

FIG. 74. Piercing implement of cop-
per, Mt. Royal. (Full size.)
27 JOURN. A. N. S. PHILA., VOL. X.


supply of staple sizes for commercial purposes been stamped from sheet metal by
the Whites, exact duplicates must necessarily occur in the mounds.
3. Strie in depressed surfaces, giving evidence of the conferring of the design
by pressure and motion as shown in the paper of Professor Cushing.
We are, therefore, of the opinion that aboriginal workmanship on the copper
of the river mounds may be conceded.


But here we are brought face to face with another question. It has been
alleged by some that, granting aboriginal work on copper, the sheets of metal may
have been obtained through White contact, a suggestion too plausible to be dismissed
without serious consideration.
We shall first consider the mechanical evidence of aboriginal work.
The sheet-copper ornaments do not present a uniform thickness, as would be
the case had the metal been rolled or hammered by European artisans and stamped,
the same specimen sometimes decreasing in thickness to almost a cutting edge on
one side. Moreover, a number of plates give evidence of a species of patchwork
where smaller sheets are joined together to furnish one of the requisite size.
While present in a number of cases, this curious aboriginal custom is especially
emphasized in the case of the copper breast-plate found by us at Mt. Royal.
This breast-plate, it will be recalled, was exhumed from the great Mt. Royal
mound at a depth to guarantee original deposit.
We have requested Dr. M. G. Miller, who was present at the discovery of
all the copper met with by us, carefully to examine and to describe the two copper
plates constituting the breast-piece, one of which was figured as frontispiece of
Part I, while the representation of the other occupies a similar position in this
"The two plates of copper composing the chest-piece were each about 10-5
inches square, and when found were separated by a woven vegetable fabric.
The anterior plate, restored in the frontispiece of Part I, unfortunately, is
now in such a fragmentary condition that a complete description of its structure is
impossible. Moreover, a thick coating of carbonate upon the surface of the frag-
ments adds to the difficulty. On certain pieces this coating presents a delicate hair-
like structure, suggesting the idea that the surface had been in contact with the
skin of some animal, but inspection under the microscope shows it at present to
consist of capilliform crystals.
"The decoration is impressed and is regular and well defined. It consists of
seven concentric circles surrounding a central perforation, and a conventional
aboriginal bird's head occupying the space between the circumference of the outer
circle and each corner. Between each head and its neighbor is a diagonally placed
elliptical figure. The edges of the decoration are slightly rounded and not sharply
cut as represented in the frontispiece. The impression varies somewhat in depth,
its maximum being about 2 mm.


"Owing to chemical changes, it is impossible to state the original thickness of
the plate, but it is decidedly thinner than its companion. Variations in thickness
are evident in different fragments, but taken as a whole they represent a sheet of
remarkable uniformity.
"At certain points, during the process of working, the plate had been beaten
or ground through, and here an interesting feature presents itself. To render the
plate fit for use it was necessary of course thoroughly to repair such a defect.
Accordingly the margin of the opening was worked to a fine edge, against the sur-
face of the plate was then applied a thin sheet of copper of size sufficient to cover
the hole, and copper rivets were driven through from -25 to "5 of an inch apart to
hold it firmly in place. The rivets were placed not only along the margin of the
patch but through the body also to fix solidly the superimposed portion of the plate.
The task was finished by pressing the margin of the hole as closely as possible
against the patch beneath, and by working off the rivet ends projecting on that
surface. The delicacy of the work may be judged by the fact that even at the
present day, after long exposure to the action of the elements, it requires the
closest scrutiny, after cleaning the surface with acid, to determine the location of
the rivets and the line of repair. This of course refers to the exposed surface of
the plate, the rivets and patch being clearly evident on the opposite side. The
patch was not cut to fit closely the part repaired but extends well beyond, and
presents a rough, irregular, unworked margin as though the piece had been ham-
mered from a small lump of metal.
"That this work was done before the completion of the plate, and not to repair
holes made during subsequent use, is shown by the way in which the patches par-
ticipate in the decoration.
But this process was resorted to not only for the repair of such defects, but
for the extension of the sheet as well. Along one side of the plate a strip over an
inch wide and several inches long has been attached by riveting along the inner
edge while the marginal portion has been turned upon itself and closely pressed
against the other surface.
"The posterior plate, about twice as thick as its companion, is from 1 to 1-5
mm. in thickness; weighs 15 ounces avoirdupois, and, thanks to its greater strength,
is comparatively well preserved.
The decoration, shown in the frontispiece of this report, is remarkable for
the symmetry of the work. It consists of a central perforation' surrounded by
seven concentric circles, while the space between the outermost circle and each
corner is occupied by an hour-glass arrangement of bars with five parallel bars upon
one side (Fig. 77).
"Certain irregularities, however, are evident. The central elevation (M) of the
figure in the lower right hand corner is oblong, while the corresponding portion of
the other figures is square in outline. The first bar to the left of M is much nar-
rower than the others. The inner half of the hour-glass in the upper figures is
1 Omitted in the illustration.


