AN ACCOUNT 
OF THE 
FRESH-WATER SIIELL-IIEAPS 
OF THE 
ST. JOHNS RIVER, EAST FLORIDA. 
B Y 
JEFFRIES WYMAN, M.D. 
REPRINTED FROM THE AMERICAN NATURALIST, VOL. II, NOS. EIGHT AND NINE. 
SALEM, MASS.: 
ESSEN INSTITUTE PRESS. 
1 8 68. ON THE 
FRESH-WATER SHELL-HEAPS 
OF THE 
ST. JOHNS RIVER, EAST FLORIDA. 
The St. Johns River, on the banks of which are to be 
seen the mounds described in the following pages, has, in 
several respects, a peculiar interest. It rises near the mid- 
dle of the eastern half of the peninsula of Florida, in two 
series of lakes and swamps of great extent, one of which 
finds its outlet through the upper portion of the main stream, 
and the other through the Oklawaha, the largest of its tribu- 
taries. These waters are separated by land scarcely rising- 
above their level, from another chain of lakes and swamps 
which have an outlet southwards through the Kissimmee, 
and thence into the great lake of Okee-Chobee, which has an 
area of about eight hundred square miles. Other waters, 
starting from the same region as the preceding, but separated 
from them by a low range of sand-hills, are discharged west- 
wards into the Gulf of Mexico, chiefly through the Withlo- 
kootchee. Though extremely crooked, the general course 
of the St. Johns is somewhat to the west of north, and in its 
various windings is supposed to traverse a distance of three 
hundred miles. Its frequent enlargements, as at Lake Har- 
ney, Lake Monroe, Lake George, and its great breadth 4 
FRESH-WATER SHELL-HEAPS 
from Palatka to its mouth, almost justify the designation of 
it as a chain of lakes rather than a river. Flowing through 
© © 
a region which is nearly of a dead level, its stream is neces- 
sarily sluggish. 
There is much dry and arable land, but so little is this 
raised above the level of the river, that, were it depressed 
live or six feet, the ocean would reassert its sway over a 
large part of the eastern portion of the peninsula, leaving 
only narrow ridges along the sea-coast, and inland, here 
and there low islands. As it is, immense tracts are under 
water throughout the year, and the whole area drained by 
the St. Johns is a combination of dry land, swamps, lagoons, 
and creeks. Open prairies, pine barrens, palmetto ham- 
mocks, mixed forest growths, chapporals of saw-palmetto, 
thick jungles, and large tracts overgrown with tall reeds or 
rank grass vary the surface. From the absence of a change 
of level in the land, the distant views on the river are ex- 
tremely monotonous, while the near ones are often of great 
beauty, because of the windings of the river, and the sub- 
tropical vegetation. The creeks and lagoons, with their rank 
vegetation, and also the wilder shores of the river, shelter 
vast numbers of water and shore birds, and also countless 
alligators, water moccasins, frogs, and other reptiles. 
Of animals suitable for the food of man there is an abun- 
dance, as will be seen farther on, so that along the banks of 
the river and its tributaries, hunter-life could be easily sus- 
tained. The aborigines were, however, planters as well as 
hunters, for the hrst explorers found the land largely tilled, 
and the "Indian-old-helds” which can still be traced bear 
witness of the fact. Of all the American races none appear 
to have occupied a region more nearly equally divided be- 
tween land and water, or one which had been more newly 
lifted above the level of the ocean, than natives of the shores 
of the St. Johns.* 
* For a description of the physical features of the St. Johns River, the reader is re- 
ferred to an article entitled Cursory Remarks on East Florida. By Major Henry OF ST. JOHNS RIVER, EAST FLORIDA. 
5 
The shell-heaps we are now to describe were visited dur- 
ing the months of February and March, 1867, in company 
with Mr. G. A. Peabody, of Salem, Mass., and Mr. George 
H. Dunscombe, of Canada West, to both of whom the writer 
is largely indebted for aid in making explorations and for 
valuable contributions to his collections. The heaps tire dis- 
tributed over a distance of more than one hundred and fifty 
miles, between Palatka and Salt Lake, and are nearly all 
situated on knolls, seen here and there on the borders of the 
river, though a few are built in swamps or on dry land, at 
some distance from it. They tire composed almost exclu- 
sively of one or more of the following species of shells, 
namely, Ampullaria depressci of Say, Paludina midtilineata 
Say, and TJnio Buckleyi Lea. Besides these, a species of 
Melania and a few Helices are found, but they, as well as a 
few marine shells, must be considered as accidentally pres- 
ent. The mounds vary much in size, from circular heaps 
fifteen to twenty feet in diameter, and a few inches high, to 
long ridges several hundred feet in length, and having a 
height from a few inches to four or five, and in some cases 
to fifteen feet. They are generally overgrown with oaks, 
maples, palmettos, bays, magnolias occasionally, and other 
forest trees, and not unfrequently with groves of the wild 
orange. The last, bearing a fruit both bitter and sour, has 
been supposed to be indigenous; but it would appear from 
the researches of Mr. G. R. Fairbanks, a gentleman thor- 
oughly versed in the history of the peninsula, that they 
were introduced by the Spaniards.* We personally visited 
more than twenty-five of these heaps, but only a few of them 
Whiting, U. S. A. American Journal of Science, Yol. XXXV, p. 47. To those who wish 
an introduction to, and a digest of, the literature pertaining to the whole State, the ex- 
cellent work, Notes on the Floridian Peninsula, by Daniel G. Brinton, A. B., Phila- 
delphia, 1859, is invaluable. This work also contains an account of the author’s own 
investigations of the shell-heaps of the sea-coast. 
* Mr. Fairbanks has observed that they are confined to the best camping-places on the 
river, and it does not appear that they are described or referred to in any of the Span- 
ish records, which it is presumed they would have been, had the Spaniards found them 
there, since they so particularly mention other fruits. They are probably the Seville 
Orange run wild. FRESII-AVATER SHELL-HEAPS 
will be described, as they are nearly all essentially alike; 
an enumeration of the whole series will, however, be given 
at the end of the article. 
The mounds of oyster-shells on the sea-coast of Florida 
have long since attracted attention ; some of them have been 
described by Dr. Brinton, who has clearly set forth the 
grounds for the conclusion that they are of human origin.* 
The fresh-water shell-heaps have received but comparatively 
little notice, and have generally been supposed to be either 
fhiviatile or lacustrine deposits, for which any one might cer- 
tainly be excused for mistaking them at first glance. That 
they are the works of man the following observations are in- 
tended to show. Count Pourtales, however, visited the shell- 
heaps at Old Enterprise, Lake Monroe, in 1858, when he 
obtained from among the shells fragments of pottery, and of 
the bones of animals. He has not published an account of his 
observations, but informs me that he came to the conclusion 
that this mound was artificial. 
The existence of shell-heaps in other regions consisting 
of the remains of fresh-water species, though from time to 
time noticed, have not been generally recognized. The first 
observation that we have seen with regard to them is by 
Atwater, who described mounds of mussel-shells on the 
banks of the Muskingum River, containing various articles 
of human workmanship, f Dr. Brinton, while connected with 
the Army of the Cumberland in the war of the rebellion, ob- 
served mounds of mussel-shells which had served to supply 
food to the Indians and during the last year the writer, in 
company with Mr. Ralph Waldo Emerson, Mr. Elliot Cabot, 
and others, examined a similar deposit on the banks of the 
Concord River, in Massachusetts, consisting of Unio compla- 
natus, and containing charcoal, pieces of worked bone and 
flint.§ I am also informed by Professor J. D. Whitney, the 
* Notes on the Floridian Peninsula, p. 166. 
f Archseologia Americana, Vol. I, p. 226. 
J Smithsonian Reports, 1866, p. 356. 
§ Proceedings of the Boston Society of Natural History, Vol. XI, p. 243. OF ST. JOHNS RIVER, EAST FLORIDA. 
7 
chief of the Geological Survey of California, and Dr. William 
II. Brewer, botanist to the same survey, that vast numbers 
of fresh-water shell-heaps exist there. Indeed there is an 
abundance of evidence for the belief that they are widely 
scattered throughout the United States. 
I. SHELL-HEAPS. 
King Philijp’s Town. This place was in a wild state until 
quite recently, and derives its name from a Seminole chief, 
who, it is said, once occupied it. The shell-heap, now con- 
verted into an orange grove, is on the left bank of the river, 
about a mile below the outlet of Lake Harney. Its situation 
is favorable both for hunting and fishing; the river is here 
sixty or seventy yards wide ; opposite is the mouth of Deep 
Creek,* rising far to the eastward, and pouring into the St. 
Johns an excellent quality of water; to the rear and west- 
ward are open prairies and pine lands, and in the distance, 
to the north, is a large lake. The river contains an abun- 
dance of fish, but generally of a poor quality, except in the 
month of February, when vast numbers of shad pass on their 
way to Lake Harney, two hundred miles from the mouth, 
to spawn. While we were encamped here, the splashing of 
the water by shoals of these fish could be heard at all hours, 
from evening twilight to early dawn. 
The shell-mound is about four hundred and fifty feet in 
length, and from a hundred to a hundred and twenty in 
breadth. It stretches at right angles to the river, borders a 
lagoon on the south, and on the north merges into cultivated 
fields, over which its materials have become somewhat scat- 
tered. Its greatest height is about eight feet. Fragments 
of pots may be picked up anywhere on the surface, and, with 
these, bones of various edible animals. As all such remains 
may have been deposited on the mound after its completion, 
excavations were made at many points from a few inches to 
* There is another creek of the same name which enters the St. Johns on the right 
hank, between Pilatka and Picolata. 8 
FRESH-WATER SHELL-HEAPS 
several feet in depth, to ascertain if similar objects were 
buried in its interior. The most unequivocal evidence that 
this mound, while in the process of formation, was occupied 
by the aborigines, was obtained from a pit between four and 
five feet in diameter, and from five to six feet deep, which 
was dug near the centre. Not only were fragments of pots 
and bones found at all depths, but at a depth of three feet 
the remains of an old fireplace were uncovered, consisting 
of a horizontal layer of charcoal, beneath which were per- 
fectly calcined shells, and near these others more or less 
blackened with heat. Still farther oft’ were fragments of the 
bones of deer, of birds, turtle and fish, all just as they would 
naturally have been left around a fire, where cooking had 
been for some time carried on. In addition to the above 
statement it may be mentioned as a matter of negative evi- 
dence, that not a single article was discovered which could 
have been attributed to the white man. Several excavations 
made in other portions of the mound yielded similar results. 
Black Hammock. One of the largest shell-heaps on the 
St. Johns is to be seen here. It is situated on the borders 
of a large lagoon, on the left bank of the river just above 
the outlet of Lake Jessup,* and seven miles above Lake 
Monroe. Besides the principal deposit of shells, there are 
two smaller ones. At the westerly end is the first, a few 
inches thick, extending one hundred and fifty feet along the 
shore, and some thirty or forty feet inland. This is sepa- 
rated from the rest by a small watercourse, the outlet of a 
morass. The shore then takes a northerly direction for 
about two hundred feet, and consists entirely of sand; at 
the point where the shore again takes an east and west di- 
rection, is a second but smaller deposit, extending only a 
few feet to the eastward. One hundred and eighty feet from 
the point just mentioned is a small burial-mound, and a 
*This lake was discovered by Lieutenant Peyton, of the U. S. Army, during the 
Florida war, and at first bore his name, which ought not to have been discontinued. It 
were better to preserve the Indian names if they can be learned, but if not, no one has 
a right to usurp a name which has been given by others in honor of the discoverer. OF ST. JOHNS RIVER, EAST FLORIDA. 
9 
little more than a hundred feet from this begins the largest 
of the heaps, which measures about nine hundred feet in 
length on the river side, and has a breadth varying from 
one hundred to one hundred and fifty feet. It has been 
largely undermined, and sections, in some places from three 
to four feet in thickness, exposed. It is not improbable 
that originally this and the smaller deposits were continuous, 
the intervening portion having been washed away by the 
river. If this were so, the mound could not have been less 
than twelve hundred feet in length. It is intersected by a 
small stream near the centre, and is bordered by another 
at its easterly end, both outlets of small morasses in the rear 
of the mound. 