composed of seven bars, while in the lower figures there are eight in the correspond-
ing space.
"On this plate also the carbonate forms a thick coating, and interferes with a
thorough examination. However, a close scrutiny shows the plate to be made of
two sheets overlapping along the margin. The larger sheet had been split to the
depth of about half an inch near the middle of one edge, and into this slit had been
inserted the edge of the smaller sheet, so that one-half of the margin of each sheet

FIG. 77. Lower plate, copper breast-piece, showing use of rivets, Mt. Royal. (Half size.)

overlapped on one surface and the other half on the opposite surface. In this
position they were riveted together.
"The arrangement is indicated on Fig. 77, in which the line A, B, represents
the margin of the smaller sheet inserted into the split edge of the larger sheet at C.
The heavy portion of the line, A, C, indicates the free margin of the smaller sheet
on the presenting surface of the plate, while the dotted portion, B, C, is the con-


tinuation of the same margin on the opposite surface. The line, D, C, represents
the margin of the larger sheet in the same plane as A, C, while E, C, is the con-
tinuation of the edge on the other surface. At X are the rivets which have been
driven through to hold the sheets together. Those represented are such as could
be detected by a close inspection of the surface, and were subsequently determined
positively by the use of acid. Undoubtedly a thorough cleaning of the surface
would bring others to light.
Here we have an excellent illustration of the production of a plate of desired
size by the process of joining by rivets two smaller sheets, the larger of which con-
stitutes about two-thirds of the finished plate. The same procedure has been fol-
lowed in the production of several of the smaller ornaments in the collection, while
in no instance is there evidence of either soldering or brazing, methods which one
would certainly expect to note were the articles of European origin.
At several points on the surface, H, K, L, where a tendency to exfoliation was
evident, rivets were used to hold the loosened edges in place. This is especially
marked near the margins and in the corners. The letters indicate only the most
evident of the rivets, many others undoubtedly being hidden under the coating of
"The exfoliation, it will be observed, arose not after long exposure in the
mound, but during the process of construction of the sheet as evidenced by the
rivets. This condition is frequently observed in hammered masses of copper, and
in copper ornaments and implements obtained from mounds, and may aid in deter-
mining such articles as of aboriginal origin.
"At several points on the depressed surface of the ornamentation may be seen
striae, which have probably come from the rubbing of some implement used to force
the copper into corresponding depressions of a mold.
In finish, this plate does not equal its companion, though both surfaces had
received a certain amount of attention. The line of union of the two sheets can
be readily followed throughout the greater part of its course; the conjoined edges
are irregular, and were not so carefully pressed into the underlying sheet."
From this description we are of the opinion that these plates, with their
ingenious system of repair and extension' by the aid of rivets, may not under any
circumstances be attributed to the handiwork of artisans of Europe.


In estimating the epoch of construction of mounds, the mounds themselves
are the most reliable witnesses, and any class of objects found therein is best known
by the company it keeps. By consulting the detailed descriptions of the copper-
bearing mounds of the St. John's given in this report, it will be seen that the objects
of copper were from the base and the body of the mound, associated with products
distinctly of aboriginal art, such as polished hatchets of stone, vessels of earthen-
Professor Putnam has referred to the use of rivets for repair, and recently has discovered the ex-
tremities of a bracelet joined by rivets. See, also, Fifth Annual Report, Bur. Eth., page 99.


ware, and ornaments of shell, but never with articles indicating a knowledge of
Europeans as exemplified by implements of iron, beads of glass, brass, pewter, and
glazed pottery. It is, indeed, a hypothesis not to be entertained that later Indians
in possession of these products of the Whites, through gift, barter, or plunder, in
these mounds interred European copper alone, and sedulously refrained from placing
with the dead, iron axes, glass beads, or the paint of commerce, and many kindred
articles, all of which they so highly prized, and all of which are found on the St.
John's with the superficial burials.
As we have pointed out in our note on the probable comparative age of the
river mounds, archeologists are agreed that when in a class of mounds no article
of European manufacture is met with other than superficially, these mounds may
be classed as of a period antedating the coming of the Whites, and Professor Put-
nam has so clearly expressed this view that we quote here a portion of a personal
letter from him.
"Just after I wrote my little paper on copper in the Museum as the beginning
of a series of papers on the use of metals, copper began to come in from our Ohio
explorations in a wonderful manner, until we now have copper in such abundance
that a paper on the subject would be a volume. We have it hammered and cut into
all manner of shapes-implements and ornaments-and with it have come several
lots of ornaments made of meteoric iron-implements and ornaments-and also
considerable silver (ornaments) and a little of gold. All these metals are ham-
mered and cut, and we have the copper in all stages from the rough nuggets, through
those partly hammered to the sheets and the objects cut from them. To consider
this the work of Europeans is an absurd perversion of the facts before us; and yet
just because the facts do not agree with the theories of some who would have all
facts drop into their theories, or else throw them out of consideration, these objects
are spoken of as unquestionably of European origin, traded to our old mound
building people of the Ohio valley by Whites since the settlement of the country.
"I am confident that you are right in your conclusion, when, after the careful
examination you made at Mt. Royal, you did not find anything derived from the
white man, such as glass beads, brass, etc., that all the work was native, the copper
plates included.
"I have explored several sites (villages and burial) which were known histori-
cally, and I have always found glass beads, brass kettles, pewter mugs and plates,
brass buttons and iron knives and axes, etc., etc., in the graves or in the refuse piles
and hearths of the wigwams, mixed with native objects, and when, after a thorough
and extensive exploration of any place such objects of European origin are not found,
we have no right to doubt as native what we do find, no matter what peculiar things
may occur.

We are, however, not compelled to base our conclusions as to the origin of the
copper solely upon the testimony we have adduced. In the investigation of copper


found within the limits of the United States, for some reason, analysis has been
hitherto entirely overlooked, and we shall now endeavor to throw some light upon
copper from a chemical point of view.
Comparative Purity of Copper.-Before proceeding to discuss the copper of
the mounds, early post-Columbian copper and native copper, it will be necessary
for the reader to have a clear idea of the percentage of pure copper usually found
in these commodities, that he may definitely draw his conclusions.
Unfortunately, so far as the presentation of a striking case is concerned, the
difference between the purest of smelted copper and the most impure is relatively
small, its range not greatly exceeding two per cent., thus eack tenth of one per cent.
is of marked importance.
At the present time, with the most approved methods of smelting, a copper 99
per cent. pure1 is considered of inferior grade, good commercial copper averaging,
let us say, 99-5 per cent. pure.
On the other hand, opinions of many experts, backed by results of analysis,
lead us to conclude that in early post-Columbian times commercial copper of a
much higher degree of purity than 99 per cent. was not produced in Europe. As
we shall see, lead was at that period intentionally introduced during treatment,
while arsenic and silver, loath to part company with copper, were not successfully
treated by the processes of those days. In fact, it is asserted that in recent years
old German copper coins have been profitably remelted for their silver, and yet it
was in Germany that smelting processes were best understood during early times.
It is evident, then, that a copper purer than the average of that produced under
modern methods of smelting cannot have been derived from Europe during the six-
teenth or seventeenth centuries.
The reader must bear in mind that analyses of copper from native metal, as
given in works on metallurgy and the like, are not results obtained from specimens
of mass copper prior to melting, but analyses made from ingots of cast metal which
may have lost proportions of certain elements during treatment, or received others
through accidental or intentional introduction. We have been unable in any work
to find the result of a single analysis of native copper made prior to the ingot
state, and have therefore submitted to various chemists a considerable number of
authentic specimens from different sections.2 The percentage of pure copper in the
native state previous to treatment is from 99p65 to 99"994.
The Copper of the Mounds.-So great a period of time has elapsed since
the deposit of copper in the mounds that many of the sheets are carbonated and
oxidized through and through, rendering absolute determination as to purity impos-
sible. Where, however, a part of the metal remained we have been able to arrive
at definite results.