That the Indians confined their encampments, or, at all 
events their cooking, almost entirely to these mounds, is 
proved by the fact, that fragments of pots were picked up 
in large numbers along the shore wherever the shells are 
seen in the bank, and, though careful search was made for 
them, not elsewhere. To make the evidence of the human 
origin of the whole deposit complete, pits Avere sunk at dif- 
ferent points. One of these, about four feet in diameter, 
Avas dug entirely through the shells into the sand beneath, 
which Avas reached at the depth of four feet and three inches. 
Seventy-five fragments of pots and six pieces of the bones of 
the deer, thirteen of turtles, and tAvo of the alligator Avere 
thrown out. These were scattered through the Avhole thick- 
ness of the shell deposit, but not a single specimen was 
found after the sand Avas reached. In a second pit of simi- 
lar size, ninety-seven pieces of pots, six fragments of the 
bones of the deer, eleven of the turtle, and nine of the alli- 
gator Avere found. The shells found here are chiefly Palu- 
dinas, though Unios and Ampullarias are met Avith. 
Old Enterprise is situated on the north-eastern shore of 
Lake Monroe, and forms a distinct bluff consisting entirely 
of shells. It has a front of about one hundred and sixty 
feet on the water side, and at the western end rises some- 10 
FIIESII-WATER SIIELL-IIEAPS 
what abruptly to the height of fifteen feet above the lake ; on 
the top is a plateau, on which formerly stood a hotel and 
several outbuildings, and to the eastward the surface falls off 
by a gradual slope. On this side there is an extensive 
swamp, separated from the lake by a beach-wall of shells, 
consisting of the same species as those found in the bluff, 
and extending several hundred yards along the shore. As 
there are mingled with these shells the fragments of pots 
and bones of animals, they were all no doubt derived from 
the mound, and have been scattered by the action of the 
water. On the westerly side is a spring discharging highly 
sulphuretted water, and flowing into the lake through a small 
morass. The mound extends back from the shore about five 
hundred feet, but is of a very irregular shape, being much 
narrowed in its middle, and spreading out again in the rear 
portion into two unequal and irregular transverse ridges. 
While on the front the mound is composed of the three kinds 
of shells, the rest consists almost exclusively of Paludinas. 
That a large portion of this mound has been destroyed, and 
that the shore of the lake is receding, is obvious from its 
abrupt front, the distribution of its material along the shore, 
and the fact that twelve palmetto trees to the eastward of it 
are now surrounded by water, and their roots denuded to 
the depth of from two to three feet. 
Iii consequence of the undermining of the front, and the 
looseness of the materials, which generally are neither com- 
pact or stratified, excavations were easily made. They were 
continued through several days, many cartloads of material 
were moved, and collections made from all depths below 
the surface, of whatever objects were mingled with the 
shells. These objects consisted of the articles already men- 
tioned in connection with the other localities, and in addition 
various fragments of worked shells which will be described 
farther on. Although several arrow-heads and many flakes 
or "chips” of flint were picked up along the shore, none 
were actually found in the mound. OF ST. JOHNS RIVER, EAST FLORIDA. 
11 
Excavations made in the ridges at the rear of this shell- 
heap did not yield precisely the same, nor so decisive results. 
The shells, consisting almost entirely of Paludinas, were 
much more compact, and the objects found in them much 
fewer. In certain directions there were appearances of some- 
what extensive removals of material having been made, but 
whether by the Indian or the white man, we could not learn. 
To the westward of Old Enterprise, which name applies 
to the bluff just described, is an orange grove, and beyond 
this an "old-field,” which rests upon a thin deposit of shells, 
distributed somewhat uniformly over the surface. Excava- 
tions made here in many places gave the same results as 
were obtained at the bluff. 
Horse Landing is a shell-mound on the right bank of the 
river a few miles above Palatka, and eight miles below Lake 
George; it is three hundred feet, in length, one hundred in 
breadth in the widest part, and rises abruptly in every direc- 
tion. On the front it shows a vertical wall about eight feet 
high, giving a good section of its whole structure, the result 
of the action of the river which here makes a sudden bend. 
Underneath the shells is a layer of sand rising about four 
feet above the water, which at the time we visited the local- 
ity, was not much below its highest mark. In its general 
appearance the mound has the aspect of a geological deposit, 
in consequence of the compactness of the materials, the 
greater decomposition of these than is seen elsewhere, and 
above all, from its distinct stratification. The upper portion 
of the sand on which it rests is more or less mixed with 
fragments of shells, and still higher are alternate layers of 
these, and of shells mixed with sand; it is this condition 
which gives the whole its stratified appearance. At one place 
six such alternations were counted, but in others they were 
less numerous. None of the strata extended continuously 
through the whole length of the mound. Two explanations 
of this appearance are suggested: first, successive over- 
flows of the river; second, interrupted occupation of the 12 
FRESH-WATER SHELL-HEAPS 
mound. The first seems quite improbable. The water is 
not now known to rise above the lowest limit of the shells, 
nor in fact could it rise much beyond this, since the configu- 
ration of East Florida is such, that any unusual flow of water 
becomes at once spread out over the immense tracts rising 
only a few inches above the level of the river. Nothing 
short of subsidence of the land could bring the water to the 
level of the highest of these strata. The second is the more 
probable, but in the absence of proof can only stand as a 
reasonable conjecture. 
Iii view of these facts the search for the evidence of man’s 
work was important, and especially as the mound had the 
appearance of great age. The whole front, in which all the 
objects were undisturbed, was therefore most carefully exam- 
ined, and with the following results : First, excepting within 
a few inches of the surface and in the vegetable mould, not a 
fragment of pottery was discovered; second, a few bones of 
the deer, more or less broken, were found, and one of 
burned; those of the soft-shelled turtle, alligator, and gar- 
pike, as also numerous fragments of charcoal, were obtained 
at various depths between the top of the mound and the sand 
on which it rested. If to these we add an ornament made 
of bone, to be described farther on, we have the scanty evi- 
dence derived from the materials, for the conclusion that the 
mound was built by man. Mr. Peabody, however, made an 
important discoveiy which confirms this conclusion. He ob- 
served a piece of Hint projecting from the sand just beneath 
and quite near to the lowest deposit of shells. It is to be 
remembered that in this part of Florida flint does not natu- 
rally occur, in fact that there is nothing but sand in which 
even pebbles are seldom seen. Before the flint was removed, 
we both carefully examined all the surroundings, and were 
satisfied that the flint and the sand in which it was embedded 
had not been disturbed since the mound was begun. The 
front of the mound was vertical, the section was recent, and 
the small talus which forms below it is constantly removed. OF ST. JOHNS RIVER, EAST FLORIDA. 
13 
Anything once detached is carried away by the current 
which is here somewhat brisk. When removed, the flint 
had all the evidence of having been "chipped,” and was evi- 
dently the result of a rude attempt at an arrow-head. We 
cannot, therefore, in view of all the facts resist the conclu- 
sion that the mound was of human origin. 
The only shell-heap visited by us in which we failed to 
And satisfactory traces of man, was on the left bank of the 
river, a few miles below Hawkinsville (formerly Oceola). 
This deposit is one hundred and fifty to two hundred feet in 
length and eight feet high, has a swamp in the rear from 
which it rises very abruptly; on the front it has been so 
much undermined by the river that it presents a nearly ver- 
tical face, showing a good section through its whole length. 
A series of excavations had been made along the summit 
during the rebellion, for military purposes, so that there 
were unusually good opportunities for examination. Not- 
withstanding all this, we failed to find any pottery or other 
works of man at any point, except within a few inches of the 
surface. The contrast with Black Hammock and Old Enter- 
prise was very striking. The mound was composed almost 
entirely of Paludinas, and, in some points, of these mixed 
with sand, forming a solid conglomeration. In this last we 
saw fragments of the tibia of a deer, which had been broken 
in the same manner as the bones from the other shell-heaps. 
The abruptness with which the mound rose from the level 
surface on the rear gave it the appearance, and this was the 
only circumstance which did, of artificial origin. 
II. ARTICLES TAKEN FROM THE SHELL-HEAPS, SHOWING 
HUMAN AGENCY. 
Pottery. In the old world no traces of pottery have been 
found associated with the earliest flint-implements, and it is 
therefore concluded that the men who wrought these were 
ignorant of it. When the European first came to America, 
some of the tribes were found to be destitute of this art. The 14 
FRESH-WATER SIIELL-IIEAPS 
Patagonians had no earthen vessels either for cooking* or 
holding water. Instead of such the Esquimaux used wooden 
bowls, and the natives of the North-west Coast, Oregon and 
California, water-tight baskets, substituting heated stones for 
the direct action of fire. But with few exceptions pottery, 
as an art, was practised by a large majority of the tribes. 
If, as daily experience tends to show, man, when first in- 
troduced upon the surface of the earth, was at best a pure 
savage without experience, it follows as a natural conse- 
quence that there must have been a longer or shorter time 
when instruments were unknown to him. We have no ade- 
quate grounds for any other belief, than that his knowledge 
and his inventions have been progressively developed, and 
analogy, as we think, legitimately suggests that the most 
simple inventions are signs of actual progress, and point 
back to an earlier state out of which he has emerged. The 
discovery of the oldest of man’s works, either in the form 
CY" 
of worked flints, earthen vessels, and of fire-hearths, do not 
carry us back to his beginning; if we would attain to a 
knowledge of this, it must be sought for in the remains of 
his own body, older than all his works. 
We have as yet no data for determining the time or the 
order of his inventions. But of all his works thus far dis- 
covered, flint-implements are the most ancient, and earthen 
vessels the next. The invention of fire and cookery appears 
to have preceded that of pottery, the proof of the existence 
of the former being the oldest. The determination of how, 
and the period when, fire was first made available as an 
agent, would be one of the most important contributions to 
the history of the early progress of the human mind. 
The shell-heaps oil the St. Johns River, like those from 
the other parts of the United States, show that those who 
inhabited them were not, strictly speaking, primitive men. 
They had already made some progress in the useful arts, and 
however rude their instruments, these were nevertheless in- 
ventions, and such, too, as could only have been the result of 15 
OF ST. JOHNS RIVER, EAST FLORIDA. 
experience extending through considerable periods of time. 
They not only used worked stone, bone and shell, but their 
pottery had passed out of the first and rudest stage into that 
of comely forms with outward ornament, and, as the follow- 
ing table shows, exhibits some little variety in the composi- 
tion of the materials. 
MATERIALS. 
SURFACE. 
LOCALITY. 
■o £ 
rd 
v ■ 
© 
rt-d 
O o 
Olaj 
Claj 
Sa 
8?S? 
o> 
*3 
s 
c3 
H 
H 
s? 
l| 
Oso 
s 
Lake Harney, .... 
10G 
0 
0 
8 
2 
90 
0 
0 
Burial Mound do., 
38 
0 
2 
32 
i 
7 
0 
0 
Watson’s Landing, . 
G2 
0 
15 
G7 
0 
10 
0 
0 
Black Hammock,. 
210 
0 
0 
142 
0 
68 
0 
0 
Old Enterprise, 
23 
2 
92 
G4 
50 
8 
0 
0 
Old Town, .... 
126 
0 
13 
GO 
1 
78 
0 
0 
St. Johns Bluff, 
27 
28 
0 
14 
1 
16 
18 
12 
Total number of pieces, 
592 
30 
122 
387 
55 
272 
18 
12 
For the purposes of comparison we have included in the 
enumeration, articles obtained from St. Johns Bluff, where 
the shell-heap is made up of salt-water species. The table 
shows that more than three-fourths, eighty per cent., of all 
the pieces were made of clay without the admixture of any 
other substance, and that when another substance was added, 
it was most commonly palmetto fibre. The use of sand was 
almost exclusively confined to St. Johns Bluff, where, too, is 
found the most highly ornamented work, characterized by 
the most complex figures. The only pieces marked with the 
impression of a cord were also found at the same place. This 
kind of ornament was extensively used over the United 
States, as we have specimens from Illinois and Massachu- 
setts, and has also been observed on the pots from tumuli 
belonging to the Pre-Koman period of .Great Britain.* We 
have seen no evidence that, as has been frequently asserted, 
these markings indicate that the pots had been formed in nets. 