SAt the present time "India sheets" and "brazier sheets," of a very low grade of copper, are shipped
from England to India and China.
2 Since writing, our attention has been called to a statement by Rammelsberg (Mineralchemie, page
5) as to the impurities in Lake Superior copper. He reports iron 0-31 per cent. and silver entirely absent!


A portion of a piercing implement of hammered sheet copper from Mt. Royal,
Florida, was submitted for analysis to Booth, Garrett and Blair, with the following
"Copper 99-897 per cent.
Lead None.
Bismuth None.
Iron 0057
Gold None.
Silver 00012 "

99-9039 "
They state that the sample was much corroded and although cleared with acid
before analysis there was probably a little oxide remaining which they did not
attempt to determine.
A second and smaller fragment of the same implement was submitted to A. R.
Ledoux, M. S., Ph. D., whose determination was as follows:-
"This sample consisted of a pure copper core coated with a film of oxide and
carbonate. We removed the film by dissolving in dilute acid until only the unoxi-
dized core remained. This was analyzed and showed:-
"Copper 99-85 per cent.
Silver Trace.
Iron Trace.
"No arsenic, antimony, lead, tin, zinc, nickel or cobalt is present. This is a
very pure copper indeed."
The reader's attention is especially called to the great purity of this copper
and to the absence of arsenic and antimony; and in this connection we introduce a
portion of a letter from Dr. A. R. Ledoux, who has had the widest experience in
such matters: "I now authorize you, if you wish, to quote me for publication to
the effect that in my opinion the smelting processes of two or more centuries ago
could not have turned out as pure an article as this from copper ores, and that in
my opinion the piercing instrument was made from native copper of exceptional
purity as is found in some of the Lake Superior mines and at a few other points."
A number of fragments of various ornaments of sheet copper from all parts of
Mt. Royal mound were analyzed as a whole by Ledoux and Company, with the
following result:-
"Arsenic 0038
Iron 0170
Nickel and cobalt 0030
Silver Trace.
NOTE : The sample contains very little metallic copper-not sufficient to make
a satisfactory determination-consisting practically of a crust of carbonate and
oxides of copper.


"The sample was specially examined for antimony, tin, lead, bismuth, and
zinc, none of which was found present.
"The above figures are on the basis of the sample as received."
The metal used in this analysis was so altered that we must not regard the
result as quantitative in relation to the copper as it originally existed. It is sub-
mitted as a qualitative analysis to show presence or absence of certain elements.
We have referred in another portion of this report to the discovery of articles
of copper in the great Grant Mound, near Mill Cove, Duval Co., Florida. These
objects, all in good state of preservation, we were unwilling to sacrifice for purposes
of analysis, and are, therefore, doubly grateful to Mrs. Martha A. Millspaugh,
whose residence is near the foot of the mound, for a piercing implement of copper,
8-25 inches long and -12 of an inch in thickness, apparently made from a strip of
thick sheet copper, and a fragment of another piercing implement 4 inches in
length, with a thickness of -2 of an inch, made, as the reader will recall, from
numerous thin sheets of copper closely hammered together. These objects were
taken from the mound by a relative of Mrs. Millspaugh.
According to the analyses of Ledoux and Company the metal of these objects
had the following composition:-
Fragment of piercing implement of sheet copper, Grant Mound, Florida.
[The larger specimen.]
"Copper 99-730 per cent.
Iron 0-034 "
Silver 0023 "

NOTE : The sample was examined for lead, bismuth, arsenic, antimony, nickel,
cobalt, none of which was present.
"The sample also contains some oxygen, as it was impossible to entirely re-
move the oxide of copper from the surface of the sample. This oxygen is due to
superficial oxidation, and is not an essential ingredient of the metal."
Fragment of smaller piercing implement from Grant Mound, Florida.