Although the meshes are often regular, there are no signs of 
knots at the point of crossing of the threads, which there 
*Sir John Lubbock. Prehistoric Times. London, 1865, p. 113. 16 
FliESII-WATER SIIELL-IIEAPS 
certainly would have been if nets had been used. Traced 
pottery was confined almost wholly to Old Enterprise, the 
figures being made with a point, and consisting of combina- 
tions of straight lines. These were sometimes combined 
with indentations. We saw no specimens of pottery made 
in baskets, though frequently told that such are found. The 
absence of pounded shells, as one of the ingredients of their 
pottery, is worthy of notice, especially as shells were in 
daily use among the natives of the St. Johns. 
The plain-stamped pottery was universally distributed, but 
was most abundant at Lake Harney and Black Hammock, 
and is characterized by square, oblong, or lozenge-shaped 
impressions, regularly arranged, the stamp being of suffi- 
cient size to make a large number of them at once, but very 
often the figures are confused in consequence of the instru- 
ment having been applied twice to the same region. In one 
case the apex of the spine of a Paludina had been used as a 
stamp. The complex figures on the pieces from St. Johns 
Bluff, consist of combinations of square, with more or less 
rounded or curved impressions, giving the whole surface an 
intricate series of markings, but which we were unable in 
any specimen found, to reduce to a definite plan. They, 
however, resemble in their general style the pottery de- 
scribed by Schoolcraft* as coming from the sea-coast, and 
remind one of Mexican forms. 
The size of the vessels, as indicated by the curvature of 
the fragments, varied from between two and three to twelve 
inches. The more common kinds appear to have been either 
shallow like a common pudding-dish, or deep enough to be 
used as seething-pots, and both are figured in the illustra- 
tions to the Brevis JVarratio of LeMoyne.f 
Fig. 1, PI. 10 (natural size), represents a rude attempt at 
ornament, consisting of two irregular parallel spiral lines 
starting from the same point. From Old Enterprise. 
* North American Indians, Yol. Ill, PI. XLV. 
fDe Bry, Hist. Amer. Francforte ad Mcenam. Pars. 2da, pp. 4 and 5. OF ST. JOHNS RIVER, EAST FLORIDA. 
17 
Fig. 2, PL 10 (natural size), also from Old Enterprise. 
In this, as was not unfrequently the case at the locality just 
mentioned, straight lines are combined with indentations 
made with a round point. 
Fig. 3, PI. 10 (natural size), represents one of the in- 
stances of complex figures from St. Johns Bluff. This was 
made either by one large complicated stamp, or by a series 
of different stamps, since none of the details are exactly re- 
peated. 
Articles of Shell and Bone. The natives of the upper por- 
tions of the river were in constant communication with the 
coast, and, as might be expected, carried marine shells into 
the interior, some of which were converted into useful arti- 
cles, especially Strombus gigas, Pgrula carica, and JP. per- 
versa, the last acquiring a length of from twelve to fourteen 
inches. 
Fig. 4, PI. 10 (half natural size), one of the most common 
instruments, is made of a triangular piece cut from 1\ carica, 
so as to comprise a portion of the rostrum, serving as a handle, 
and a portion of the swollen part of the body, which is the 
useful part of the tool. The sides and apex are smoothed and 
rounded, while the base is regularly curved and ground to an 
edge like that of a gouge, but with the bevel on the inside. 
A specimen presented to me by Dr. H. P. Bowditch, and 
which he obtained at Old Enterprise, shows quite clearly that 
it was detached from the shell by first cutting a groove, and 
then breaking off the fragment. Length from 80 to 90 m.m., 
breadth from 60 to 70 m. m. 
Fig. 5, PL 10 (half natural size), represents a species of 
Pyrula, with thick and heavy walls; the lip and nearly the 
whole of the rostrum are ground off, and a somewhat irregu- 
lar oval hole with rounded edges is made between the first 
and second row of tubercles, and quite near to the mouth. 
Though such an instrument would give resonance to the voice, 
the position of the hole is not such as to adapt it most favora- 
bly to be used as a horn. It may, nevertheless, be the instru- 18 
FRESH-WATER SHELL-HEAPS 
ment which Bartram states was still in use when he visited 
the St.Johns, and with which, he says, "on one and the same 
day, early in the morning, the whole town is summoned by 
the sound of a conch-shell, from the mouth of the overseer, 
to meet in the public square,” for the purpose of entering 
upon the work of cultivating the soil.* 
Fig. 6, PI. 10 (natural size), is a portion of the rostrum of 
Py rula, 60 m. m. in length, the two ends of which have been 
obliquely ground. 
Fig. 7 (natural size), a piece of bone with a central cavity, 
into which a hole has been drilled at each end. This was 
found at Horse Landing, midway between the top and base 
of the shell-heap, and was the only object found actually 
within the shell-heap, which was clearly the work 
of the human hand. Nearly similar forms are 
figured in the plates of the Brevis JSfar ratio, as 
forming a part of the 
necklace worn by the 
natives, f 
Fig. 8, front view; 
fig. 9, side view Nat- 
ural size), represents 
an instrument made 
of shell, which, from 
the exterior markings 
seen in some, and the 
thick ridge on the inside in near- 
ly all, appears to have been cut 
from the borders of the mouth of Strombus gigas. Several of 
these were found, but all more or less broken. When whole 
the length was about 150 m. m., breadth from 50 to 60, and 
the thickness 25 to 30 m. m. The broad end is ground to a 
blunt edge like that seen in most of the stone chisels from the 
other States, and the other as ground to a blunt point. The 
Fig. 7. 
Fig8. 
Fig. 9. 
* Travels in Florida. Philadelphia, 1791, p. 512. 
t Plates XXXVI, XXXVIII, XXXIX. or ST. JOHNS RIVER, EAST FLORIDA. 
19 
instrument closely resembles the shell-adze used by the 
Kingsmill islanders, specimens of which, with their handles 
attached, can be seen in the Smithsonian collections. One of 
the specimens has been twice perforated by a Lithodomus, 
and thus so far weakened as to lead to fracture. These per- 
forations were undoubtedly made before the instrument was 
wrought. Its outer surface is largely bored by worms. 
A large specimen of Pyrula jperversa, from which the 
interior whorls had been broken out, was found at Blue 
Spring. Such as this were used as drinking horns, and 
are mentioned by Le Moyne, though his figures, drawn from 
memory, as might be expected, do not agree with this or 
any other species. 
Besides the implements of bone already mentioned, a por- 
tion of the radius of a bear, which had been divided by cut- 
ting a groove around the outside and breaking the rest, was 
found at Old Town; and Mr. Bo wd itch gave me the antler 
of a deer which had been similarly treated, and which he 
found at Enterprise. 
Articles of Stone. The collection of stone implements was 
quite small, only twenty-five or thirty pieces, nearly all of 
which were picked up on the shores near Old Enterprise, 
only a few being actually dug from the mounds. A single 
chisel of the ordinary form, and with a remarkably sharp 
edge, was found at Old Town, but all the other articles were 
either arrow or spear-points, and none of them had unusual 
shapes. No pipes or fragments of them were found at any 
place. 
Fig. 10, PI. 10 (half natural size), represents the rude at- 
tempt at an arrow-head, mentioned on p. 403, and found by 
Mr. Peabody under the lowest portion of the shell-heap at 
Horse Landing. 
We will add to the above two pieces of worked shell, both 
of which were, however, taken from the burial-mound at 
Black Hammock, near the shell-heap, but were undoubtedly 
in common use among the natives. FRESH-WATER SHELL-HEAPS 
Fig. 11, PI. 10 (natural size), is an ornament cut from that 
portion of a Pyrula, namely the suture, where one whorl 
joins the preceding, and is bent to nearly a right angle; the 
length of the upright portion is 45 m. m., and the disk at the 
bottom measures 31 by 24 m. m. 
Fig. 12, PI. 10 (natural size), a disk of shell, 18 m. m. in 
diameter, and 5 m. m. thick, with a hole drilled through the 
centre. A similar one is figured by Schoolcraft.* 
Bemains of Animals. The subjoined table gives a com- 
plete list of the different kinds of animals, indicated by the 
bones found in the different mounds. The species most com- 
monly met with are the Deer ( Cervus Virginianus), the Ter- 
rapin (Emys Floridana), Soft-shelled Turtle ( Trionyx ferox) 
and the Alligator (Alligator Mississiypiensis). The condi- 
tion of the bones in many instances, particularly those from 
Old Enterprise and Horse Landing, indicated that they had 
been long buried, inasmuch as they had lost nearly all their 
organic matter, and when exposed to heat scarcely changed 
their color. In many instances they were incrusted with a 
deposit of lime, and had the shells in which they were em- 
bedded cemented to them. The bones of birds are quite rare, 
even those of the wild turkey and of the various species of 
ducks, which in the winter frequent the rivers and lakes in 
immense numbers. Of fishes, the species most commonly 
represented are the gar-pikes (Lepidosteus), and a cat-fish 
( Pirnelodus). 
Iii the illustrations to the Brevis Narratio of LeMoyne, 
PI. XXIV represents a lire over which is built a frame, and* 
on this, exposed to heat and smoke, are several animals, 
among which can be recognized the deer, a small mammal, 
the mouth of which resembles that of the opossum, an alli- 
gator, an eel or a snake, and several species of fish. Several 
Indians are standing near, one fanning the lire, and another 
holds an alligator under his arm. On PL XXIII, natives 
are represented carrying food in baskets, one of which con- 
* Notes on the Iroquois. Albany, 1847, p. 243. OF ST. JOHNS RIVER, EAST FLORIDA. 
21 
tains a deer, a flsh, and an alligator. This is quite too large 
a load for one basket, and too much importance must not be 
attached to these plates, since they were drawn from mem- 
ory, but they may be taken as an indication of what the 
kinds of food were. In the text, the writer states that they 
"ate freely of the flesh of the alligator, which is white and 
clean, and which we should have eaten often had it not been 
too redolent of musk.”* This objection we have found from 
personal experience to be a valid one. 
SPECIES OF ANIMALS FOUND IN THE SHELL- 
MOUNDS. 
>4 
0) 
X z 
c3 
r~ bl 
Black 
Hammock. 
Enterprise 
Blue 
Spring. 
M’d above 
Osceola. 
Oldtown. 
Horse 
Landing. 
Deer, Cervus Virginianus, 
» 
* 
* 
* 
* 
* 
* 
* 
Bear, Ursus, ..... 
* 
Raccoon, Procyon lotor, .... 
* 
* 
* 
Opossum, Didelphys, .... 
* 
* 
Turkey, Meleagris gallopavo, . 
* 
* 
Birds, not known, .... 
* 
* 
* 
* 
Terrapin, Emys Floridana, 
* 
* 
* 
* 
* 
* 
* 
* 
Soft-slielled Turtle, Trionyx ferox, 
* 
* 
* 
* 
* 
* 
Species of Turtle not known,. 
* 
* 
Alligator, Alligator Mississippiensis, . 
* 
* 
* 
* 
* 
* 
* 
Catfish, Pimelodus, 
* 
* 
Gar-pike, Lepidosteus, . . • 
* 
* 
* 
* 
* 
Fish, not known, ..... 
* 
* 
That the animals of the shells which form the materials of 
the mounds were used as food, there seems to be no reason- 
able doubt. Unios are known to be edible, and, almost ex- 
clusively, form the shell-heaps on the borders of other rivers 
as the Ohio,f the the Concord, etc.§ We are 
not aware of any evidence that Ampullarias and Paludinas 
have been so used elsewhere than in Florida, but their asso- 
ciation with pottery, and charcoal, and the bones of edible 
animals, seems to be decisive. If the inference we have 
drawn be correct, then it follows that the animal food of the 
ancient inhabitants of Eastern Florida was very largely de- 
rived from these species, and especially the Paludinas, since 
* Ibid., p. 5. 
f Atwater, Archaeologia Americana, Vol. I, p. 226. 
$ Brinton, Smithsonian Publications, 1866, p. 356. 