"Copper 99720 per cent.
Iron 0044
Silver 015

NOTE: The sample was examined for lead, bismuth, arsenic, antimony, nickel,
cobalt, none of which was present.
"The sample also contained some oxygen, as it was impossible to entirely re-
move the oxide of copper from the surface of the sample. This oxygen is due to
superficial oxidation, and is not an essential ingredient of the metal."
We have thought it well to include with these analyses, for comparison, results
obtained from copper from mounds other than of Florida.
We are indebted to Thomas Wilson, Esq., of the National Museum, for a num-
ber of fragments of sheet copper "from grave A, Little Etowah Mound, Georgia."
28 JOURN. A. N. S. PHILA., VOL. X.


These fragments, submitted to Booth, Garrett and Blair, were reported as fol-
"Gold None.
Silver 0013 per cent.
Lead None.
Copper 75-050 per cent.
Silicious matter 0480 "
75-543 "
"This sample contained but a very little amount of copper in the form of
metal, the great mass being either oxide or carbonate.
Since we found no other metal present' in the sample than those given above,
we feel justified in saying that the original metal was nearly pure copper, over 99
per cent."
To those who have examined or have read of the wonderful deposit of copper
ornaments, etc., in the Hopewell Mound, Ohio, the results of thorough analysis will
be of peculiar interest. We are indebted to Professor Putnam for a fragment of
a 'breast-plate' hammered and cut, a fair representative of the copper of the
This specimen was exhaustively examined by Ledoux and Company, who
write as follows:-
The piece of ornament from mound in Ohio submitted to us for analysis con-
"Copper 96-3100 per cent.
Antimony 00070 "
Silver 00450 "
Nickel and cobalt 0060 "
"NOTE: The sample was carefully examined for arsenic, tin, lead, bismuth,
and zinc, none of which is present. This analysis was made by removing adhering
oxides and carbonates as far as possible, but the sample contained sand adhering to
it and penetrating it to such an extent that it was impossible to remove it all.
This accounts for the low percentage of copper; the metal would have undoubtedly
run over 99 per cent.2 if freed from mechanical impurities and oxygen."
Warren K. Moorehead, Esq., has kindly furnished us with an implement of
copper from an Ohio mound which yielded to a partial analysis by Booth, Garrett
and Blair:-
"Gold None.
Silver 0011 per cent.
Lead None.
Bismuth None.
Copper 99-678 per cent."
In this case no search was made for other elements.