§ J. Wyman, Proceedings of Boston Society of Natural History, Vol. XI, p. 243. 22 
FRESH-WATER SIIELL-IIEARS 
the remains of fish, turtles, alligators, and deer, form so 
insignificant a portion of the whole heap. 
In view of the vast number and size of the shell-heaps 
now known to be scattered along the Atlantic coast,* and 
the vast quantities of shells which compose them, it is quite 
clear that the aborigines must have depended largely upon 
shell-fish for food. In fact such was obviously the case with 
the early inhabitants of the old world as well as new. Of the 
extent to which vegetable substances were made use of, the 
shell-heaps offer no evidence ; but it seems certain, that until 
the bow and arrow, the trap or the net were invented, the 
animal food must have of necessity been derived from such 
species as could most easily be obtained, and among these 
the shell-fish and the more sluggish reptiles would first at- 
tract attention. 
III. AGE. 
No satisfactory data were found for determining the age 
of the shell-heaps. The appearance of great age which some 
of them have, as at Horse Landing and Old Enterprise, is 
important; the same may be said of the fact that the bones 
embedded in them had lost nearly all their organic matter, 
and at both of these places were incrusted with calcareous 
deposits, in some instances forming a conglomerate. The 
time required for these last results is not necessarily very 
great, but the organic matter of bone is destroyed very 
slowly, and is largely present in those of some of the extinct 
animals. We have obtained a larger quantity of animal 
matter from the bones of the Mastodon than from those of 
the deer at Old Enterprise. 
The most trustworthy records are found in the forest trees 
growing upon the mounds. These give us a minimum age 
with some approach to accuracy. The live-oaks (Quercus 
* Dr. Joseph Leidy, Pi’oceedings of Academy of Natural Sciences, 1866, has de- 
scribed the shell-heaps at Cape Henlopen, and should have been cited in our commu- 
nication in the Naturalist for December, 1867, but at that time we had not seen it. OF ST. JOHNS RIVER, EAST FLORIDA. 
23 
virens) are not only long-lived, attaining an age of many 
centuries, but their wood is the most durable of all the forest 
trees of the United States. One of these, which had fallen 
from the effects of age, lies upon the top of a mound in the 
woods near Blue Spring, and measures live feet and six 
inches in diameter. As it was on the summit of the mound, 
it could not have begun to grow until the mound was nearly 
or quite finished; it may have begun many years later. It 
had been dead for a long time ; its bark, all of the small and 
most of the large branches had disappeared. These trees 
after they are dead still remain erect for many years. Some 
of them girdled more than thirty years since, can still be 
seen standing firmly in the Indian-old-fields. It certainly 
would not be extravagant to say that the tree in question 
had been dead more than half a century. Fragments of pot- 
tery were found in the earth and shells contained in the 
upturned roots of this tree, and on sinking a pit in the place 
formerly covered by the upright trunk, others were found at 
a depth of from two to three feet. We had neither the tools 
nor the aid for making a section of this trunk to count the 
number of annual rings. Through the kindness of Commo- 
dore John lingers, of the United States Navy, we have re- 
ceived a section from a tree nearly a century and a quarter 
old, and find that at the beginning of the second century 
there are about fifteen rings to the inch. In later periods of 
the life of the tree they would of course be more numerous. 
Assuming fifteen to the inch as the average, a half diameter 
of thirty-three inches would give 495 rings, or nearly five 
hundred years; if to this we add fifty years for the time 
since the tree died, there can be no doubt that the mound 
was substantially as complete as now more than a century 
before the discovery of the country. 
We know of no data based on the quantity of materials 
of which the mounds were formed, on which to estimate the 
time required to build them; to this end, it would be neces- 
sary to know the number of persons occupying the place, FRESH-WATER SHELL-HEAPS . 
and the daily or annual consumption of food. If, as is the 
case of mounds built up in the swamps, they were resorted 
to only by those who could find camping conveniences upon 
them, the number must necessarily have been very small. 
The later aborigines had no traditions with regard to these 
shell-heaps, or the burial-mounds which are sometimes near 
them. They ascribed them to a former race. Florida, how- 
ever, has been more than once overrun by exterior tribes, 
and the absence of traditions might in this way be accounted 
for, since these would be likely to be lost with the change of 
inhabitants. Under the most favorable circumstances tradi- 
tions form Jin uncertain basis for history. If, therefore, on 
the one hand there is no proof of great antiquity, it may still 
be claimed that there is nothing inconsistent with it, and 
that the appearances of the mounds, and facts connected 
with them, largely favor it. 
It was the special object of this paper to describe only 
fresh-water shell-heaps, but as we have visited two deposits 
consisting of marine species, chiefly oysters, we will add a 
few words with regard to them, especially the above-men- 
tioned locality. The one at Fernandina, on the northerly 
end of Amelia Island, has already been described by Dr. 
Brinton,* who has given the most satisfactory proof of its 
human origin, and of other similar deposits on the Atlantic 
and Gulf coasts of Florida. The result of our own observa- 
tions at Fernandina are confirmatory of what Dr. Brinton 
has recorded, and afford some additional evidence from the 
earthworks thrown up during the rebellion, and the mounds 
over the soldiers’ graves in the rear of Old Fernandina, 
in making both of which, portions of the shell-heaps were 
uncovered, and the contents, similar to those previously 
noticed, exposed. 
IV. ST. JOHNS BLUFF. 
St. Johns Bluff has a twofold interest, for it was not only 
* Floridian Peninsula, p. 177. OF ST. JOHNS RIVER, EAST FLORIDA. 
25 
a favorite resort for the Indians, but was the scene of two of 
the most tragic events in the early history of the continent. * 
It is situated on the right bank of the river, and about live 
miles from the mouth. Like all the adjoining shores, it is 
composed of a line yellowish silicious sand. It is about forty 
feet high on the front, and at the eastern end rises quite 
abruptly out of a marsh, and to the westward, i. e. up the 
river, descends at first by a rapid, then a gentle slope, which 
merges into a nearly level plain, backed by the thickly- 
wooded hills; beyond this is a marsh, which, still farther to 
the westward, is bordered by a creek.f The base of the 
bluff is washed by a swift current at every tide, so that it is 
constantly undermined, and is rapidly disappearing. Earth- 
works thrown top during the rebellion have already 
begun to fall. I was told by a man living near by that an 
oleander tree, which I saw lying at the water’s edge to the 
westward of the bluff, a few years since was thirty feet from 
the shore in the middle of a garden. 
At present the bluff itself must greatly differ from what it 
was when the French came, and it is highly probable that 
more of it has been destroyed than remains. The site of 
Fort Caroline has not been identified, and has probably disap- 
peared. The bluff* presents a front of clear sand, is overgrown 
with trees except where military works were thrown up, 
and beneath the vegetable mould, a few inches thick, is a 
layer of oyster shells, with a very slight admixture of sand, 
extending from two to three hundred feet along the more 
easterly portion, and varying in thickness from a few inches 
to three feet. A second and much thinner layer is seen to 
* It was here that the French, under Jean Ribault, in 1504, built Fort Caroline with a 
view to establish a Huguenot colony, which in less than eighteen months Menendez, 
writh the purpose of impeding the progress of Protestantism captured, put the garrison 
to the sword, and set up the inscription, “ not as to Frenchmen, hut as to Lutherans.” 
Two years later Dominique de Gourgues avenged the atrocity, by retaking the fort, 
killing the captives, leaving behind attached to a tree another inscription, “not as to 
Spaniards or mariners, hut as to traitors, robbers, and murderers.” See Parkman, 
Pioneers of France in the New World. Boston, 1865, p. 157. 
fMr. Parkman’s description of St. Johns Blufl', in the work already cited, is admira- 
ble for its portrayal of the general landscape as well as the individual details. 26 
FRESH-WATER SHELL-] IE APS 
the westward, where the land rises only eight or ten feet 
above the water. It is not improbable that the two deposits 
were originally connected, the intervening portion having 
been washed away. Fragments of pottery which have fallen 
from the banks arc scattered along the whole shore in front 
of these deposits, and on examining fresh sections made by 
the falling of the bluff, and also in making excavations in 
undisturbed portions, similar fragments were found in place, 
and so there can be no doubt that the shells and pottery 
were simultaneously deposited. After careful search no 
tlint or other implements were found during my visit, either 
in the bluff itself or along the shore, neither were the bones 
of edible animals found minified with the shells. Flint im- 
plements have, however, been obtained in considerable num- 
bers, and an arrow-head was given me by a negro, who had 
picked it up near by. The various excavations for military 
purposes, revealed the existence of shells several hundred 
feet to the rear of the present front of the bluff, and beyond 
the creek to the westward of the marsh is a farm, where pot- 
tery and shells may be seen loosely scattered over a tract of 
many acres in extent, wherever the plough has turned up the 
soil. 
The shell-mounds of the sea-coast, as well as of the inte- 
rior, seem to have passed almost unnoticed by the early 
writers on Florida. Dr. Brinton quotes a single passage, 
the only one met with by him relating to the subject, from 
Cabeza de Yaca, in which it is stated that the houses of the 
Indians were "built of mats on heaps of oyster shells.”* 
ENUMERATION OF THE SHELL-HEAPS VISITED. 
Besides those mentioned in the following list, there are many others 
not visited by the writer, some of which are said to be of even di- 
mensions than any seen by him. 
The localities are mentioned in the order in which they stand on the 
river, beginning with those nearest the sources. 
1. llattlesnake Hammock, on Salt Creek, right bank, and near the union 
of the creek and the St. Johns. 
♦Floridian Peninsula, p. 17!). OF ST. JOHNS RIVER, EAST FLORIDA. 
27 
2. Solee’s Landing, right shore of Lake Harney. 
3. King Phillip’s-town, left bank of the St. Johns, a mile below the out- 
let of Lake Harney. There is a large burial-mound near this locality. 
4. Another shell-heap, one mile below preceding. 
5. Watson’s Landing, right bank between Lakes Ilarney and Jessup. 
G. A mound one mile above preceding, on the same side of the river. 
7. Black Hammock, left bank, just above the outlet of Lake Jessup. 
There is a small burial-mound here. 
8 & 9. Two mounds on the right bank and below the preceding, but 
separated from the river by a large lagoon. 
10. Spear’s Landing, about live miles above Lake Munroe, left bank. 
There is a burial-mound at this place. 
11. Buzzard’s Roost, left bank, near entrance to Lake Munroe. 
12. Doctor’s Island, right shore of Lake Munroe, above Enterprise. 
13. Old Enterprise, l’ight shore of Lake Munroe. 
14. Outlet of Lake Munroe, right bank. 
15. Wekiva, right bank. 
1G. Blue Spring, right bank. 
17 & 18. Two mounds in the woods below Blue Spring, with a wide 
swamp between them and the river. A third but small mound was found 
about a half mile from them. 
19. Mound above Hawkinsville, left bank formerly, and still ought to 
be called Osceola, or, as Dr. Bi'inton writes the name, Ass-se-he-ho-la, 
Rising Sun, after the celebrated chief who was prominent in the Florida 
War. 
20. Mound below preceding, left bank, having the usual appearance of 
the other shell-heaps, but in which we failed to find signs of its artificial 
origin. 
21. Old Town, left bank, seven miles below Hawkinsville. 
22. Small mound in the woods in the rear of the preceding. 
23. Mound above the outlet of Lake Dexter, left bank. 
24. Mound below the outlet of Lake Dexter, right bank. 
25. Fort Butler, left bank. 
26. Volusia, right bank. 
27. Rope’s Island, right bank, entrance of Lake George. 
28. Drayton’s Island, now Rembrandt’s Island, at the outlet of Lake 
George, left bank. 
29. Horse Landing, right bank, eight miles above Palatka. 
30. Palatka, left bank, one hundred miles from the mouth of the river. 
31. St. John’s Bluff, right bank, five miles from the mouth of the river. 
32. Old and New Fernandina, at the northern end of Amelia Island. American Naturalist. 
Fig. 3. 
Fig. 1. 
Vol. II. PI. 10. 
Fig. 2. 
Fig. 5. 
Fig. 6. 
Fig. 12. 
Fig. 11. 
Fig. 10. 
Fig. 4. 
"WYMAN ON THE SHELL-HEAPS OF FLORIDA.