1 The italics are ours.
SThe italics are ours.


During the present summer (1894) Gerard Fowke, Esq., has obligingly con-
ducted for us investigations in a "double mound" on the farm of Mr. J. M. Van
Meter, three "miles south of Piketon, Pike County, Ohio.
With human remains, below the base, in a central position in the mound, were
five sheets of copper averaging in size about 3-5 by 2-5 inches. One of these
sheets, forwarded to Ledoux and Company, was reported upon as follows:-
"Copper 99-9130 per cent.
Silver 00198 "
Arsenic 00026 "
Antimony Trace.
Iron 00233 "
Nickel and.cobalt 00080 "
"NOTE: The above analysis was made after removing the superimposed film
of oxides and carbonate from the sample. Special examination for lead, bismuth
and zinc shows that none of these is present."
From these analyses of copper from the mounds we note the very higk percent-
age of copper, the constant occurrence of silver and of iron, the totat absence of lead,
the occasional presence in minute quantities of arsenic, of antimony, aud of nickel
and cobalt. In no case, however, does the amount of any impurity present, with the
exception of silver and of iron, greatly exceed a trace.
Copper after White Contact.-It is a noteworthy fact that in a great majority
of cases where the discovery of copper with articles giving evidence of White con-
tact is reported, the metal is in reality brass.
During all our researches on the St. John's in but one case have we found cop-
per (and this was possibly bronze) in association with glass, iron and other products
of Europe.
In the mound at Bayard Point, where alone on the river we have found, other
than superficially, articles distinctly European, brass was met with, but no copper.
Professor Putnam, with the collections of the Peabody Museum to draw upon
and with the warmest desire to aid us, has been able to furnish us brass alone from
post-Columbian Indian village sites and graves.
From a post-Columbian Onondaga village site, Rev. W. M. Beauchamp, the
well-known authority, kindly forwarded us a number of ornaments, etc., all of
which were likewise of brass. We have it from this gentleman and from other
sources that the copper arrow heads, so called, are of the same material, as also
are the kettles usually spoken of as copper.
"Wood, in his 'New England Prospect,' published in 1634 (page 90), distinctly
states that the Indians obtained brass from the English for their ornaments and
arrow heads."1 We see, then, that much that has heretofore been considered cop-
per is in reality brass, and that brass and not copper was as a rule furnished the
Indians by the Whites, the natives being quick to take advantage of the superior
character of the alloyed metal.
1 Squier, Aboriginal Monuments of the State of New York, page 188, et seq.


There is little doubt that we must look to Europe as the source of supply of
such copper as came to this country during the possible post-Columbian mound
building period.
Now the sulphide ores of copper are universally distributed throughout the
world, supplying more than four-fifths of the copper in demand, while Europe has
no great available district of wholly native copper as is found in the Lake Superior
region of this country.
Of the ores of copper, the sulphide most stubbornly resists treatment, and in
association with it are almost invariably certain elements, arsenic, antimony, and
sometimes bismuth, whose absolute elimination even at tkis day it is impossible to
accomplish by any process of smelting. In fact, the comparatively recent electro-
lytic process does not appear to entirely eliminate these elements when present.1
Even in the more amenable oxide ores from certain districts, a trace of arsenic,
after refining, still remains, not sufficient, it is true, to interfere with the drawing,
the rolling and the stamping of the metal, but nevertheless distinctly recognizable,
and this we mention to disabuse the reader of the idea, prevalent in certain
quarters, that oxide ores are necessarily non-arsenical. It is evident then that
early European copper for commercial purposes must have contained ponderable
quantities of arsenic or antimony or of both, with occasional presence of bismuth, if
at the present day, with improved methods, these elements are to a certain extent
present in the metal.
At the time of the discovery of America, the extraction of copper in Europe
was practised by the Germans and the Italians, whose supplies were almost exclu-
sively sulphide ores. Later the German processes were established in Wales.
Professor James Douglas, of New York, whose unceasing kind offices have so
materially aided us in this paper, has placed in our hands a work2 containing much
original and curious information relative to the introduction of smelting processes
in Great Britain.
From this work we learn that the smelting of copper in the Swansea district
(where fully nine-tenths of the copper of Great Britain is said to be reduced from
the ore) was begun at Neath, in the year 1584; at Swansea, 1717-1720; at Taibach,
1727, etc.
On page 25 et seq. of the same work we find a curious communication regard-
ing the earliest English smelting, containing references to the presence of sulphur,
showing the ores to be sulphide, and allusions to the roasting of ore, which is not
practised in the case of oxides.
As to antimony and arsenic we shall quote the quaint wording of the original:
"And that water doth not only drawe the vitriall and coppris from the ure, but also
divers other hurtfull humors, being by nature enemyes to the Copper; as arsenick,
1 "On the Analysis of American Refined Copper," by Harry F. Keller, Ph. D., Journal of the
Franklin Institute, July 1894, page 54.
SThe Smelting of Copper in the Swansea District of South Wales, from the Time of Elizabeth to
the Present Day; by Col. Grant-Francis, F. S. A. Second edition, London and Manchester, 1881.


sulpher, antimony, allome, and ironn." And again, "The ij corrupt humor is
Arsineque, by nature a kinde of poyson, being in like manner a mineral substance,
wilbe consumed w'th fire in to Smoke, w'ch is a vere dangerous ayer or savor, and
by his force maketh the copper white and brother then the sulpher doeth. This
Arsenieque is not only in great quantitie in our copper ures, but is by nature so
forceable of it self, that it is Lorde and Ruller over all the rest, and consumes both
ye sulpher, and antimony, so y't thei ar not to be seene."
We have devoted considerable time during the preparation of this paper to the
investigation of the range of impurities in the German and English copper of com-
paratively recent times, since it is evident that all impurities found in this copper
at the present time must have existed to a greater extent at an earlier period.

Of many recent analyses of German copper we give three

99-325 per cent.
0-012 '"
0-1166 "

99-357 pe

representative ex-

Scent. 99-6125 per cent.
" 0-0292
" None.
S 0-0172 per cent.

S 0-0200
" 0-0039


I. and II. Copper from Oker. Analyses by W. Hampe, and given in his
" Beitrdge zur Metallurgie des Kupfers," Zeitschriftfiir Berg-, Hiitten- und Sa-
linenwesen, Vol. 21, page 252.
III. Mansfeld copper (from the Saigerhiitte at Hettstedt). Analysis made
under Hampe's direction. Loc. cit., p. 255.
In Percy's Metallurgy1 we have various analyses of modern copper of commerce.
In twenty-nine specimens of English tile, best selected, sheet copper, etc., all
but two, in which traces are present, contain ponderable quantities of arsenic.
Antimony is included in traces in eleven instances and in eight by weight. We
find bismuth in twenty-five of these analyses in quantities sufficient for exact de-
termination, and traces in the remaining four.
Copper ores, mainly sulphides, are found in various parts of Italy, though most
of the copper used by the Romans was obtained from the Island of Cyprus.
According to G. Jervis (I tessori sotteranei dell' Italia, Turin, 1874, Vol. II,
page 439), the mines of Monte Catini in the Province of Pisa were worked about

SFrench translation, Vol. V., pages 383, 384.



the middle of the fifteenth century. Toward the close of the century, in 1494,
they were abandoned, but reopened early in the sixteenth century and operated
intermittently. According to Jervis these are among the richest copper mines of
Europe. The ores are sulphides.
Another copper mine of importance is that of Ravamonte, near Agordo, in the
Venetian Alps, which was operated, according to Jervis (op. cit., Vol. I, page 332),
so far back as the fifteenth century. The process employed there for the extraction
of copper is described by Haton, Percy, and Rivot.
Percy (Metallurgy, Vol. I, page 439) says: "The process appears to be of
comparatively ancient date; but I have not been able to trace its history with cer-
tainty. At the present time (1861) it appears to be carried on with great skill at
Agordo, where it is stated to have been first introduced in 1692 by a Prussian of
the name of Weyberg."
The copper, according to Schnabel's Metallhiittenkunde, contains :-
"Arsenic 064 per cent.
Antimony 0.04 "
Lead 020 "
Silver 010
Sulphur 004
We shall not occupy space with farther analyses of Italian copper, the impuri-
ties of which, as a product of sulphide ores, the reader is by this time in a position
to divine.
Now let us consider results of analyses of copper used by Indians after contact
with the Whites.
We are indebted to David Boyle, Esq., of the Canadian Institute, Toronto, for
a fragment of copper taken by him from a grave of the Tobacco Hurons, near that
place. In these graves are articles unquestionably of European origin.
This fragment submitted to Dr. A. R. Ledoux, was analyzed and reported upon
as follows:-
"Copper 98-970 per cent.
Silver 0084 "
Iron 0057
Arsenic 0160 "
Antimony 0130 "
Nickel and cobalt 0070 "
Lead 0300 "
"The sample contains no bismuth, tin or zinc. The low percentage of copper is
partially due to oxygen which it was impossible to remove entirely from such thin
samples. This copper is, as you will note, of inferior grade."
The considerable collections of post-Columbian Indian relics of Mr. W. W.
Adams, of Cayuga Co., N. Y., largely made on the spot by him, contained no
specimens of early post-Columbian copper other than ears of copper on two kettles
of brass, and in one instance copper rivets joining together two sheets of brass


which had probably formed part of one of those kettles which archaeologists usually
incorrectly describe as of copper.
The copper ears of a kettle from a Cayuga Iroquois grave, Venice, N. Y., found
in association with glass beads and other articles of European manufacture, yielded
to the analysis of Ledoux and Company the subjoined result:-
"Copper 98-520 per cent.
Lead 0362 "
Silver 026
Antimony 136 "
Arsenic 0127 "
Iron 0016 "
Nickel and cobalt 0211
Bismuth Trace."
The copper rivets from Iroquois brass kettle, Fleming, N. Y., afforded too small
a quantity of material to permit a complete quantitative analysis. Ledoux and
Company report upon them as follows:-
"Copper .. 97-03 per cent.
Lead Present.
Silver Present.
Iron Present.
Cobalt Present.
Nickel Present.
Arsenic Present.
Antimony Present.
Bismuth None.
"NOTE: This is an exceedingly impure copper. The above elements, while de-
termined only qualitatively, are present in very considerable quantities, apparently
greater than in any of the samples we have hitherto examined for you. The lead
is especially high."
We regret that, after many months of earnest endeavor, including extended
correspondence with those in all parts of the country likely to have it in their
power to aid us, we have been unable to secure other specimens of European copper
which had seen service with American aborigines, though abundance of brass was
We shall now take up the important question of the presence of lead in
European copper. The introduction of lead during treatment was included in the
German method of smelting and with this method was adopted in England. Es-
pecially was lead a feature in sheet and drawn copper.
The translators of Percy's Metallurgy,1 in summing up the results of many
analyses, inform us that lead constituted an almost invariable element in wire and
sheet copper, basing their conclusion on the results of a considerable number of
analyses of English sheet and drawn copper, all of which show the presence of
lead, and we may add here that all similar works consulted by us, which cover the
1 Loc. cit.


ground to a period, say thirty years back, refer to the union of lead with copper
used for the production of sheet copper and wire. Moreover, various experts con-
sulted by us, consider that, while the introduction of lead was not invariable, its
absence from any considerable number of specimens is good negative evidence
against a European origin in early times. Even at the present day, many analyses
of copper show the presence of lead, though sheet copper from certain sections or
where the material has been obtained by electrolytic deposit, may give no evidence
of it. But such recent copper, of course, is not germane to our subject.
Lead has never been discovered in copper from the mounds nor is it ever present
in native copper.
We shall now examine results of partial analyses made for us, bearing strongly
upon this point.
A number of fragments of sheet copper from various parts of Mt. Royal, sub-
mitted to Booth, Garrett and Blair, furnished the following result:-
"Silver 00315 per cent.
Gold None.
Lead None.
"The remainder is chiefly red oxide of copper.
"Assuming that all the copper is in the form of red oxide, the percentage of
silver would be 0-0372 in the copper in the melted state."
Several fragments of thin sheet copper from the mound west of Duval's,
Lake County, Florida, were submitted to Dr. Ledoux with a view to determination
as to lead. The result was negative.
General Gates P. Thruston has kindly forwarded us several small fragments
of sheet copper from the stone graves of Tennessee, in 15,000 of which, we are in-
formed, no European object has ever been brought to light. Copper is far from
abundant in these graves, and we are especially indebted to General Thruston for
the sacrifice of his specimens. Unfortunately, the thin sheet copper was com-
pletely oxidized; lead could not be detected by analysis.
Professor Putnam, for whose deep interest in this investigation we must again
express our gratitude, has placed at our disposal a fragment of a copper breast-
plate from a Tennessee stone grave." A partial analysis showed the presence of
silver, the absence of lead.
We are indebted to Professor Putnam also for a number of specimens of cop-
per from Ohio, which are reported upon by Booth, Garrett and Blair as follows:-
B. Piece broken from nugget of copper found with several others in a
mound in Ohio:"-
"Silver Present.
Lead None."
"C. Fragment of a copper ornament from a mound in Ohio:"-
"Silver Present.
Lead None."

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs