LIBRARY

OF THE

UNIVERSITY OF CALIFORNIA.

RECEIVED BY EXCHANGE

Published monthly by the

New York State Education Department

BULLETIN 338 APRIL 1905

New York State Museum

JOHN M. CLARKE Director

Bulletin 87
ARCHEOLOGY 10

PERCH LAKE MOUNDS

WITH NOTES ON OTHER NEW YORK MOUNDS, AND SOME ACCOUNTS
OF INDIAN TRAILS

BY

WILLIAM M. kEAUCHAMP S.T.D.

List of authorities . • • • • 3

Perch lake mounds. 5

Other New York mounds 24

Trails 33

Addenda 48

Explanations of plates 53

Index 77

ALBANY

NEW YORK STATE EDUCATION DEPARTMENT
1905

Mi 47111-84- 1 500

Price 20 cents

STATE OF NEW YORK
EDUCATION DEPARTMENT

Begents of the University
With years when terms expire

1913 WHITELAW REID M.A. LL.D. Chancellor - - - New York

1906 ST CLAIR MCKELWAY M.A. L.H.D. LL.D. D.C.L.

Vice Chancellor --- Brooklyn

1908 DANIEL BEACH Ph.D. LL.D. Watkins

1914 PLINY T. SEXTON LL.D. Palmyra

1912 T. GUILFORD SMITH M.A. C.E. LL.D. Buffalo

1907 WILLIAM NOTTINGHAM M.A. Ph.D. LL.D. - - Syracuse

1910 CHARLES A. GARDINER Ph.D. L.H.D. LL.D. D.C.L. New York

1915 CHARLES S. FRANCIS B.S. Troy

1911 EDWARD LAUTERBACH M.A. - New York

1909* EUGENE A. PHILBIN LL.B. LL.D. New York

1916 LUCIAN L. SHEDDEN LL.B. Plattsburg

Commissioner of Education

ANDREW S. DRAPER LL.D.

Assistant Commissioners

HOWARD J. ROGERS M.A. LL.D. First Assistant Commissioner
EDWARD J. GOODWIN Lit.D. Second Assistant Commissioner
AUGUSTUS S. DOWNING M.A. Third Assistant Commissioner

Secretary to the Commissioner

HARLAN H. HORNER B.A.

i

Director of Libraries and Home Education

MELVIL DEWEY LL.D.

Director of Science and State Museum

JOHN M. CLARKE LL.D.

Chiefs of Divisions

Accounts, WILLIAM MASON
Attendance, JAMES D. SULLIVAN
Examinations, CHARLES F. WHEELOCK B.S.
Inspections, FRANK H. WOOD M.A.
Law, THOMAS E. FINEGAN M.A.
Records, CHARLES E. FITCH L.H.D.
Statistics, HIRAM C. CASE

New York State Education Department

New York State Museum

JOHN M. CLARKE Director

Bulletin 87
ARCHEOLOGY 10

OF THE

UNIVERSITY

OF

PERCH LAKE MOUNDS

WITH NOTES ON OTHER NEW YORK MOUNDS, AND SOME ACCOUNTS OP

INDIAN TRAILS

LIST OF AUTHORITIES

Abbreviations at the left are used in the bulletin in exact reference to works in the following list.

Amidon Amidon, R. W. Letter to Rev. Dr Beauchamp.

Beauchamp Beauchamp, W. M. The Aborigines. Jefferson County

Hist. Soc. Trans. Watertown 1887.
Benedict Benedict, A. L. Mound Builder Remains on Cattaraugus

Creek. Erie county, N. Y. American Antiquarian.

Chicago 1901.
Boyle Boyle, David. Annual Archaeological Reports, 1901 and

1903. Toronto.
Cammerhoff Cammerhoff, Frederick. Diary of the Journey of Bro.

Cammerhoff and David Zeisberger to the Five Nations

from May 3-14 to August 6-17.

Manuscript.

Cecil Cecil, Harry B. Letter to Rev. Dr Beauchamp.

Clark Clark, John S. Notes in various papers.

Emerson Emerson, Edgar E. Our Country and its People. Boston

1898.

French French, J. H. Gazetteer of New York. Syracuse 1860.

Getman Getman, A. A. Letters to Rev. Dr Beauchamp.

Harris Harris, George H. Aboriginal Occupation of the Lower

Genesee Valley. Rochester 1884.

First issued in Semi-centennial History of Rochester, ch. 1-15. Roches-
ter 1884.

Hough Hough, F. B. History of Jefferson County. Albany 1854.

Jennings Jennings, C. P. Letter to Rev. Dr Beauchamp.

Jordan Jordan, J. W. Spangenberg's Notes of Travel to Onon-

daga in 1745. Pa. Mag. Hist, and Biog. v. 2. Phil. 1878.
Mackay Mackay, John. Letter to Rev. Dr Beauchamp.

Marvin Marvin, D. S. Mounds at Perch Lake, Jefferson County.

Jefferson County Hist. Soc. Trans. Watertown 1887.
Mattern Mattern, J. E. Letter to Rev. Dr Beauchamp.

O (\ A A f- n

NEW YORK STATE MUSEUM

Mayhew Mayhew, G. F. Bulletin of Natural History of New

Brunswick. St Johns N. B. 1884.
Morgan Morgan, L. H. League of the Ho-de-no-sau-nee or

Iroquois. Rochester 1851.
O'Callaghan O'Callaghan, E. B. Documents Relative to the Colonial

History of the State of New York. Alb. 1853-87.
Putnam Putnam, F. W. nth Report of Peabody Museum; quoted

by Cyrus Thomas in Report on the Mound Explorations

of the Bureau of Ethnology. Wash. 1894.
Relations Relations des Jesuites. Quebec 1858.

Skinner Skinner, A. B. Letter to Rev. Dr Beauchamp.

Smith Smith, Harlan I. Letter to Rev. Dr Beauchamp.

Thomas Thomas, Cyrus. Report on the Mound Explorations of

the Bureau of Ethnology. U. S. — Ethnology, Bureau of.

I2th An. Rep't. Wash. 1894.

Twining Twining, J. S. Letter to Rev. Dr Beauchamp.

Wallbridge Wallbridge, Thomas C. On some Ancient Mounds

upon the Shores of the Bay of Quinte. Can. Jour, of

Industry, Science and Art. Toronto 1860.
Wilson Wilson, James Grant. Arent Van Curler and his Journal

of 1634-35. Am. Hist. Ass'n. An. Rep't. 1895. Wash.

1896.

Woodworth Woodworth, Henry. Letter to Rev. Dr Beauchamp.
Zeisberger Zeisberger, David. David Zeisberger's and Henry Frey's

Journey and Stay in Onondago from April 23d to

November I2th, 1753.

Manuscript.

/ OF THE

I UNIVERSITY

OF

'

PERCH LAKE MOUNDS

In all histories of Jefferson county, N. Y. there are slight notices
of the curious mounds about Perch lake. When Squier wrote his
account of the antiquities of New York they had not attracted atten-
tion, for they were inconspicuous and remote from ordinary travel.
Mr F. B. Hough seems to have been the first to mention them, a few
years later, and he said there were several at the mouth of Lowell
creek, Perch lake, about 30 feet across and with depressed centers.
No creek is now known by this name to the oldest inhabitants, but
he probably meant Hyde creek at the head of the lake, where there
are yet a number. He added that there are some on Linnell's island.
In these were found pottery, burnt stone and charred corn. Hough,
p.io

Linnell's island is not in the lake, but is an extensive elevation in
the great swamp west of its foot and north of the outlet, as shown in
plate i. It lies between two large streams and is now occupied by
farms. Some mounds still remain on those owned by Messrs Gailey
and Klock. No charred corn has been reported by any accurate
investigator, and small coals may have been mistaken for this. Very
little pottery has anywhere been found, but charcoal and burnt stone
appear in all. In French's Gazetteer it is said that-" in the vicinity of
Perch Lake have been found several barrows, or sepulchral mounds."
French, p. 360. It would not have been surprising if some of the
larger ones had had a secondary use for burial, being well adapted
for it in such a region, but no evidence of this has yet been found.

Regarding these Mr J. S. Twining wrote me in 1886 of a more
extensive distribution of these mounds than has been given by others.
He said :

We have extensive vestiges of a much older race than those who
built the forts and made the pottery. They are scattered along Black
river, some 6 miles from Copenhagen, and also on the hills back
of Perch lake, some 10 miles from Watertown, on the farms of John
Gailey and A. Klock. On the latter are the largest and most perfect.
They are the remains of camp bottoms, with a depression in the
middle, with a true circle of camp refuse and burnt stones around
them from 2 to 5 feet high, and with a diameter of from 20 to 30 feet.
I have never found a piece of pottery in any of them, but plenty of
flint chips. Beauchamp, p. 113

6 NEW YORK STATE MUSEUM

Mr D. S. Marvin made a day's exploration of the Perch lake
mounds in August 1886 in company with Messrs Carter, Wood-
worth and Woodard. The results he embodied in a paper read before
the Jefferson County Historical Society, Mar. 15, 1887, adding
a few facts from the earlier explorations of Henry Woodworth and
J. S. Twining. The lake is a small one, part of the shores high and
rocky, but much more low and marshy. The mounds occur only on
the higher part. The outlet is 6 miles long, and mounds have been
reported near this. At the natural bridge, near its mouth, are exten-
sive camp sites with abundant bone articles and fragmentary pottery.
The most important part of Mr Marvin's paper is quoted here as
follows :

The objects that arrest our attention and interest us the most are
the so called Indian mounds, observed along both shores of the lake,
and more or less down the outlet. They are situated upon the bluffs
overlooking the water, and reach back from the lake sometimes a
hundred rods; they number some two hundred in all. These so
called mounds are all round, usually from 50 to 90 feet in circum-
ference ; some of them double, and so near that their edges coalesce.
They are elevated or raised above the summits of the hills they occupy
from 2 to 4 feet. Where the land has not been cleared, ordinary
forest trees of all ages are seen growing around and upon the
mounds, ranging from yearling growths to trees several hundred
years old. The debris usually observed about old Indian villages is
found buried in the soil, old bones and broken pottery; the organic
remains though seem to have mainly rotted and gone to decay. The
broken pottery observed was of the usual patterns, but it is only
sparingly observed, for around some of the mounds none could be
found. A few of the small mounds were flat topped, but the usual
shape and appearance is a ring of earth, with a depressed or basin-
shaped center.

In opening cross sections, or digging trenches from the outside to
the center of the circles, as the centers are approached, remains of
fires, charcoal, ashes, etc., were observed, sparingly though in the
case of the largest mound. There was observed no disturbance of
the soil below the level of the natural surface. The dirt of which
the mounds had been constructed, is the common country soil, none
of it seemingly brought from a distance, similar in character and com-
position to the soil of the adjacent land, made up of clay, sand and
small fragments of the underlying limestone, belonging to the Tren-
ton group, as near as I could determine from a cursory examina-
tion of the contained fossils, with here and there an occasional

PERCH LAKE AND OTHER NEW YORK MOUNDS 7

transported or drift pebble. The only observable difference was a
darker color, caused by an increase of decayed organic matter and
burned earth. No graves or human bones were observed. No lines
of entrenchments were to be seen. Nor have there been any metal
objects or utensils found.

The explanation of the phenomena observed here, that has seem-
ingly puzzled several generations of white men, seems to be plain
and simple. There is no necessity for bringing farfetched theories
to explain the observed facts. Whoever has been to California and
noted the singular rings of earth, with their basin-shaped centers,
that are known to be the remains of the old rancherias of the Dig-
ger Indians, can readily see here in the close resemblances the original
forms of Indian houses, belonging to the lower stages of barbarism,
and probably a more or less universal style of house belonging to
this stage of advancement, usually occupied only during the winter
months, or generally deserted for nomad life during the warmer
summer months. This style of house was constructed with a frame-
work of poles set upon end and meeting at the top, and covered with
dirt, leaving an uncovered space at the top to serve for the exit of
smoke.

The writer once visited one of these dirt houses in California, large
enough to hold several hundred people, but perhaps not larger than
the remains of one of those observed at Perch lake. Professor
Thomas has described the remains of similarly constructed houses in
Mississippi, Alabama and Georgia. . . I have -also observed near
Burrville, within a strongly fortified enclosure, circles of toadstools
that had grown up from organic matters, old bones, etc., buried in
the soil, showing that similar round houses once existed within forti-
fied enclosures, but unfortunately both ditches and circles are now
leveled by the plow. Marvin, p. 58

I add some notes sent me in October 1901 by Mr Henry Wood-
worth, one of the party mentioned, but whose conclusions are dif-
ferent. Both these gentlemen are careful observers :

I visited the mounds with Mr Marvin and Clarence Woodard, and
we spent one day at the lower end of the lake, on the south side. We
found a very large mound on a ridge in the woods. Some large
maple trees were in it. Distance from the lake was 10 or 15 rods.
We did the most of our digging in that one, but we dug in others
that were hollowed on top, as most of them are. We found but little
to pay us for our labor. The ashes and coal that would naturally
accumulate were very light. For that reason I think they were occu-
pied only for a short time in the summer, for fishing and hunting. If
they had been used to winter in, the accumulation would have been
much more. , I and my son dug some in the mounds on the Gailey

8 NEW YORK STATE MUSEUM

farm at another time, but we found nothing to satisfy our curiosity.
No shells were found in any of them by any one. We found no flint
in any mounds. Mr Gailey said some had been found, and stone
pipes, but who has them I do not know. He says but little was ever
found. I found no bone articles. We found some animal bones, but
they were so decomposed that they easily crumbled to pieces. About
the outlet and lower end of the lake are a number of mounds on the
Gailey farm, of different sizes. Some are flat on top, but most have
a depression in the center. Mr Gailey said there are over 200 up
the creek and around the lake. I think there are mounds below Mr
Gailey's, but I never visited the upper end of the lake.

He said there was no accumulation of burned stones in the mounds,
probably meaning the small ones used in heating water. From this
and the lack of pottery he concluded that no cooking, or but little,
was done in them. Most of the many stones found show the action
of fire, but they are usually of some size.

In a letter dated Aug. 4, 1900, Dr Getman said:

We were at Perch lake a few days ago, and examined the mounds
that are found at each end of the lake. We were at the north end
and along the banks of Hyde creek. They are situated near the bank
of the lake, extending upwards on to a high bluff of sandstone, and
gravel of the same, along the banks of Hyde creek. They are 25 to
30 feet across, 3 to 5 feet high, with a central depression of 8 to 10
feet in diameter. This depression is paved with the usual firestones.
We saw one that was on a gravel bed, and had been partly removed.
It was uniform in thickness, simply burned sandstone, gravel and
black earth. The earth is different from the surrounding soil, being
burned. We saw no pottery, bone, or anything that would give us a
clue to the builders. Hough says broken pottery and bones are
found there. This I think a mistake. Some have been dug to the
center, and we were informed they had found flint and stone imple-
ments. There was only one that showed evidence of large timber
growing from the site. We counted 15 in a piece of woods, and the
trees (maple) were mostly small that were growing on the banks.

In a recent history of Jefferson county the mounds at the north
end of the lake are again mentioned, but with little additional in-
formation save that of partial location. The editor says that at the
lake 8 or 10 mounds are on the lands of George W. Sherman and
Alonzo Van DeWalker 10 or 15 rods from the shore. They are cir-
cular, 2 or 3 feet high, 2 to 4 rods in diameter, and with the central
holes 2 feet deep. The largest is said to be on the Sherman farm,
near the ruins of the old La Farge mansion. Emerson, p-738

PERCH LAKE AND OTHER NEW YORK MOUNDS 9

There are two large groups north of any of these, and but one
mound was observed by me over 40 feet in diameter. The fine pair
in front of the old mansion are by no means of the largest size, either
in hight or width.

Before adding notes of personal observations to these, it may be
well to take notice of some kindred groups on the north shore of
Lake Ontario, which I had planned to examine some years since.
Mr Thomas C. Wallbridge read a paper " On some Ancient Mounds
upon the Shores of the Bay of Quinte," Mar. 3, 1860, which was
printed in the Canadian Journal for September of that year. These
mounds had then been locally known as artificial for 50 years, but no
account had been previously published. Commencing at Redners-
ville they could be traced along the bay about 8 miles to Massassaga
point. This space, with the islands of Big bay, included about 1.00
distinct mounds, but others could be seen at intervals from the
eastern to the western end of the Bay of Quinte. Others were
reported at one place on the River Trent. Mr Wallbridge said :

As far as has yet been ascertained, there is but one class or form
of mounds in this part of the country, and the truncated cone is the
shape they assume. In size they vary from a diameter at the base of
30 to 50 feet, to a diameter at the apex of 12 feet. Each mound has
a shallow basin or circular depression upon its summit, which, what-
ever be the size of the work, has a diameter of 8 feet ; and no mound
under my observation possessed an altitude of more than 5 feet. It
is a remarkable peculiarity of these works, that in almost every
instance they occur in groups of two, and at irregular distances the
one group from the other. Irregularity is likewise observable
between any one mound and its fellow, these being sometimes found
in juxtaposition, and again from 50 to 100 feet asunder. The two
of the same group are always of one size. With respect to the sur-
rounding country they are situate apparently without design, now
at the foot of a commanding hill, then halfway down the side of a
bank, and again so near the shore that in several instances they have
been destroyed by the action of the water. Twice they have been
found in very low or swampy ground, and in those cases they occur
singly. Wallbridge, p.m

He opened five of these at Massassaga point in August 1859. A
cut was made 33 feet long, 2 feet wide and 3 feet deep, to the original
surface. Under a few inches of mold was a heap of broken gneiss,

IO NEW YORK STATE MUSEUM

conforming to the outer shape of the mound. The stones varied
from i to 20 pounds each, but those forming the bottom of the
basin were the smallest of all. Some showed the action of fire,
but there were no traces of this in the mound. In making a cross-
section some fragments of birch bark and bone were found above
the stones. He said:

The other mounds examined agreed in all particulars of con-
struction with that above described, excepting in one pair where it
was evident from what remained that the inside margin of the basin
of each mound had been surrounded with flat stones placed vertically
and touching at their edges, as if designed to prevent the earth
falling into the hollow. Similar stones, perhaps used for the same
purpose, were observed lying near most of the other mounds in this
vicinity.

He thought these had been displaced by diggers, and added :

In several instances the builders have been forced, from the nature
of the surrounding country, to carry their material from a distance,
but to obtain the usual covering of mold for the pair of mounds
last mentioned they have bared the smooth underlying rock of its
scanty soil, in a well defined circle about the works. The use of
broken gneiss for a building material, to the almost entire exclusion
of limestone, is a noticeable feature.

Limestone was most easily procured, but I think its change by
fire may have made it objectionable. Large trees grew on some
mounds, one oak stump being 8 feet around.

So far the likeness to the Perch lake mounds is that of external
form, size and situation, with a tendency to pairs. The interior
differs in the character and arrangement of the stones, and the
absence of coals. Similar ones were examined later, but one was
of a sepulchral nature. This was excavated from the center to the
natural surface. Some of the diagrams made are here partially
reproduced.

Plate 10, figure 3, shows a section of the mound in which skele-
tons were found, with general features of the construction of all.
In this appear the interior stones, the overlying soil and the central
depression. Figure 2 shows the position of some articles found, and
the central chamber. Ground was broken at ro, and a little below
the surface was a flat, horizontal limestone, with fragments of human

PERCH LAKE AND OTHER NEW YORK MOUNDS II

bones and birch bark, and a bone awl 8 inches long. These were
probably from intrusive burial.

Another flat stone was found 2 feet from the surface, with three
human skulls underneath, in a rude box of flat limestones. Many
of the remaining bones were found, and five well preserved
crania were secured. One skeleton at 6 was in a sitting position,
with a pile of articles by it. Among these was the upper part of a
bone comb, several teeth of the same, a unilateral bone harpoon,
and three long shell beads. These articles do not indicate a high
antiquity, and are much like those of New York. The burial was
clearly intrusive.

A sketch of this interesting group has been given because it is
little known, and partly because, being not far distant and in a very
similar situation, it may have some relation to those of Perch lake.
The latter seem to have gradually increased in hight; according to
Mr Wallbridge the former would seem to have been of nearly the
same size from the beginning. This hardly seems probable, nor is
it likely no fire was used in them, judging from what is found else-
where. I saw no ashes in those of Perch lake, and in some cases
the coals were so blended with the soil as to be hardly distinguishable.

In the spring of 1901 I visited Perch lake, where the old La Farge
mansion once stood, at a considerable distance north of a large
stream which enters the lake on the east side. Quite a point extends
into the lake near this, back of which is a rocky bank, and thence
the land rises eastward in low and broad terraces. On the green-
sward of one of these, not far from the bank, two of these mounds
are conspicuous, one being a little above the other, and the edges
meeting. At this spot they are the only ones in sight, and both
have the characteristic circular form and depressed center. A little
digging has been done in each, but this has affected the appearance
very little. Though a little shaded they are practically in open
ground. They are not of the largest size. The upper and eastern
one has an extreme breadth of 34 feet, and an inside diameter of
14 feet from the interior slope. This is about 2 feet deep, within
and without. The western one is of the same outside hight, and is
32 feet across the base. The inside width is 17 feet, and the depth

12 NEW YORK STATE Ml/SEUM

3 feet. Some digging has been done in the center. The disturbed
earth is black, containing burnt stones, but there are no signs of
organic or artificial remains. Not far away there are many spots
where the flat rocks form the natural surface, and about these the
spring saxifrage was abundantly in bloom. The low, symmetric
mounds themselves formed a pleasing feature of the scene, full in
view from the modern ruins as the land descended toward the lake.
The spot is so convenient and beautiful that one might have expected
to find more there, but for the evident tendency to place them in
pairs or small groups.

It was late in the morning, and no satisfactory photographs could
be obtained from lack of shadows. At a subsequent visit many
mounds were examined in the rain, and others in the depths of
woods and undergrowth. Many sketches were made, some of which
are here given, but in no place could the camera be used to any
great advantage.

A second visit was made in the middle of September 1901. A
map of the vicinity had been secured, on a scale of a mile to the
inch, and the general grouping will appear on this, shown on plate I.
As there was no special plan in the location of these structures, no
necessity is felt for more exact details. They were placed where per-
sonal or family taste or convenience required. No rule appears in
this except ease of access to the lake or streams. Some were on
quite elevated land ; others on broad hummocks, surrounded by
marshy spots but little above the lake. In a few cases they were
on the high banks above rocky streams, at some distance from the
shore. The unpropitious weather prevented a personal examination
of those at the south end.

As far as I could ascertain there are none now remaining on the
west side of Hyde creek and northwest side of Perch lake. The lake
may once have been higher than now and thus larger, but this did
not affect the situation of the mounds, nor their probable age.
Beginning on the west side of Hyde creek a long line of cliffs runs
parallel with the present shore toward the southwest, and between
these and the lake is a broad expanse of swampy land, well covered
with trees. No one could have lived in this swamp, nor was access

PERCH LAKE AND OTHER NEW YORK MOUNDS 13

to the shore through it in the least easy. I examined the undis-
turbed land at the top of these cliffs for a long distance, without
finding a trace of aboriginal life. Every favorable indication was
carefully examined, but nothing appeared. There may have been
obliterated dwellings in the cultivated land farther back, but this is
not probable. The swamp was an undesirable barrier to the lake.

Farther north, on the west side of Hyde creek, the case was
different. That stream came fairly near the rocky uplands, afford-
ing an easy passage to the lake. Accordingly a few mounds were
reported there, though none seem to remain. Certainly they were
few. I was told of two mounds leveled by my informant on the
A. J. Dillenbeck farm in 1901. These were 5 rods west of the
swamp and 30 rods from the lake. In plowing there he found a
broken flint knife, a fragment of pottery and a pottery rim, all of
which he gave me. From the character of the rim I think there is
an error of location. These were all the mounds of which I could
learn on that side.

Mr S. Getman said that he found a celt near two mounds he plowed
up on the south part of his farm, at an early day, on the higher ter-
race east of the creek. I found no existing mounds as far north as
this. A celt and arrowheads were reported from two mounds
destroyed in 1900, on the upper terrace of the Timmerman farm.
These had disappeared. The dual arrangement may be observed in
all these mounds. It is probable that many mounds have long dis-
appeared from this higher cultivated land. Those remaining are on
the stony lower terraces. Commencing south of the Getman farm
they extend along the shore to a stream called Ruff's creek by some.
South of this swampy lands again appear by the lake. This eastern
shore is mostly high and rocky, rising thence in terraces, and the
mounds appear here and there all the way. Some mounds may have
escaped my attention in the undergrowth on the Van de Walker
farms.

A medium sized mound was opened on the farm next south
of S. Getman's. Plate 4 shows this, on the second terrace east of
Hyde creek and not far from it. It is 30 feet across and 2 feet high,
with a broad central depression. A rectangular fireplace in the

14 NEW YORK STATE MUSEUM

center was 8 feet across and edged with upright flat stones. This
went down 2 feet below the present surface, the earth having been
removed for the fireplace, and cast back as a foundation for the ring.
Plate 12, figure 2, shows the surface plan. A is the outer slope, B
the top of the ridge, C the inside slope, D the fireplace edged with
stones, which is not an invariable feature. There were many coals
in the black earth, no ashes and no vestige of anything else. There
were many large stones. A trench was carried through to the
original surface, and shorter cross-sections were made. A little
southeast was another of similar size, rather flat and not prominent.

Another, farther south, is near the north line of the Timmerman
farm. This is about 36 feet across and 2 feet high. It is flatter than
most on top, but shows the usual depression. Small trees are grow-
ing on it, and there are some large stones along the edge. They may
have been dug out of it, for most mounds have a few such stones.
Plate 5 shows this.

A mound on the Timmerman farm has a large hemlock stump on
it, and some small trees. It is a continuation of a low ridge, so that
its exact dimensions are modified by this. As measured it is 28 feet
wide by 2 feet high. The hemlock stump might show how old it
must be, but not how old it might be. In these descriptions the gen-
eral course is from north to south. A low ring, 19 feet across and
on the same farm, tends to show that growth in hight and width was
slow, and by removals of matter from center to circumference; pos-
sibly by additions without The depression is 8 feet across, but it
was not noticed whether there was an inclosed fireplace. Plate 6 is
of this. Another, west of the fence and this, is broad and low.
Still another small one is on the lower terrace, not far away. Both
these are northwest of the next.

A high mound on the edge of the upper terrace, and just west of
the fence which crosses it, was not measured across, but is 3 feet high
and with a deep central depression. A large stump is on the south
side of this. Part of its effect is lost from its surroundings. There
is an obscure one on the lower terrace a little west. Another low one
with a wide and deep depression is on the same farm, and is
shown in plate 7. It is about 21 feet wide, the hight being usually

PERCH LAKE AND OTHER NEW YORK MOUNDS 15

in proportion to the width, and is an excellent example in its
symmetry.

On the same farm is another 27 feet wide and 2 feet high, having
a deep depression. A small one is just west of this, and perhaps
related.

Another is 30 feet wide and 2 feet high, having a depression.
There are some large stones outside of this. As the mound ro'se and
the ring grew, it may have taken in loose boulders around, which
had no relation to it. There is an obscure one north of this, and near
a shanty in the woods. One on the south side of the shanty was 2
feet high.

A gravel bed, which has been opened in the woods, was cut
through one of these mounds, in such a way as to give a good
exposure. The bed reached above the mound on the east or upper
side, the mound terminating a ridge, and nothing is seen in the
exposure there. Another cut has been made in the mound below.
At the base is coarse gravel, in its natural condition. Over this is a
stratum of black earth, 3 feet deep and about 27 feet wide. The top
and the extremities of the mound remain. Nothing was found in
obtaining the gravel, nor was anything discovered in our farther
digging. Not far from this, by the fence on the upper terrace, was
another large one.

The finest mound on the Timmerman farm is near its southwest
corner, in an open field and near the head of Perch lake. It is 33
feet wide and about 5 feet high. Plate 8 is of this. It is at the foot
of a bold hillside, and itself on high ground. Digging had been done
there, and the Rev. Mr Scott is said to have obtained pottery and
other things in 1901. We dug but little, and found nothing. Other
low mounds were near toward the lake, and there is a large flat one
quite a distance north. South of this, group is low land for some
distance, crossed by a rapid stream.

Leaving this stream and the low land the woods on the Sayles
farm are reached, where there are many evergreens and a rocky shore
along the lake. In these dark woods are other mounds. One is near
the north end, and has its east side more elevated than the other,
apparently from the slope of the land. This has quite a deep

1 6 NEW YORK STATE MUSEUM

depression, and is of the usual size. A low and broad one is north of
this, and two smaller ones south. Another large one is farther
southwrest. The depression is deep.

South of these woods is lower land and a brook, both north of the
large point. A fine mound is on a low bluff in this bay. There are
several mounds not far off. One is on the low bank near the shore ;
another south of this on a knoll or slightly higher land. Another,
beyond the last and on the lake side of a knoll, has a very deep and
large depression, 13 feet across. There is another at the south end
of this ridge, and another in the low land east. Two are on the
lower bank farther south, and there is a confluent group at the
north base of the high terrace on which the La Farge mansion stood.
The situation of these in these low lands is singular, though some of
them are dry enough, and the spot is unusually sheltered.

South of the brook, as the large point is turned, there appear deep
depressions and slight rings of an undecided character, and then
come the two fine mounds in front of the ruined La Farge mansion,
already described. Plate 9 shows the upper and larger one. South
of this is low land, through which a large stream enters the lake.
Beyond this creek no mounds appear for nearly half a mile, though
some have probably been destroyed. Then one with a deep and
broad depression appeared on a high bluff in the edge of the woods.
It was about 27 feet across. High rocks here fringed the lake, with
terraces above, on which was much undergrowth.

Nearly a quarter of a mile beyond was a doubtful one, not having
a complete circle, and a similar one was on the edge of a knoll
beyond. It is difficult to estimate distances while looking for mounds
in thick woods, and it may be another quarter of a mile to a large
and fine one on the second terrace. This was 27 feet wide, 2 feet
high, and has a depression of 3 feet in the center. As before
observed, it was usual to dig out the center in beginning these
mounds. Not far away are two large ones, close together on the
high bluff above the lake. Another just beyond is 36 feet wide, and
the depression is 16 feet across. Another obscure one is farther
south near the high bank of the lake. Between this and the cottages
and boathouses beyond, is one 36 feet wide, 3 feet deep inside, and
2 feet outside. This is fine.

PERCH LAKE AND OTHER NEW YORK MOUNDS I/

A stream enters the lake at the boathouses. South of this and east
of the shore is a large and high mound in which digging has been
done. In the freshly turned earth no vestiges of early occupation
could be seen except black earth and burnt stone. This mound is
about 5 feet high and 40 feet across. Common pottery was found
in small camps by the shore. Beyond this is another low mound.
Probably some in these woods were not observed.

After leaving the woodland the swamp is soon reached, and some
mounds may have been obliterated in the open fields. North of the
swamp flows a small rocky stream through these fields. On the brow
of the upper terrace, on the north side of this is the largest and
deepest mound I saw, measuring 45 feet across. Another is close
beside this on the east, and another on higher ground still, in the rear
.of these and toward the road. On the south side of the stream,
farther down and overlooking the water from a high bank, is
another small but deep one under a tree. It is a pretty situation.

A large mound is near a shanty in the sugar camp, toward Ruff's
creek. This is 40 feet across and 3 feet high. A good deal of dig-
ging has been done there, but seemingly without results. A smaller
one is near the shanty. This ended the explorations on that side of
the lake. The oldest inhabitant knew of nothing taken from mounds
south of the La Farge mansion. In all 54 mounds were observed,
and 6 obliterated ones reported, or 60 in all. Other unobserved or
obliterated mounds might much increase this number, but it is not
likely to reach the higher estimates made for the whole territory.

At the north end of the lake is one spot deserving of a few words,
and yet probably not connected with the general subject. In the edge
of the swamp at the northeast angle of the lake, is an immense mass
of rock which can be reached by a boat. In some of the depressions
of this rock are many small flint chips, showing that it was a favorite
spot for arrow makers. What weapons the makers of these mounds
used is uncertain, but it is probable that the visitors to Squaw island,
as some call it, were of another people. The spot commands' a view
of nearly the entire lake.

Dr A. A. Getman and Oren Pomeroy, of Chaumont, kindly made a
close examination of the group I could not visit and with much the

l8 NEW YORK STATE MUSEUM

same results. Both are experienced and careful observers, and for
this reason I give Dr Getman's account, written Nov. 6, 1901.
He wrote:

We went to Linnell island today. It is a limestone terrace, sur-
rounded by lowland and swamp. . . Mr Gailey says the island
contains about 500 acres, with three farms at present. The soil is
clay and a gravelly loam, with abundant outcroppings of rock (Chazy
limestone). From the map you see the mounds dot the crest of the
terrace all around the island ; some at least % of a mile apart. Some
of them appear to be built on the rock. In fact we dug the center of
one down to see that it was started on the solid rock. We dug on
this one near the barn. It had no central depression. We dug the
center to the bottom ; pit 4 feet square ; then commenced at the edge
on the south, and opened to this pit. There are less stones and more
soil than at the head of the lake, but we found lots of large hard
heads, 8 to 12 inches across, about halfway from the edge to the
center. All were burned. The depth was 2 to 4 feet from circum-
ference to center. There were streaks of very dark earth and char-
coal. Some of the charcoal was small limbs, 2 to 4 inches long, by
i inch through. Three of the mounds have been removed for road
building. They appear to make excellent roads. (These are marked
A, A on the map, plate 2.) We saw some recent plowing which
exposed three more. They were near those removed.

We looked over the three that had been used for roads. They had
been only partially removed. The manner of working had been to
plow the soil loose and shovel on the wagons ; three men to beam the
plow. In the plowed field some were smoothly plowed; on others
the plow ran out. We found in the one near the house of J.
Gailey, marked S, a skull and teeth of a muskrat, badly decayed,
and a piece of broken stone that looked as if used for a nut stone.
That is absolutely everything, except burned stones of all
descriptions, charcoal and dirt; unless a few pieces of reddish
crumbly pieces of stones were paint stones. S also shows some
excavated mounds.

We looked over the plowed field, pawed over debris of the road
mounds, and looked over the road that the stones were used on — a
private road to the Klock farm.

How many mounds there are I have no idea. We counted about
20, but there have been and are many more. I think, as a general
thing, that they are smaller than those at the head of the lake, and
of less depth. Mr Clarence Gailey claimed to have found two
arrows when working the road business, but could not produce the
find. It is very perplexing that no authentic relics can be found and
handled. Pomeroy says the mounds are similar to the one we saw
on Fox island ; that is the contents, stone, soil, etc.

PERCH LAKE AM) OTHER NEW YORK MOUNDS 19

Mr R. D. Loveland, of Watertown, found a few small fragments
of pottery in the large mound near the boathouse, but did not pre-
serve them, as he obtained larger pieces of the usual types near the.
shore, where these might be expected. Had none been found else-
where those in the mound might be thought intrusive. A clay pipe
was also found near the shore.

In the Bulletin of Natural History of New Brunswick for 1884,
p. 14, Mr G. F. Mayhew gives his ideas of how such hut rings were
gradually raised. I infer that these might have been well known
there, but am not certain on this point. He supposed that in some
circular lodges moderate cleanliness was preserved, not by removing
refuse altogether, but by drawing it back and filling in the center
with fresh gravel. A constant repetition of this would preserve the
circular form and the central depression as the mound rose above
the surface. Bone needles found 'near the edge he thought had been
stowed at the back of a couch. Pottery was much like that found
elsewhere. Arrowheads were made by the fire, few flakes being seen
out of doors. Most ordinary aboriginal implements were found.

This general mode of elevation reasonably accounts for the form
and growth of the Perch lake mounds. A pit was made in the
center for the fire, and a large circle was thrown up at the edge of
the lodge to carry off water. A slope from this to' the fireplace gave
an easy position to the reclining people within. It was necessary to
remove or rake the embers away, and the edge gradually rose. To
make it cleaner it was as easy to bring in a fresh supply of dirt as
to carry accumulations away. In all this there was a natural over-
flow which enlarged the borders of the mound. The original fire-
place was all the time retained, and so the largest mounds are the
deepest.

Mr Harlan I. Smith suggests a likeness in these to some he recently
examined on the Pacific coast, and I give plate 10, figure I, to illus-
trate this point. In his explorations he found that up to a recent
date the Thompson river Indians made huts of this kind. In this
section a is an excavation in the ground, which increased the hight
of the interior of the lodge, and supplied material for its covering.
Around this excavation an arch b was raised, resting on the surface

2O NEW YORK STATE MUSEUM

c. This arch had a frame of saplings and branches underneath,
covered with dirt and sods. In the center of this strong frame above
an opening, d, was left. This was reached by a primitive ladder, e,
niade of a notched pole which gave strength to the roof. This was
the only means of ingress or egress for light, smoke and Indians.
He found one still standing in a dilapidated condition, but observed
the remains of many. When the roof fell in, a low mound was
formed, with a marked central depression. These would probably
differ from the Perch lake mounds in the size of this depression, the
hight of the circle, and the evidences of fire throughout. The latter
were probably simple tepees, pitched from time to time on the same
spot, but not continuously occupied. The accumulation was gradual,
but earth might have covered the lower part of the wall.

In his report on Mound Explorations, Prof. Cyrus Thomas
described some mounds of this class closely connected with larger
mounds in the Welch group, Brown county, Illinois. The group
" consists of six mounds, and a number of small saucer-shaped basins
surrounded by low, earthen ridges, doubtless the sites of ancient
dwellings or wigwams." Thomas, p.iiS. He adds that " the dwell-
ing sites vary considerably in size, some being as much as 70 feet
in diameter, and some of them 3 feet deep in the center after 50
years of cultivation." In describing those on the Big Mary river,
111., he adds something on their situation and origin :

These are situated upon a flat topped ridge, about 30 feet higher
than the creek bottoms. They are low, with the usual depression in
the center, but the outlines are rather indistinct. Mr Gault of Sparta,
who has long resided here, states that when he first moved to this
section, the Indians lived in houses or wigwams which, when de-
cayed, left such remains as these. They hollowed out a shallow cir-
cular cavity in the surface soil, then, standing poles around the
margin of this basin, brought them together at the top, and having
covered them with bark or other material — in other words having
constructed wigwams of the usual circular form — covered them in
whole or in part, specially the lower portion — with earth. He also
said that after a camp was abandoned and the wood rotted away, it
left these rings of earth. Thomas, p. 141

In one inclosure near Lakeville, Stoddard county, Missouri, he
says,

PERCH LAKE AND OTHER NEW YORK MOUNDS 21

Nearly the whole space between the walls is occupied by the hut
rings or circular depressions. They are of the usual size, 20 to 50
feet across, and i to 3 feet deep. Thomas, p. 174

These contained ashes, pottery, etc., and he mentions no ridges.
In another group the rings varied from 21 to 34 feet across. In an-
other large group we get a more definite idea of the elevation, a fea-
ture in which most of these seem to differ from those of Perch lake.
This is at Beckwith's fort, Mississippi county, Missouri. After
describing the inclosure' he says of the hut rings :

These almost cover the remainder of the area, the only open space
of any considerable size being the 200 feet square just east of the
large mound. They are not confined to the natural level of the in-
closure, as some are found on the level tops of the mounds. They are
circular in form, varying from 30 to 50 feet in diameter, measuring
to the tops of their rims, which are raised slightly above the natural
level. The depth of the depression at the center is from 2 to 3 feet.
Near the center, somewhat covered with earth, are usually found the
baked earth, charcoal, and ashes of ancient fires, and around these
and beneath the rims split bones and fresh-water shells. Often
mingled with this refuse material are rude stone implements and
fragments of pottery. The similarity in the size, form, and general
appearance of these depressions and earthen rings to those of the
earth lodges of the abandoned Mandan towns along the Missouri
river, leaves no doubt that they mark the dwelling sites of the people
who formerly occupied this locality. Thomas, p. 187

These mere depressions illustrate but one feature of the Perch
lake mounds, and we have a closer correspondence in those described
by Prof. F. W. Putnam in the nth report of the Peabody Museum,
and quoted by Mr Thomas. They were some observed by the
former in Tennessee, and thus described:

Scattered irregularly within the inclosure [the earthen wall which
inclosed the area] are nearly one hundred more or less defined
circular ridges of earth, which are from a few inches to a little over
3 feet in hight, and of diameters varying from 10 to 50 feet. . .
An examination of these numerous low mounds, or rather earth
rings, as there could generally be traced a central depression, soon
convinced me that I had before me the remains of the dwellings of
the people who had erected the large mound, made the earthen
embankment, buried their dead in the stone graves, and lived in this
fortified town. Thomas, p.662

22 NEW YORK STATE MUSEUM

Professor Thomas adds that these hut rings " are seldom, if ever,
met with except on the site of an ancient village, and often one that
was defended by an inclosure." This again differentiates the
western and southern forms from those of New York. The latter
are scattered or in very small groups, have the depressed center
very little below the natural surface, are usually of considerable
hight, show the action of fire, but rarely contain ashes or relics of
any kind, have no bones or shells, and the earth of which they are
composed has been gradually gathered from year to year. With
all this difference there is an unmistakable likeness, and no hesitation
is felt in calling them the foundations of early lodges.

Two things naturally arrest attention. There are no bones or
shells revealing the food of the inhabitants, though the conditions
are favorable for their preservation. Most of them contain no
articles made by man. The favorite fresh-water clam of the New
York Indians was U n i o c o m p 1 a n a t u s . It is so widely dis-
tributed that it probably occurs in Perch lake, though I observed
none along the shores. If it is not found there that part of the
problem is solved. But these aborigines were there for the fish of
the lake, as well as for the game in the woods. Their homes had
an easy access to the water. Why are no bones of any kind found
under these conditions? The Iroquois sites yield them abundantly.
It may be due to an Algonquin superstition. All will agree that
these were not Iroquois homes, for they rarely used the circular
lodge, which the Algonquins commonly preferred. There were
differing tastes and beliefs of other kinds. The Iroquois left bones
of every description on the ground. The Algonquin scrupulously
gathered up many kinds, and either threw them in the water, or
burned them in the fire.

A single quotation from the Relation of 1634 will illustrate this.
The missionary said:

The savages do not throw the bones of the beaver to the dogs,
or of female porcupines, at least certain special bones ; in short they
very carefully take pains that the dogs shall eat no bone of birds
or* other animals which are taken in a snare. Otherwise they will
take no more except with the greatest difficulty; besides there are
within a thousand observations, for it is important only that the

PERCH LAKE AND OTHER NEW YORK MOUNDS 23

vertebrae or the rump alone should be given to these dogs, the rest
must be thrown into the fire; still, for a beaver taken in a snare it
is better to throw his bones into a river; it is a strange thing that
they gather and pick up these things, and preserve them with so
much care that you would say their hunt had been lost had they
gone contrary to their superstitions. As I ridiculed them and told
them that the beavers did not know what was done with their bones,
they replied to me: You do not know how to take beavers, and
you wish to talk about them ; before the beaver is entirely dead,
they said to me, his soul comes around by the cabin of the one who
killed him, and notices carefully what they do with his bones; that
if one gave them to the dogs the other beavers would be warned
of it; that is why they would render themselves hard to catch: but
they are very glad if they throw their bones into the fire or into a
river, the snare especially, which has taken them, is well pleased.
I told them that the Iroquois, as is done among us, threw the bones
of the beaver to the dogs, and yet they very often took some, and
that our French, beyond comparison, were accustomed to take more
game than they, and yet our dogs were accustomed to eat the bones.

The Algonquins, of that day, extended this rule to fish, and it may
have had wider applications still. To leave no permanent memorial
it was necessary only to care for the bones on the lodge site. Out-
side of the circle they would soon perish, and this superstition pre-
vented their casting them there. These lodges had no dumping
places; everything was disposed of on the spot.

In referring these mounds to the Algonquin family another fact
is explained. These nations may not have been without earthen-
ware, and perhaps most of them were not, in a limited way, but it
was not so common as with the Iroquois and others. They were
nomadic, and the lightest vessel possible suited them best. It was
particularly necessary to have one not easily broken, and that could
be readily replaced on a journey. Toward and north of the St
Lawrence the canoe birch abounded, and of this material their cook-
ing vessels were formed. Their cooking was not very thorough,
and hot stones, dropped into the water, heated it enough for their
needs.

Why arrowheads are not found, nor other stone implements as
a rule, is a more difficult question, but capable of various answers.
There were careful aborigines, those who lost little, as well as those
careless and wasteful. Articles were not so readily lost, but more

24 NEW YORK STATE MUSEUM

readily found, in a cabin than in a village. The wooden arrow
might have sufficed for most of the needs of the place. Some have
suggested that -the huts were those of a recent day, and that no
purely Indian relics may be expected. I do not assent to this view,
nor am I prepared to say with Mr Marvin, that these forest men
have left us traces of the oldest habitations in the State. The fact
seems to be, however, that we must make these very modern, with
but little to sustain this view, or place them before the Iroquois
occupancy of New York and the St Lawrence. Till the Iroquois
sold their lands there has been no time within the last 300 years
when it would have been safe for Algonquins to have habitations
on Perch lake. For a century before that, at least, Jefferson county
was occupied by the Iroquoian family, and they had no wish* for
intruders. How much these mounds antedate the last four centuries
is a harder problem. I think they may safely be placed within the
past 500 years. Traditionally the Algonquin and Iroquois family
arrived here nearly together, and at no remote period. An exam-
ination of the sites of their camps and towns seems to substantiate
this, and these mounds suggest a period antedating that of their
inveterate hostility. They were undefended, long used, and yet
wrere in a territory claimed and held by the Iroquois for hundreds of
years.

Two maps of the vicinity are given ; one from a large county map,
and the other from the public topographic survey, conspicuously
differing in some respects. In the latter, plate 3, the lake is much
shorter than in the former, and streams which enter the river
in one flow into the lake in the other. The difference may be
accounted for by the fact that part of the swampy shores were once
included in the lake, when the water supply was greater. On the
former map the general range of the mounds is indicated by the
usual sign.

OTHER NEW YORK MOUNDS

A few supplementary remarks may be made on other mounds in
New York, the larger part of the State having none, and most of
those found being of small size and simple character. In some
cases natural formations have been mistaken for these, having been

PERCH LAKE AND OTHER NEW YORK MOUNDS 25

used for burial or camps. In 27 counties some form of mound
has been reported and a summary of these follows. They are most
frequent west of the center of the State, and will be mentioned by
counties.

Several occurred- in Allegany county, and thence westward they
were frequent. In the town of Conewango, Cattaraugus county,
was a tumulus 13 feet high, with a diameter of 61 by 65 feet.
Skeletons were found with relics. In the village of Randolph was
a burial mound 10 feet high and 35 feet in diameter. In the town
of Bucktooth, north side of the Allegheny, was a burial mound, 39
feet in diameter and 10 feet high. Another was in the town of
Napoli, on Cold Spring creek, which was 120 feet around. At
Olean were several of these, one being 40 by 60 feet in diameter
and nearly 10 feet high. One in Dayton was of the same hight,
and 1 20 feet in circumference. Another was on the west side of
the Allegheny river, in the town of Cold Spring. This has been
reported as 200 feet around and 20 feet high ; probably an exaggera-
tion. On Cold Spring creek, 2 miles from the Allegheny, were two
burial mounds, 10 feet high and 100 feet around. Others were in
the towns of Leon and Conewango, in one of which were 8 sitting
skeletons.

Quite a number were in Chautauqua county. One at Cassadaga
lake was 7 feet high and 30 feet in diameter. A stone mound near
a fort in Ellington was 4 feet wide and 5 feet high. Two mounds
near Griffith's point, Chautauqua lake, were once 12 feet high and
40 feet in diameter. A number of similar mounds have been
reported on both shores, and two near Jamestown. Another, near
Rutledge, was 20 feet in diameter and 6 feet high. One in the
village of Fredonia was 7 feet high. Another at Fluvanna seems
recent. Most mounds west of Cayuga lake were sepulchral.

Near Spring Lake, in Cayuga county, were small mounds with
human remains, but these may have been incidental, as in some
other places. On the high land of Howland island, near the river,
are one or two suggestive of Perch lake. One is not very distinct,
but the other stands out plainly. It is a circle with a diameter of
37 feet and an elevation of 30 inches, inclosing burnt earth and

26 NEW YORK STATE MUSEUM

stone, but yielding" no relics. The earth is in its natural condition
for a considerable distance around. Some pits within the circle
may be the work of explorers. This I examined July 18, 1902.

The noted burial mound in Greene, Chenango county, was 40
feet in diameter and 6 feet high. Several hut rings have been
reported along the Chenango river, similar to those at Perch lake,
but those at Indian brook, a little south of Greene, prove to be
caches.

Columbia and Schoharie counties both had the stone heaps to
which Indians added stones in passing.

Erie county had its full share of mounds. One at the mouth of
Cattaraugus creek was used for burial, but was probably natural.
It was 50 feet across and from 10 to 15 feet high. The relics were
modern.

There were several burial mounds on the east side of Cattaraugus
creek, two of which were excavated by Dr A. L. Benedict of Buffalo,
in 1900. As good accounts of such work by competent observers
are rare in New York, his plans are given in plate n, and his report
is summarized from the American Antiquarian for 1901, p.99-io7.

No. i, a truncated mound in an open field when I saw it, is a mile
north of the creek, and 600 feet north of the high bank of the ancient
valley. It is nearly circular and about 70 feet in diameter. The
central hight is 4 feet, 8 inches, but he thought it was originally
10 or ii feet high. It was made of sand loam, and there were
depressions north and south in the general level of the field. It
had been disturbed. At A were animal bones, ashes and charcoal
at 3 feet, 5 inches from the surface; also bones of the aboriginal
dog. At B was a heap of flat pieces of Hamilton slates, some of
them waterworn. A rib and sacrum under these he thought those
of the musk ox. At C was a fragmentary human skull, with other
human bones, at a depth of 3 feet. Near this were flint arrows and
knife, flint chips occurring elsewhere in the mound. Dr Benedict's
plans ^ have the top to the south.

No. 2 resembles the last and has been reduced by plowing. It is
quite near the creek, and a central shaft was sunk below the original
soil in 1875 by William C. Bryant of Buffalo. Gravel was found

PERCH LAKE AND OTHER NEW YORK MOUNDS 2/

4 feet below the level at A by Dr Benedict, and this occurred at
4 feet, 9 inches at H. At F was an oblong fireplace of water worn
stones. Between the top stone and one on the west side of the
inclosure part of a pottery rim was found. There were small
sherds at H. In the ashes under the top stone were calcined bones.
A human astragalus was found at B, 4 feet southwest of the central
stake, at a depth of I foot, 9 inches, covered with several round
stones, 6 inches to a foot in diameter and an inch thick. A
calcined long bone was found 15 feet south of the stake, which
seemed part of a human tibia. At 7 feet, 10 inches south from the
stake the bottom of the mound was of burnt clay and gravel, about

6 inches thick. Below this was a hollow space, beginning 3 feet,

7 inches from the surface of the mound. This was 9 or 10 inches
deep, and extended every way 2 or 3 feet. The floor of this was
of coarse gravel, about the size of hickory nuts, blackened, but
showing no disturbance. Charred ,wood was occasionally found,
some of considerable size. There were also small bits of mica.
These seem hardly true burial mounds, though containing human
bones.

Other mounds have been reported in Erie county 15 to 16 feet
high and from 45 to 54 feet in diameter. One near the Indian fort
at Buffalo was 5 or 6 feet high, and from 35 to 40 feet across. It is
probable that Dr Benedict's diameters may be too great, allowance
not being made for increase at the base by washing down from
above.

On St Regis island, Franklin county, was a mound 8 feet high,
and another opposite, on the east bank of St Regis river. Burial
mounds were frequent along the St Lawrence.

Small mounds have been reported on Tonawanda creek, in
Genesee county, but they may not have been artificial, though used
for sepulture. The mound at the Bone fort, near Caryville, was
6 feet high and 30 feet wide, almost entirely composed of bones.

Two small mounds have been reported in Jefferson county, and
many hut rings on the east bank of Black river, Lewis county,
opposite the Deer River station. These are like those at Perch lake.

In the summer of 1903 an early and notable ossuary was discov-

28 NEW YORK STATE MUSEUM

ered by Mrs R. D. Loveland of Watertown, near the long carrying-
place at the head of Chaumont bay. A curious depression arrested
her attention, and a little digging brought to light a human skull.
She then turned over the search to others, who unfortunately had
not her knowledge and skill, and no clear description is available
from them. Dr R. W. Amidon afterward visited the place, saw
the relics, and obtained what information he could. Its importance
comes from its age, the relics being mostly of early types. The pit
is near and below the end of a ledge of Trenton limestone. At
least 8 skeletons of vigorous adults were unearthed, from 2 to 4
feet below the surface, and mostly covered with boulders and flat
stones. Two skulls were fractured, as though by a war club. A
perfect clay vessel was destroyed by the diggers, but it was of a
frequent form. A bird amulet, of green striped slate, was found.
This was $l/2 inches long, rather broad, and with the head and tail
almost on a plane with the body. There was also a bar amulet of
sandstone, 6 inches long, and a perfect soapstone pipe of a frequent
form. A flat bone bead, bone and horn implements, flint arrow-
heads or knives, were among other articles. This ossuary thus
gives us some idea of what other things were used by those who
had these amulets.

In Livingston county there was once a mound in the road from
Dansville to Groveland, which was 4 or 5 feet high and 30 feet in
diameter. Another was midway between Dansville and Scottsburg.
A burial mound was 2^ miles southeast of the head of Hemlock
lake. One at Mt Morris was used for recent sepulture, but
may not have been artificial, as it is said to have been 100 feet
across and 8 to 10 feet high. Some accounts make the relics of
early types, and it is probable it was used at various periods. On
the Genesee river, near the Wheatland line, was a burial mound
8 feet high. Two mounds are also on the Wadsworth farm near
Geneseo. One is 3 feet high, but not quite 25 feet across ; the other
is much smaller. Both have been reduced in size. A stone heap
at Lima traditionally had a memorial character.

In Madison county, on Oneida lake, are supposed Indian mounds,,
which are probably natural formations.

PERCH LAKE AND OTHER NEW YORK MOUNDS 29

There was a mound in Monroe county, a few miles northwest
of Scottsville. Two small mounds were west of Irondequoit bay,
on high land, the largest being less than 5 feet high. A large one
was east of the bay, and another, east of the village of Penfield, was
originally 40 feet in diameter and 8 or 9 feet high. Two burial
mounds were on the east bank of Genesee river, half a mile below
the lower fall. They were 4 feet high and 20 to 25 feet wide.
There were other mounds in that vicinity. In Pitts ford was a pile
of large limestone boulders, the heap being about 12 feet square.
Between Irondequoit landing and the lake was a cemetery of 200
small grave mounds, arranged in rows. The further character was
not reported, but single graves are usually depressed. A mound
was on the bluff south of Dunbar hollow, which contained stone
implements. Mr Harris thought a small island on the west side of
Irondequoit bay was mostly artificial, as proved by excavations arid
grading. It was 90 feet long, 32 wide and 17 feet high, but was
not sepulchral, though it contained many fine articles at a depth
of 15 feet.

A mound described in Cambria, Niagara county, should be called
an ossuary and contained metallic articles. A stone mound has
been reported a mile west of Lockport, and an ordinary one at Gas-
port. Two burial mounds of large size were on Tonawanda island.
Another was in Wilson, and two in the town of Lewiston. In Sep-
tember 1903, the one marked D on Schoolcraft's map was opened.
He called it " a small mound or barrow," but if it ever had much
elevation cultivation had long before removed all signs of this. As
it has not before been described a brief account of it will be given
here. The first skull was found 6 or 8 inches below the present
level of the ground, and the skeletons were estimated at over 300.
Over 200 skulls were secured and none had been injured, the place
representing well the ossuaries of Canada. The date may have been
not far from 1620, perhaps a little later, while the Neutral nation
still occupied land in New York. The pit, excavated by Mr John
Mackay of Niagara Falls, was about 18 feet long and from 12 to
14 feet wide, with a depth of 3^ feet from the surface. The form
was an irregular ellipse, and the bottom was of rock and clay. To-

3<3 NEW YORK STATE MUSEUM

make more room the pit had been widened about 18 inches from the
bottom, and the smaller bones were placed in this addition. There
were no traces of any lining to the pit, nor any suggestions of
Jesuit contact, while earlier articles of European trade had reached
the spot, possibly from the Dutch through the Five Nations. There
were 24 iron axes, several brass kettles, 3 sword blades, 24 large
and curious brass rings, 5 cylindric brass or copper beads, with other
ornaments of shell. Through the kindness of Mr Mackay I exam-
ined a number of these. The rings are simply short brass cylinders,
bent in circles, and the beads are long brass tubes, precisely like
those occurring in the Mohawk valley. One of these is 1 1 inches
long and ^ inch in diameter. Mr Mackay has an interesting collec-
tion, well repaying study.

Some burial mounds have been reported in New York city,
apparently natural elevations used for sepulture.

Some supposed mounds in Oneida county are also of doubtful
character, nothing having been determined by examination.

In Onondaga county, near Baldwinsville, were two large stone
heaps, covering human bones, and two burial mounds were on the
west side of Onondaga outlet. One was circular and stood out
prominently from the bank behind it. The other was oblong, being
12 feet long and 3 feet high when I sketched it, and had then been
somewhat reduced.

At the modern Seneca castle near Geneva, where Johnson built
a fort in 1756, is an artificial mound about 6 feet high and used as
a cemetery. It is probably rather graded than built up. There was
a small recent mound at Clifton Springs.

In Carlton, Orleans county, on the north bank of Oak Orchard
creek, is a small oblong mound, 20 by 30 feet in diameter. Another
small mound was 30 rods away.

Bone hill, at Oswego Falls, was a place of sepulture, now known
to be of natural formation. It was 6 rods in diameter and 40 feet
high, and has been removed.

In Unadilla was a supposed Indian monument, 20 feet in diameter,
10 feet high, and of a conical form. There was a mound at
Oneonta, and a supposed burial mound at Cooperstown.

PERCH LAKE AND OTHER NEW YORK MOUNDS 31

In Tioga county there was a circular burial mound at Owego.
Several mounds were in the vicinity of Newark Valley. One of
these was 15 feet high and 250 feet around, suggesting natural
formation.

In Wayne county I examined several mounds July 18, 1902. One
was northwest of Savannah and in the midst of camp sites. It is
circular and but slightly separated from the ridge behind. It is 60
feet across and 3 feet high. Another burial mound north of Crusoe
creek and northeast of this, is now small and low, but distinct.
Another was examined 2^/2 miles south of Savannah. It is at the
south end of a ridge containing caches, from which it has been
separated by excavation. The bodies were apparently laid on the
surface and the earth heaped upon them. It is 30 feet across and
about 7 feet high. The first of these mounds shows little work.

In Wyoming county is a burial mound about 4 miles south of
Portage.

In Yates county a small burial mound on Bluff point is 9 feet
long and 4 feet high.

These are all the burial or monumental mounds thus far reported
in New York, as distinguished from defensive earthworks. Very
few indeed resemble those of Perch lake, and this led to the special
examination of the latter. Their peculiar character is emphasized
by this comparison with New York mounds elsewhere, and though
scattered examples may yet be found here, it is quite probable that
nowhere else in the State will they be. seen in such numbers or in
such fine preservation. They form a unique group, well worthy of
further study, though offering little in the way of fine relics, or
indeed of any at all.

By way of caution it should be remarked that the hight of mounds
is commonly made too great unless accurately determined ; and there
is also a disposition to consider any symmetric elevation of moderate
size an Indian mound. Even when human bones are found in them
they are not always artificial.

A curious spot il/2 miles west-southwest of Unadilla may be
described here, having never been mentioned before. For the
account and chart, thanks are due to Mr Harry B. Cecil of that

32 NEW YORK STATE MUSEUM

place. It is on the farm of Enoch H. Copley and in. a woodland
of about 33 acres, the whole of which is a series of moraines and
kettle-shaped hollows. In the largest of these hollows is a shallow
pond, marked A in the diagram, plate 12, figure I. The shaded
part B has been partly filled in for the Delaware & Hudson Rail-
road. The pond is surrounded by moraines, C C C, about 100 feet
high, and a road D, follows the north and east margins. At E, F, G,
are rude stone walls from 2 to 4 feet high. Mr Cecil said :

At one time I supposed these had been constructed to get rid
of the rocks that were in the way, but this could not be the case,
as the stones could have been dumped into the pond very much
more easily, and it would have materially helped to widen the road
D. The oldest residents say that these piles and walls have always
been there. At H, until a short time ago, were two circles made of
rocks loosely thrown together. They measured 10 feet across and
were contiguous, having openings at the remote parts of their cir-
cumferences. I turned these over carefully, but failed to find
anything of Indian workmanship and the soil beneath was appar-
ently undisturbed. At I was another stone wall. At J is a heap
of undisturbed rocks. At K is a carefully made road, about 8 feet
wide and extending about 300 feet in a westerly direction, gradually
ascending to 50 feet above the pond level. No explanation can be
given of this unless it was part of a trail. Below this road and
above the wall at E, is a stone heap, and above the road is a large
hollow filled up with stones of all descriptions. I am positive that
these heaps are not natural. All these remains are included in
about half an acre.

This account is free from extravagance and suggests the use of
the spot as a pound for deer, terminating a driveway. These and
other animals would naturally resort there to drink. With or with-
out contracting hedges they would follow their own paths, and the
roadway would turn them toward the double walls, I, F, when driven.
Some would escape only to encounter other hunters at the wall G.
In the press others might turn back and meet hunters at the wall E.
The circles may have been the foundations of hunting lodges, and
the season of wild fowl would afford a secondary use. The usual
course was to make a pound of stakes and branches, but the primi-
tive hunter was quick to avail himself of natural advantages, and
was not sparing of work.

TRAILS

In the League of the Iroquois, Mr Morgan gave a definite and
interesting account of the principal Indian trails of New York, over-
looking some things which seriously affected his scheme. Sites of
Indian towns were constantly changing, and trails of necessity
changed and were forgotten. An abandoned forest path is soon
obliterated. All his towns were not properly located if the record
is to be considered two centuries old.- Much of the time Indians
took a general rather than a fixed course in going from place to
place, for the advantage of hunting or for other reasons. Thus
trails were very faint in some places, becoming plainer as they
approached towns. Remembering such things, Mr Morgan's gen-
eral plan will serve as a proper basis for some remarks on Indian
trails. At some period it may have been essentially correct, but in
the nature of things this was but for a brief time. With this reser-
vation it deserves high praise.

His scheme makes the trail leave Albany along the old turnpike,
going to a spring 6 miles west and thence to Schenectady, crossing
the river at the ford, where a bridge was afterward built. This
may be allowed, though it may not have been Van Curler's exact
route in 1634. Yet it is doubtful whether there was any trail or
much travel there before the Dutch came, for the Mahicans at
Albany were at war with the Mohawks west of Schoharie creek,
and made their footprints as light as possible. At Schenectady the
trail probably divided, when there was one, following both banks
of the Mohawk. When Van Curler followed that stream westward
in 1634, all the Mohawk towns were west of Schoharie creek, and
the Indians did not care to ford that for some reason. Curler
crossed the river, followed the north bank till the creek was passed,
and then recrossed to the south bank, where all the towns were.
When all these were on the north side, a few years later, there was
no use for the southern trail. When the south bank was occupied
it was used again. Mr Morgan's scheme places but one small
village on the north.

In 1634, and for some years later, all the Mohawk towns were
between Schoharie creek and Spraker's. His scheme places

34 NEW YORK STATE MUSEUM

Te-hon-da-lo'-ga at the mouth of the creek, Canajoharie on the
east bank of Ot-squa'-go creek at Fort Plain, and the upper Mohawk
castle in Danube, Herkimer county. Thence his trail went to Utica,
Whitesboro, Oriskany and Rome. This was a very recent route.

On the north he supposed that the trail turned off to Johnstown.
a modern feature, returning to the river at Fonda, and going thence
to Rome. This does not allow for the fact that as early as 1600
one Mohawk town was far up Cayadutta creek, another still farther
on the Garoga, and a third on the Ot-squa'-go, all several miles
from the Mohawk. Rome was not an objective point till western
trade became vigorous, and there was probably little travel that way
till the 1 8th century. Van Curler, in going to Oneida in 1634,
certainly left the Mohawk east of present Canajoharie, crossing the
hills to the upper waters of Oneida creek. Later accounts show
that this was long the only great trail, and this fact Morgan over-
looked. This affected his scheme beyond the portage. He said :

From Rome the main trail, taking a southwest direction, passed
through Verona, Te-o-na-tale', and finally came out at Oneida Castle.
This was the principal village of the Oneidas.

Knowing the latter was a recent town, Gen. John S. Clark placed
Old Oneida in the southwest corner of Vernon, where Sauthier's
map shows it. On his map of 1700 Colonel Romer marked his
route as leaving the Mohawk at the third castle, thence southwesterly
near the head of Schuyler lake, thence west to Old Oneida, whence
a branch trail led to the portage. Being on horseback his party took
the main road west to Onondaga.

From the modern Oneida Castle, Morgan's trail went through
Canastota, Canaseraga, Chittenango and the Deep Spring, Manlius
and Jamesville, to Onondaga Valley. No colonial traveler mentions
Deep Spring, though one of Gansevoort's officers spoke of it as
" Sunken Spring in the road," in 1779, and the" Onondagas tell me
that their favorite resting place was at Green lake, near Kirkville
and north of this route. Johnson and the Moravians alike show
that the main trail in 1756 was a mile south of Jamesville, entering
Onondaga valley far south of the turnpike. The Moravian journals
show that there were several trails between Oneida and Onondaga,
touching the Tuscarora towns.

TRAILS 35

Mr Morgan's three Onondaga villages are not well located, as
is easily proved. Ka-na-ta-go'-wa, or great village, is now where
he placed it, but it was 3 miles north of this in 1779, the farthest
north of all. There was no village at Gis-twe-ah'-na. In this
scheme the trail passed up the hill west of the present village of
Onondaga Valley, northwest through the sites of Camillus and
Elbridge, thence through Sennett and Auburn, crossing Owasco
creek just above the prison, following the old turnpike halfway to
Cayuga lake, then going direct to the old Cayuga ferry, half a mile;
above Cayuga bridge. It crossed the foot of the lake about 4 miles
farther north, at the old fording place near the lower bridge. This,
however, was not the trail used by the Moravians and others
in the middle of the i8th century. That went directly over the
hills from Onondaga, at that time 4 miles south of the present
canal, passed some miles south of Marcellus village, crossed the
foot of Skaneateles lake and that of Owasco, reaching the village of
Ganiatarage il/> miles north of Union Springs, intersecting there
the trail which connected the Cayuga villages east of the lake. This
was also Colonel Gansevoort's route in going eastward from the foot
of Cayuga lake to Fort Stanwix in 1779. The principal Cayuga
village was at Great Gully brook, 3 miles southeast of Union
Springs. From the mouth of this stream the lake was usually
crossed in canoes, and the trail went on to the foot of Seneca lake,
passing through the Cayuga village of Nuquiage, not far from that
lake. This is an historic route from Onondaga to the first Seneca
castle. That given by Mr Morgan seems much later.

In his scheme, after fording the foot of Cayuga lake, the trail
followed the north bank of Seneca river to Seneca lake. He noted
another trail, crossing the lake in canoes, and running west from
the shore to Seneca Falls. Thence it followed the south bank of
the river, intersecting the other trail at the lake, and running il/2
miles north to the first Seneca castle, near Geneva. Thence it fol-
lowed nearly the line of the turnpike to Canandaigua, at the foot of
that lake.

The Moravian journals make quite a difference here. From the
foot of Seneca lake they went 4 miles west-southwest to a deserted

36 NEW YORK STATE MUSEUM

town-, where the trail divided, one path going to the head and the
other to the foot of Canandaigua lake. There were others to
various points. At last they found the right one, but a very bad
road. This took them through old Onnachee in Hopewell, and
Canandaigua lake was then called by this name. The outlet was
crossed on a rude Indian bridge. Ganataqueh (Canandaigua) was
a few miles beyond, on a hill. Thence they went to Hachniage
(Honeoye) near the foot of Honeoye lake. Still going west they
crossed Noehnta creek, the outlet of Hemlock lake, and came to
Ohegechrage or Conesus lake. Ten miles farther they reached
Zonesschio (Geneseo) on the Genesee river, but not the later site.

This is essentially one of Morgan's two trails. One of these went
southwest from Canandaigua to the foot of Honeoye lake, then in
sight of Hemlock lake, passing the foot of Conesus lake, crossing
the Genesee at the present Geneseo, and leading to Little Beard's
town, at one time the largest Seneca village. This had no existence
in 1750 on its later site, and Geneseo was then near the mouth of
Canaseraga creek.

The other trail, considered the main one by Morgan, went from
Canandaigua through the villages of West Bloomfield and Lima, to
Avon on the Genesee, crossing the river a few rods above the bridge,
and following the bank to Ganowauges a mile above. This is satis-
factory for quite recent times, but it leaves out the important early
villages near Honeoye Falls and in Mendon, as well as the great
fort and town near the village of Victor. The great trail certainly
once included these. Guy Johnson's map of 1771 has two of these
trails from Canandaigua, and a third one farther south. Pouchot's
map of 1758 is of more interest than value, as he probably had not
been over the road, but notes on these various charts may be
reserved for an appropriate place.

The remaining section of the main trail, as given by Morgan,
lies west Of the Genesee river, in a region where there were no
Seneca villages in 1650. It led from the river to the great Caledonia
spring, then through the village of Leroy, crossing Black creek
near Stafford and striking Tonawanda creek a mile above Batavia.
Passing through that place, it turned northwest through Caryville,

TRAILS 37

and led to the present Seneca village of Tonawanda. There it
"branched. One path led northwest, through the creek and swamp,
past Royalton and then to the Cold spring 2 miles northeast of
Lockport. Continuing northwest it reached the ridge road and ter-
minated at the Tuscarora village near Lewiston. This latter trail
of course dates from the occupation of that reservation.

The other branch went southwest from the Indian village to
Akron and Clarence Hollow, thence to Williamsville and the head
of Main street, Buffalo, where it ended. Mr Morgan said:

This trail was traced through the overhanging forest for almost
its entire length. In the trail itself there was nothing particularly
remarkable. It was usually from 12 to 18 inches wide, and deeply
worn in the ground; varying in this respect from 3 to 6, and even
12 inches, depending upon the firmness of the soil. The large Jrees
on each side were frequently marked with the hatchet. Morgan,
P429

It remains to notice his lake and river trails. From Oswego one
followed the lake ridge to Irondequoit bay, turning up the bay to
its head, crossing the Genesee at the Rochester aqueduct, striking
the ridge road at the lower falls and going west to the Tuscarora
village, a recent town.

A trail followed the Genesee river on each side, connecting the
recent Seneca villages occupying the valley. These need not be
mentioned here, and for many interesting details the reader is
referred to Morgan's valuable work.

Trails naturally converged at Tioga point, where the Chemung
united with the Susquehanna, and these became important thorough-
fares for the Iroquois in their later wars. From this point he
named two trails up the Susquehanna. One followed the north
bank, crossing the Chenango at Binghamton, thence to the Unadilla,
and there intersecting the Oneida trail. The frail again branched
at Charlotte river, one branch going to the Cherry Valley creek
and then to Canajoharie. The other followed the Charlotte to
its head, crossed to the Cobleskill, intersecting the Schoharie trail .
at Schoharie creek, ending at the lower Mohawk castle. A branch
turned up Foxes creek, crossed the Helderbergs and ended at

3 NEW YORK STATE MUSEUM

Albany. This was the Indian Ladder road. Another crossed the
town of Middleburg to the Catskill, following that stream to the
Hudson.

It is evident that these trails came from the recent occupation of
Schoharie creek by the Mohawks, of the Susquehanna by the
Oneidas and Tuscaroras, and of the Chemung by the Cayugas ; in
part by all. In 1753 there was no road at all along the north bank
of the Susquehanna from Owego to the Chenango river.

It is so obvious that most trails came from the situation of towns,
changing as these did, that it can be assumed that trails connected
all friendly towns of any given period. Their rapid obliteration
may also be inferred from the fact that no one pretends to locate,
by physical features or oral tradition, any trail used 250 years ago.
We know the general course of some yet older, but not because any
one has seen their traces. In still more cases we know where early
towns stood, but have no hint what thoroughfares led to other
places. These certainly existed, but have left no visible signs.
Indeed it is quite probable that the later trails had their prominent
features more from the white man's shoes and the hoofs of cattle
than from the Indian's moccasin. Woodland paths are common
now by every lake and stream ; when and by whom were they made ?
Ask our farmers, hunters and fishermen.

That Indian trails were well defined from Utica westward, soon
after the Revolution, no one will doubt. That some of them afforded
the best lines of travel for pioneers is just as clear. The Indian
was a well trained woodman, and the white man profited by his
skill, but in the nature of things the great results would have been
much the same had the latter been left to himself. The New York
Central Railroad would have gone from Albany to Buffalo had there
never been an Indian trail in the State.

Some of these early routes have interest, and the location of some
on early maps may be mentioned. A few local trails will be also
referred to, but it is evident that no complete account could be given
unless we knew the age and nationality of every town.

The earliest trail which can be traced westward from Albany is
from the itinerary of Arent Van Curler in 1634. His recorded miles

TRAILS 39

are each equal to about 2 English miles, and the latter will be used
here. There were then no white settlements west of Albany, and
the Mohawks were all west of Schoharie creek. The first day,
having traveled 16 miles, he was near the Mohawk river. The
second he went 2 miles, crossed the river and walked 20 miles more.
The third he recrossed the Mohawk, and a mile farther came to
Onekagoncka, the first Mohawk castle. A mile farther Canowarode
was passed, and Senatsycrosy at the end of another. Canagere,
the second castle, was 3 miles beyond, or 44 from Albany. The
third castle, Sohanidisse, was 3 miles farther. Osguage was a mile
beyond, and Cawaoge still another. The fourth castle, Tenotoge,
was 2 miles farther, east of the present Canajoharie, and about 51
miles from Albany. For at least 13 miles he had followed the south
bank of the river. Like later travelers, he now soon left it. Leav-
ing Spraker's he took a westerly course, crossing Canajoharie creek
but passing south of the next large stream, the Otsquago. That
day's tramp of 14 miles ended on high land near the west line of
Montgomery county. The next day 15 miles brought him near
Jordanville. Next day he crossed two branches of the Unadilla,
probably near North Winfield and North Bridgewater, or a little
farther south, camping a little west of the west branch, after walking
15 miles. The next day their course was near Sangerfield or Water-
ville, but at the end of 16 miles they had not quite reached Oriskany
creek. This they saw next morning, and at the end of 9 miles they
were at Oneida, east of Oneida creek and near Munnsville. They
walked through the snow, and the miles seemed long.

The records of the sojourn of Father Jogues, in 1642 and later,
tell us nothing of the main trail, still on the south side of the
Mohawk, but show many minor trails from place to place and for
many purposes, as we might expect.

The famous horseback ride of Wentworth Greenhalgh in 1677,
when he went from Albany almost to the Genesee river, shows a
little variation. There were four Mohawk castles and one small
village, all on the north side of the river. That he followed the
same trail as Van Curler when he left this, may be gathered from

AO NEW YORK STATE MUSEUM

his saying that Oneida was " about 30 miles distant from the
Maquaes river, which lyes to the northward." Onondaga was still
in Pompey, about 2 miles south of Morgan's later trail. The Seneca
towns lay north of that route, and he passed Cayuga lake by some
path available for horses, evidently on the north.

The trail from the last Mohawk castle to Onondaga changed but
little for a great length of time, and only as the two places and
Oneida changed their sites. It was always the practice to leave the
river at the upper Mohawk town, and take the direct overland traiL
This is the route laid down on Colonel Romer's map of 1700, and
when Johannes Bleeker jr and David Schuyler went to Onondaga
in 1701, they said that they got to Eghwake creek, the east branch
of the Unadilla, on the evening of June 7. Oneida had been moved
northeast of its former site, and they reached there next day. Part
of the early trail had been abandoned, but much of it was still used.

At that time there was little land travel along the Mohawk above
Little Falls, yet the portage at Rome was sometimes used. On
Colonel Romer's map a trail leads from Oneida to that spot, and
runs at right angles to the main road, from which it was a short day's
journey. When a trading post was established at Oswego it became
an important thoroughfare. This is what Romer's companions
wrote in October 1700:

loth Col. Romer told us that his instruccons were to see how
much lesse the Carrying place could be made; whereupon we
resolved forthwith to go thither, as we did, with an Indian which
we hired who shew us the way. nth d° We came by a most miser-
able path to the Carrying Place, wch we viewed as farr as the Wood
Creek, when Col. Romer resolved to go to Oneyda. I2th do. In ye
evening we came to Oneyda. O'Callaghan, 4:807

From this and the map it will be seen how far the portage was
from the main trail. On the map the trail goes from Oneida to
Onondaga, then on Butternut creek, and from that town the party
traversed two trails only: one to Onondaga lake and the other to
the fishing place on Chittenango creek.

A few years later Onondaga was moved to the east side of Onon-
daga creek, but this removal had most effect on the branch trails,

TRAILS 41

the main one changing but little, still passing the old town a mile
south of Jamesville, where the pickets long remained.

Guy Johnson's map of 1771 has a trail from Oneida, through
" Ganaghsaraga, a Tuscarora town " to Onondaga, and thence by
way of Owasco lake to the foot of Cayuga, following the north
bank of Seneca river to Seneca lake and Canadasegy. Canadaragey
(Canandaigua) next appears, and there are three trails thence to
Genesee river. The southern goes by Anarara (Honeoye) to
Chenussio; the middle one direct to the same place, and the third
to Canawagus, while another runs northeast from Canawagus to the
head of Irondequoit bay. From Geneseo the trail goes direct to'
Fort Schlosser on Niagara river, and small villages appear along
the way. A trail ascends the east side of Genesee river, and else-
where Kanestio is connected with Ganuskago (Dansville). There
are no trails on the Susquehanna, the river being used instead.

Sauthier's map of 1779, made by order of Governor Tryon, shows
some of the changes made as the frontier extended. Some of the
earlier trails still appear south of the Mohawk, but there are new
starting places on that river. The German Flats afforded two, and
there was another road on the north side. Fort Schuyler (Utica)
had become a starting point for Old Oneida and the towns beyond.
On this trail were Old Oneyda, Canowaroghare, (now Oneida
Castle), Canadasseoa and Canassaraga Castle, two Tuscarora towns.
From the latter one trail went to Three Rivers, and another to
Onondaga, then on Onondaga creek. From Canowaroghare one
went to Fort Stanwix, and another to the Royal Blockhouse by way
of New Oneyda Castle (now Oneida Valley). From the latter
place one reached Wood creek, while another went to Fort Stanwix.
Among others one went due north through the wilderness to Ogdens-
burg, then called Oswegatchie.

Lieutenant Lodge's map, made im the campaign of 1779, carries
the trail south of Conesus lake. Pouchot's map has some special
features, but they are of doubtful value. The Jesuit Relations
contribute little on this subject, though some make it clear that
there was a good trail from Salmon river to Onondaga by way of
Brewerton, and apparently one from the same place to Oneida

42 NEW YORK STATE MUSEUM

passing the other end of Oneida lake. On the Jesuit map of 1665
the Black and Oswegatchie rivers both bear the title, " R. qui vient
du coste d'agne " or the Mohawk country. By the latter route the
Mohawks took Father Poncet back to Canada in 1652. For a con-
siderable part of the way the routes were one. Gen. J. S. Clark
has elaborated the full route very clearly, in a note to Rev. Dr
Charles Hawley's translation of the Relations as they concerned
the Mohawks. He supposed that one trail from the Mohawk fol-
lowed its north bank to Rome, continuing along the line of the
present Rome & Watertown Railroad till it struck Salmon river,
10 or 12 miles from the lake. From the Mohawk another followed
the west bank of West Canada creek and near the line of the Black
River Railroad and Black river to Great Bend. Lake Ontario might
then be reached by following the stream, or by a portage of a few
miles to Indian river the St Lawrence might be entered through the
Oswegatchie. The usual war path of the Mohawks was through
Lake Champlain and the Sorel river. The lake was reached by
several trails.

The trail by which the French usually came to early Onondaga
led from the mouth of Salmon river to Brewerton, and thence it
varied as the village moved. The French at last came by way of
Oswego. Champlain also came by way of Brewerton, but where he
left Lake Ontario has been much disputed. It was far to the north-
east of Oswego, and Salmon river has been thought a probable place.
That he crossed the Chittenango at Bridgeport is likely, and that he
followed a trail is evident from encountering a party going to the
fishing place. The path probably led up the ridge east of the
Chittenango valley, but has left no traces.

The trails leading to the Susquehanna valley became important
nearly 200 years ago, when the Iroquois land claims in Pennsyl-
vania assumed a new aspect. They had been matters for diplomatic
action even in the I7th century. When the Iroquois realized that
there was money in them they sent a resident viceroy to rule their
subjects there and care for their lands. This and their southern
wars led to many journeys. As early as 1737 Conrad Weiser was
sent as an ambassador to Onondaga by way of Owego. In 1743

TRAILS 43

lie went again with quite a party on horseback, and in the party
were John Bartram and Lewis Evans. The latter made a map of
the route, the former wrote the itinerary. In 1745 Bishop Spangen-
"berg came over much the same route, and his party also rode. In
1750 Bishop Cammerhoff and Zeisberger tried a different course,
Doming in canoes as far as Waverly on the Chemung river, and
going thence to Cayuga on horseback. While previous travelers
had gone by way of Owego creek and Cortland county, they fol-
lowed Wynkoop creek, passed Cayuta lake, reached the site of
Ithaca, and went down the east side of Cayuga lake to the Cayuga
towns. All of these journals are of interest, but while some parts
of the route are easily recognized, some are hard to identify, nor
was the path always quite the same, even in going and returning.
Between Bartram's account of the road and that of Spangenberg
there is quite a difference, though they had the same guides and
made the trip but two years apart.

Each may be summarized after leaving Owego. Bartram's gen-
eral course was on the east side of Owego and West creeks, crossing
.a steep hill to a tributary of Fall creek, and passing some ponds in
the town of Cortland. From the site of Homer -he followed the
west branch of the Tioughnioga, seeing Mount Toppin but not the
ponds near by, and crossed to a branch of the Susquehanna rising
in Pompey. Part of Pompey hill was crossed and the Indian vil-
lage in La.Fayette visited. The Onondaga valley was entered from
the southeast. The route was slightly changed on the return, and
a branch trail led to Onaquaga.

In his notes on Spangenberg's journey in 1745, Mr John W.
Jordan made his route up Owego and Catatonk creeks, leaving the
latter above Candor, crossing Ganatowcherage or West creek in
Richford township, passing over Prospect hill in Harford and a
creek in Virgil which is an affluent of Fall creek, and reaching
Crandall's pond in Cortland. It may be that the route was up West
creek instead of Catatonk, as in Bartram's route. From Crandall's
pond or Lake Ganiataragachrachat they went to Big lake or
•Ganneratareske in Preble, and thence to Oserigooch, the largest lake

44 NEW YORK STATE MUSEUM

in Tully. Beyond this the trail was bad, but they went by way
of Cardiff, as later travelers did. Most of the trail to Owego was
so little used that it was hardly discernible even to Indians, who
depended as much on the lay of the land as the actual path.

Zeisberger and Frey attempted this route alone in 1753, starting
properly from Owego, but losing the faint trail so often that they
were discouraged and turned back. Afterward they were told that
if they had gone on a day longer they would " have had a good
road, because two roads meet there, and a road branches off, turning
toward Cayuga lake. It is much frequented." Their observation
on this is of interest:

The Indians had no proper trail, but where they cannot distinguish
it each one runs through the woods according to his own judgment.
Consequently it frequently occurs that from two to three miles, and
farther, there is no visible road. Zeisberger, p. 1753

On this occasion they finally went up the east branch of the
Tioughnioga by canoe as far as they could, leaving it northeast of
Cortland and crossing the hill to Onogariske creek, now called the
west branch of the Tioughnioga. When they left the river the
Indian guides " ran hither and thither into the forest, until at length
they found a path." They reached the west branch between Homer
and Preble, and " the trail that comes up from Owego, and is quite
clearly defined here." At the lake the trail divided, one branch
going to Onondaga and the other to the village of Tueyahdassoor
where other trails diverged.

It is evident that hundreds of trails have left neither trace nor
tradition, though some were once of great importance. Wherever
there were towns or frequent camps there must have been forest
paths. In a score of counties there was a network of these, old
and new, almost as complex as our own roads now. No general
scheme of these is possible, but it may be assumed that all early
towns were connected and most lakes and valleys were accessible
by them. Even distant points were reached by the most practicable
routes. There were war paths, hunting paths and paths of peace.
Very few of these are on record and it will suffice to mention those
briefly.

TRAILS 45

In Albany county the Indian Ladder road is well known, a recent
trail from Albany to the Schoharie valley, crossing the Helderbergs
in Guilderland. Five trails were mentioned in Rensselaerville in
1711, and the Schenectady trail soon became important.

In Broome county the trail to Binghamton, over Oquaga mountain
and another nearer Windsor, were worn deep. These were recent,
there being no early settled occupation of the county.

In Cattaraugus county a trail ran through Carrollton, following
Cold Spring creek and passing into Napoli on lot 41. A trail from
Allegheny river followed the same creek into New Albion. Thence
it went to Niagara Falls and Canada.

In Cayuga county some trails appear on maps relating to Sulli-
van's campaign. Gen. John S. Clark placed Thiohero " at the foot
of Cayuga lake, on the east side, at the' exact point where the bridge
of the Middle Turnpike left the east shore. The trail across the
marsh followed the north bank of an ancient channel of the Seneca
river." The early trails were very many,' and the Moravians
described some.

There are a few trails on record in Chemung county, and some
appear on the Susquehanna in Chenango county. In Columbia
county the stone heaps were by Indian trails, and on the map of
the Livingston patent a trail crosses it midway from east to west.
Reference has been made to trails in Cortland county, followed by
the Moravians.

A wide trail followed the Charlotte river in Delaware county, in
1786. In Franklin county are early and recent portages. A trail
called the Catskill Path led from Castle Heights due north to the
Coxsackie plains in Greene county. Some recent portages alone
represent the many early trails of Jefferson county, but the recent
trails of Madison county are better known. One of these was from
Oneida to Chittenango, and thence to Onondaga, passing some dis-
tance south of Canaseraga. A well defined trail went from Oneida
creek, through the west part of Hamilton, and down the Chenango
river. The Oneidas and Tuscaroras came up that river in canoes
as far as they could, and diverged to their several towns.

46 NEW YORK STATE MUSEUM

Mr George H. Harris has left us an excellent account of the trails
in the lower Genesee valley, and his judicious remarks are quoted
here:

In general appearance these roads did not differ in any particular
from the ordinary woods or meadow path of the present day. They
were narrow and winding, but usually connected the objective
points by as direct a course as natural obstacles would permit. In
the general course of a trail three points were carefully considered —
first, seclusion ; second, directness, and, third, a dry path. The trail
beaten was seldom over 15 inches broad, passing to the right or
left of trees or other obstacles, around swamps and occasionally
over the apex of elevations, though it generally ran a little one side
of the extreme top, especially in exposed situations. . . Fallen
trees and logs were never removed, the trail was either continued
over or took a turn around them. The Indians built no bridges,
small streams were forded or crossed on logs, while rivers and lakes
were ferried on rafts or in canoes. Harris, p.37

To these general rules exceptions will be found, as in the case
of bridges, and sometimes swamps. Mr Harris noted a branch trail
from Canandaigua lake to the head of Irondequoit bay, then to
Genesee falls and along the lake ridge to the Niagara river. Trails
converged above and below Rochester at two points. The trail from
Canandaigua was on the Pittsford road, dividing a little east of
Allen's creek, and going to Brewer's landing. Several branch trails
diverged from it. The other trail reached the river near Franklin
and North St Paul streets.

A trail came to South Rochester from Caledonia springs. Several
others are mentioned in and about the city, two being portage trails.
There were others about Irondequoit bay, but he differs from
Morgan only in added details. Mr O. H. Marshall also described
the trails followed by De Nonville's army in 1687.

Mr Irving W. Coates said that five trails met at Littleville, on
the Canandaigua outlet in Ontario county, and traced their general
course. Others were mentioned in Sullivan's campaign and in the
Moravian journal.

Mr Jeptha R. Simms mentioned several trails in Schoharie county,
with more details than Morgan gave. In Ulster county a great
trail started from Saugerties, followed the Esopus, crossing to the

TRAILS 47

Rondout at Marbletown, leaving that stream at Napanoch, passing
through Mamakating hollow, and reaching the mouth of the Never-
sink river at Port Jervis. In these and other cases most of the trails
are modern.

With these data before us, and admitting a general truth, we can
see that the Cayuga chief Wa-o-wo-wa-no'-onk, or Peter Wilson,
was somewhat rhetorical when he said :

The Empire State, as you love to call it, was once laced by our
trails from Albany to Buffalo, — trails that we had trod for centuries —
trails worn so deep by the feet of the Iroquois that they became
your roads of travel, as your possessions gradually eat into* those of
my people. Your roads still traverse those same lines of communi-
cation which bound one part of the Long House to the other.

ADDENDA

The work of examining collections and sites has thus far been a
personal one, the practical results of which come to the State
Museum. It is a present gratification to secure figures and descrip-
tions for record and preservation, but it is hoped that this will be
of future advantage to the public. Out of these may be selected
many for publication by the writer or others. Since the bulletin on
metallic implements was published several fine articles of native
copper have come to light and been figured, described and recorded,
but they are axes, spears and celts, much like those already figured.
A later history has been secured of the largest native copper imple-
ment found here, and drawings have been made of some recent
metallic pipes. Bone articles are still obtained, Onondaga and
Montgomery counties yielding most harpoons. One unique bone
article belongs to Mr H. A. Pride of Holland Patent. It is flat,
curved, and much like a small double pointed harpoon, with a single
barb toward each end of the concave edge. This edge expands
toward the center, allowing a perforation in the widest part. The
convex edge has notches and one surface is ornamented with lines.
Mr Pride has some fine pipes and one undrilled bannerstone of
striped slate. A fine flat and curved bone article, which he took
from a grave, he gave to the writer. It is sharp at both ends and
has parallel grooves at one.

Mr Fred C. Gabriel has found many peculiarly grooved and
worked elliptic pebbles on the west side of Seneca lake near Watkins
N. Y., the grooves being picked or ground, and usually the long
way. Some are otherwise worked. They could not have been used
as hammers and seem too elaborate for sinkers. Being usually found
in pairs they suggest the bolas, and this or the sling stone might
have been moderately used in open woods. In dense woods they
could not have been thrown, but the Indians often burned the under-
brush. Such stones occur on Cayuga and Seneca lakes, but mostly
at the head of the last. Some graves near the latter lake have been
particularly described. A fine stone bird pipe was also found near
Watkins. The claws, tail and closed wings are well defined, and

ADDENDA 49

it is 3^4 inches high. An unfinished bird amulet was also found
there, and also a cuneiform slate knife, a rarity in that region.

A fine red sandstone plummet, 3^5 inches long, is from Rowland
island, north of Cayuga lake, the farthest west any such article has
been found on Seneca river. A beautiful and slender one is from
Jefferson county. A smooth pipe of red sandstone, found near
Savannah N. Y., weighs i pound 5*4 ounces. A peculiar Onon-
daga clay pipe, with many grotesque human faces, has heretofore
been found only on a site west of Cazenovia. A stem now in the
writer's hands, came from Canoga, on the west shore of Cayuga
lake. It has the character of all the rest, but had been carried there
and used as a bead.

A grave was opened near Athens, Greene co., in 1899 by Dr A. H.
Getty of that place, on the Saunders farm. It was in a sand bed
and paved with cobblestones brought several miles. The remaining
bones and relics were about 4 feet under ground. It contained 300
globular native copper beads, 16 good sized shell beads of unusual
form, 4 longer ones perforated at each end, and a slate gorget with
one hole. All these are now in the State Museum.

In the fall of 1903 Mr E. Hollenbeck found an ojive green tube
of banded slate at Hoffman's Ferry. It is somewhat flattened and
has a groove across one end. Near this is a small lateral perfora-
tion, much like the vent of old guns in appearance and position.
In this respect it is unique. The length is 4j^ inches.

An ossuary was opened Sep. 8, 1904, in low ground, a mile east
of Macedon, Wayne co., and near the creek. Dr C. P. Jennings,
who superintended the excavation, writes : " We found there, in
one pit, at least 60 skeletons within an area of i rod square and
about 3 feet deep." Six large stones were found with these, but
no relics.

Mr J. E. Mattern, of West Rush, Monroe co., writes : " I know
of a few burial sites not mentioned in your bulletin on sites. One
is about 6 feet in diameter and I took about 20 skeletons out of it.
The bones were in all shapes. They were thrown in after the flesh
was gone. Last fall I found another about 10 feet from the one
mentioned, still larger. I took out 30 skeletons and did not dig it

50 NEW YORK STATE MUSEUM

all out. I found a flint knife, 3 shell beads and a bone awl." He
reported 10,000 pieces in his collection.

In Watertown, Jefferson co., Mr R. D. Loveland added nearly
40 pipes in 1904 to his already fine collection, and a large number
of bone articles. In both there were pieces of very great interest.
A barbed bone fishhook is now in his cabinet. Messrs Amidon, Get-
man, Loveland, Woodworth and others could furnish a notable
exhibit for the Jefferson county centennial.

The evolution of the human face on pottery has been finely illus-
trated by a complete series from Jefferson county. First are the
three circles arranged for eyes and mouth ; then three horizontal
ellipses, either in lines or excavations ; these are next inclosed in a
diamond, followed by a pentagon and hexagon ; then lines are incised
for the nostrils and sides of the nose; and last comes the human
face, as yet rare in that county. The absence of this class of pottery
from Oneida sites has long been a matter of surprise to the writer,
but now he has an example from Fish creek in Oneida county,
though none from a village site. This has the slender form of a
man, with the usual conventional body and limbs. A still more
interesting find was that of Mr A. B. Skinner on Staten Island
made last spring. On a camp site he found most of the rim of an
earthen vessel, having "rude raised human faces upon it." They are
small, four in number, and at some distance from the top of the ves-
sel, which had a pointed base, like most others there. It should be
added that i't was from a shell heap, of which Mr Skinner said:
" No horn or bone implements obtained here. Recent relics, such
as a brass arrow point, bullets, gunflints, etc., have been found."
The inference would be that it was either the work of an Iroquois
captive or of an Algonquin who had been in the hands of that people,
at a recent date.

A gouge with ridged back, from Lysander, is of interest, and long
and fine gouges have been found in Chautauqua county. One collec-
tion has several fine stone mortars from Seneca river, and another
has one which is long, elliptic and shallow. A small and deep one
came from Chautauqua county. A curious and pretty one is from
Oneida creek and may have been used for paint. It is small and

ADDENDA 51

almost globular, being 2^4 inches wide by 2 deep, flattened below,
and with an excavation nearly an inch in depth. From the same
place came a fine elliptic grooved granite hammer, 6-H* inches high
by 3^/2 thick. A rougher one was found with this. Mr Skinner
reports a number of grooved axes found on Staten Island, of various
types. One weighed 15 pounds, which makes it one of the largest
size. They are rather frequent on Long Island, and occur also along
the Pennsylvania line.

In the bulletin on wampum it was noted that a wampum belt
attached to an Oneida treaty of 1788 could not readily be found
for illustration. It is now on exhibition in the State Library with
the original record, and is well worth attention from its beauty and
known history and use. The length is about 27 inches, with 6 rows
and 4 diagonal double bars on the white ground. In the Archaeo-
logical Report of the Province of Ontario, 1901, are figures and
accounts of 4 fine Ottawa belts. One has the date of 1764 on it,
and another of 1786. A third has an officer's name, and may be
assigned to 1815. The fourth is credited to the same year as the
first. Among other belts examined is one now in the National
Museum, which is 2 feet long and 8 rows wide. An open square
at each end is joined to the other by a central black line.

Ap. n, 1903, Mr Alanson B. Skinner read a paper before the
Natural Science Association of Staten Island on aboriginal sites
found there, enumerating 24 of these, but afterward adding two
more. The following are those mentioned in his paper, but he said
that there were probably others along the south and east shores
not yet observed.

1 Village and cemetery at Pelton's cove, between Livingston
and West New Brighton. The cemetery was large.

2 Village at West New Brighton, between Cedar and Dongan
streets. Skeletons and implements were found in the spring of 1903.

3 At Mariner's Harbor was a village site between Blue-Bent
field and Western avenue, near the shore. Shell heap opened in
1902 near Arlington station. Grooved axes and other early articles.

4 Village on Tuinessen's or Old Place neck. The earthen
vessel mentioned came from a shell heap there. Recent relics.

52 NEW YORK STATE MUSEUM

5 Relics abundant in sand dunes and hills at Bloomfield.

6 Lodges about Watchogue road, near Union avenue, Chelsea.

7 Lodges along the north side of Long neck, Linoleumville.

8 Lodges on the south side of the neck. No relics in shell pits.

9 Site with graves at New Springville, on Corson's brook.

10 Camp between Journeay avenue and Annadale road, Green
Ridge. Early relics.

11 Small village on Lake's Meadow island. Early relics.

12 Small village on Sandy brook, between Pleasant Plains road
and Journeay avenue. Early relics.

13 Lodges from Cedar hill to Winant's brook, and along the
shore to Rossville. Early relics.

14 Shell mounds and cemetery near Billopp house, Tottenville.
Early and recent relics. Grooved ax of 15 pounds.

15 Site at Bunker hill, Huguenot, near Arbutus lake.

1 6 Shell heap on bluff near Seguine's point.

17 Site on Richmond avenue, near Arrochar station. Probably
others near.

1 8 Camp at Harbor hill, New Brighton, a little above Castleton
avenue.

19 Camps at Silver lake. Relics along the shore near St George.

20 Camp on Harbor hill near Harbor brook and Lafayette
avenue.

21 Camp near Bard avenue and Clove road, above Schoenian's
pond.

22 Large camp back of Richmond, near Ketchum's mill pond.

23 Shell heap on salt meadow, Oakwood, near Lake's mill.

24 Site with many triangular arrows on Ward's hill, near Cebra
avenue, Tompkinsville. There were shells and scattered relics near
most of these places.

EXPLANATIONS OF PLATES
Plate 1

54 NEW YORK STATE MUSEUM

Road map of Perch lake and vicinity, from an atlas of Jefferson
county. In this the lake differs much in outline and extent from
the contour map, but may have been fairly accurate at an earlier
day. All remaining mounds on the east side are between the high-
way-and the lake. In a few cases they are at some distance from
the shore. Quite a space separates this long group from the one
on Linnell's island. It is said that some mounds were formerly
near Seven Bridges, on Perch river.

Plate 1

K. D

X

Plate 2

56 NEW YORK STATE MUSEUM

A sketch map of Linnell's island, by Dr Getman, showing the
general arrangement of mounds observed on the Klock and Gailey
farms in 1903. This terrace is southwest of Perch lake, and was
formerly surrounded by swamps, showing a higher stage of water
in the lake in earlier days. This subsidence has greatly changed
the outline of the lake.

Plate 2

Hill

A$V,

farm c fly

A
A

Plate 3

58 NEW YORK STATE MUSEUM

Contour map of Perch lake and vicinity, showing numerous low
terraces in the thin soil. In many parts these terraced rocks are
nearly or quite bare. On the southeastern shore of the lake the
slope is quite abrupt. On the west shore the extensive swamps,
separating the cliffs from the lake, probably always hindered much
occupation, but the proximity of the river to Linnell's island made
that a desirable residence.

Plate 3

Plate 4

6O NEW YORK STATE MUSEUM

Medium sized mound on the second terrace east of Hyde creek,
and not far from it. The depression in the center of this is charac-
teristic of nearly all near the creek and lake, and excavation showed
the original rectangular fireplace, bordered with flat stones. Like
nearly all others it is in open woodland. The extreme width is
about 30 feet. A plan of this appears on plate 12.

Plate 5

62 NEW YORK STATE MUSEUM

A larger mound near the north line of the Timmerman farm.
This is about 36 feet across from the extreme points of the slope,
which is always gentlest near the edge and more abrupt as it
approaches the center. This mound is less depressed in the center
than most of those on that side of the lake.

OS
P4

Plate 6

64 NEW YORK STATE MUSEUM

A low and small mound on the Timmerman farm, shows the usual
formation, but at noonday might be passed unobserved. Early or
late in the day its character is clear. Being in the incipient stage
it is but 19 feet across.

m m *

Plate 7

66 NEW YORK STATE MUSEUM

A low mound, on the same farm as that shown on plate 6, which
is but 21 feet across. Though not large it is very symmetric. The
central depression is wide and deep.

ft

Plate 8

68 NEW YORK STATE MUSEUM

A fine mound on the Timmerman farm, at the base of a hill.
Partly excavated in 1901. This is 33 feet wide and about 5 feet
high. There are but few trees near this and it is a prominent
object.

GO

Plate 9

7<D NEW YORK STATE MUSEUM

Mound near the ruined La Farge mansion. Another joins it at
the base. It is not one of the largest size, but is in open ground on
a lower terrace than the house and stands out prominently against
the background of the lake. This mound has an extreme width of
34 feet.

+.

od

3

Plate 10

NEW YORK STATE MUSEUM

Figure I is a sketch, furnished by Harlan I. Smith, being a sec-
tion of an earth hut of the Thompson River Indians, showing how
mounds of this kind are sometimes formed.

Figure 2 is a plan of a mound on the Bay of Quinte, on the north
shore of Lake Ontario, and showing a central excavation.

Fig. 3 Section of a mound at the same place, showing the interior
filled with stones, the covering of soil and the central depression.

Plate 10

10

Plate 11

74 NEW YORK STATE MUSEUM

Truncated mound excavated by Dr A. L. Benedict in 1900, on
the east side of Cattaraugus creek, N. Y. The upper figure is of
the recent condition and probable original form. The plan shows
the position of various points of interest reached in excavating.
At A, were animal bones ; at B, pieces of Hamilton slate ; at C,
human bones. Mound diameter, 70 feet.

Fig. 2 A mound near the same creek and much like the last.
At A and H, gravel was found ; at F, a stone fireplace ; at B, were
human bones. There was also charred wood.

Plate 11

JT

Plate 12

76 NEW YORK STATE MUSEUM

Supposed deer pound by a pond near Unadilla. A, is the pond;
C, moraines; D, a road around the pond; E, F, G, I, stone walls;
H, stone circles ; K, a graded way.

Fig. 2 Plan of a Perch lake mound. A, the outer slope ; B, the
crown of the ridge; C, the inner slope; D, rectangular stone fire-
place in the center. This is the ground plan of mound shown on
plate 4.

Plate 12

^ '"^

gfe «£

s» wf*j*tw *4n»ff|<ff» +f*t 1 1 ( <i <»> ( > ' i »*» > <*>

A*-JL

» »
»* **

INDEX

The superior figures tell the exact place on the page in ninths; e. g. 37*
means page 37 beginning in the third ninth of the page, i. e. about one third
of the way down.

Akron, 37"

Albany, 33'"', &, 3&, 39\ 451
Albany county, trails, 45'
Allegany county, mounds, 252
Allegheny river, 25*, 45*
Allen's creek, 46*

Amidon, R. W., cited, 3'; men-
tioned, 28", 50*
Amulet, 28*
Anarara, 41"

Arrowheads, 13°, 19*, 283
Athens, 49*
-Auburn, 35'
Avon, 36'
Axes, 3oa, 48'', 5i2

Baldwinsville, 3O5

Bartram, John, mentioned, 431;

cited, 43*
Batavia, 36°
Beads, 30'", 495

Beauchamp, W. M., cited, 35, 5°
Benedict, A. L., cited, 3", 265, 2.f
Big lake, 43°
Binghamton, 37s, 45'
Bird amulet, 49*
Bird pipe, 48°
Black creek, 36°
Black river, 5s, 27°, 42", 42*
Bleeker, Johannes, jr, mentioned,

403

Bluff point, 3i5
Bone articles, 6s, 48'
Bone awl, II1
Bone bead, 28s
Bone fort, 27*
Bone hill, 3O8
Bone implements, 28'

Bone needles, 19*

Bones, 6°, f, S\ 87, io2; Algonquin

superstition, 22°
Boyle, David, cited, 3°
Brass kettles, 30"
Brass rings, 30'
Brewer's landing, 46°
Brewerton, 41", 42°
Bridgeport, 42T
Broome county, trails, 45^
Bryant, William C, mentioned, 26*
Bucktooth, 253
Buffalo, 37'; Indian fort, mound

near, 27"

Burnt stones, 5*, 5°, 5', 8s
Butternut creek, 40°

Caledonia spring, 36°, 46*

California, Indian rancherias at, 7"

Cambria, 2Q5

Camillus, 35-

Cammerhoff, Bishop, cited, 3"; men-'

tioned, 43'
Canada, 453
Canadaragey, 41"
Canadasegy, 4i2
Canadasseoa, 41°
Canagere, 39"
Canajoharie, 341, 37", 39*
Canajoharie creek, 39*
Canandaigua, 35°, 36% 36*, 36", 36',

4ia, 46°

Canandaigua lake, 361, 362, 46^
Canaseraga, 347, 45'
Canaseraga creek, 36*
Canassaraga Castle, 41°
Canastota, 347

NEW YORK STATE MUSEUM

Canawagus, 4i3

Candor, 43*

Canoga, 49'

Canowarode, 39'

Canowaroghare, 41"

Cardiff, 44"

Carlton, 30*

Carrollton, 453

Caryville, 27*, 36°

Cassadaga lake, 256

Castle Heights, 45"

Catatonk creek, 43"

Catskill creek, 38'

Catskill Path, 45'

Cattaraugus county, mounds, 25'-';

trails, 45"

Cattaraugus creek, 26", 26"'
Cawaoge, 39"
Cayuga, 41", 43'"
Cayuga county, mounds, 25*; trails,

454

Cayuga lake, 35*, 35", 40", 43', 44", 45*,
48", 49', 49"

Cayuga towns, 43^

Cayuta lake, 43*

Cecil, Harry B., cited, 3', 32"; thanks
to, 31"

Celts, i35, 48"

Champlain, mentioned, 42°

Charcoal, 5"

Charlotte river, 37", 45T

Charred corn, 54

Chautauqua county, mounds, 25°;
gouges, 50s

Chautauqua lake, 25T

Chemung county, trails, 45"

Chemung river, 38", 43"

Chenango county, mounds, 26";
trails, 45"

Chenango river, 37s, 38", 45°

Chenussio, 41"

Cherry Valley creek, 37"

Chittenango, 34', 45s

Chittenango creek, 40°, 427

Clarence Hollow, 373

Clark, John S., cited, 3", 34°; men-
tioned, 42'", 45'

Clay pipe, 19% 49s

Clifton Springs, 3OT

Coates, Irving W., cited, 46"

Cobleskill creek, 37''
• Cold Spring, 25\ 37l

Cold Spring creek, 25*, 25", 45"'

Columbia county, mounds, 26s;
trails, 45"

Conesus lake, 36", 36*, 41"

Conewango, 25% 25"
! Cooperstown, 30''

Copley, Enoch H., mound on farm,

Copper implements, 48**
Cortland, 43", 43", 44'
Cortland county, trails, 43:i, 45'''
CoxFackie plains, 45s
Crandall's pond, 43'''
Crusoe-creek, 31"

Dansville, 28", 41*
Danube, 34'
Dayton, 25''
! Deep Spring, 34'
Delaware county, trails, 457
Dillenbeck, A. J., mounds on farm,
134

Eghwake creek, 40*
Elbridge, 35-
Ellington, 25T

Emerson, Edgar E., cited, 3', 8a
Erie county, mounds, 26s, 27"
Esopus creek, 46"
Evans, Eewis, mentioned, 43l
i Explanations of plates, 53-76

Fall creek, 43", 43°

Firestones, 8"

Fish creek, 50*

Flint chips, 5"

Flint implements, 87

Flint knife, 13*

Flu van na, 25"

Fonda, 34'

Fort Plain, 34l

Fort Schlosser, 41*
| Fort Schuyler, 41"

Fort Stanwix, 4i7

Foxes creek, 37"
, Franklin county, mound, 277;

trails, 457
I Fredonia, 25s

INDEX TO PERCH LAKE MOUNDS

79

French, J. H., cited, 37, 5"
Frey, mentioned, 44'

Gabriel, Fred C, mentioned, 48*

Gailey, Clarence, mentioned, 18", i8u

Gailey, John, mounds on farm of,
S3, 5s, 7!>-81

Ganaghsaraga, 41"

Ganataqueh, 36"

Ganatowcherage, 43s

Ganiatarage, 35'

Ganneratareske, 43'''

Ganowauges, 36"

Gansevoprt, Colonel, route, 35"

Ganuskago, 41'

Gasport, 29"

Genesee county, mounds, 27s

Genesee falls, 46'

Genesee river, 2g\ 36*, 36", 36", 37",
37", 41", 4i "

Genesee valley, trails, 46l

Geneseo, 36", 36", 36"', 4T:!

Geneva, 3OT, 35''

German Flats, 41"

Getman, A. A., cited, 3", 8s, 18';
mentioned, 50''

Getman, S., mentioned, 13"'; mounds
on farm of, I3T

Getty, A. H., mentioned, 4Q4

Gistweahna, 35*

Gorget, 49"'

Gouge, 50"

Great Bend, 42^

Great Gully brook, 35°

Green lake, 34"

Greene, 26^

Greene county, trails, 45s; bones
and relics, 49' ,

Greenhalgh, Wentworth, men-
tioned, 39"

Groveland, 28"

Gnilderland, 45*

Hachniage, 36"
Hamilton, 45"
Hammer, 5il
Harford, 43s
Harpoons, 48*

Harris, George H., cited, 3*, 46';
mentioned, 29*

Havvley, Charles, mentioned, 42'
Helderbergs, 37'', 45l
Hemlock lake, 287, 36% 36*
j Herkimer county, trails, 34 l
Hoffman's Ferry, 49"
Hollenbeck, F., mentioned, 49"
Homer, 43", 44"
1 loneoye, 363, 4i3
Honeoye Falls, 367
Honeo}^e lake, 36*
Hopewell, 36"
Horn implements, 28'
Hough, F. B., cited, 3", 5-, 87
Howland island, 49"
Hudson river, 38'
Hut rings, 19''', 27"
Hyde creek, 53, 8'\ 12s, 13-

Indian brook, 26"

Indian Ladder road, 38l, 45l

Indian river, 42*

Irondequoit bay, 29', 37°, 4i3, 46",

467
Ithaca, 43:t

Jamestown, 257

Jamesville, 34", 4I1

Jefferson county, mounds, 5\ 27°;
trails, 45^; plummet from, 49";
bones and relics, 50'

Jennings, C. P., cited, 3"; men-
tioned, 49T

Jogues, Father, cited, 397

Johnson, Guy, cited, 34"; map of
/77/, 367, 41 l

Johnstown, 34"

Jordan, John W., cited, -3"; men-
tioned, 43s

Jordanville, 39'"

Kanatagowa, 35l

Kanestio, 41*

Klock, A., mounds on farm, 55, 5s

La Farge mansion, 89, nr>, 16*, 17"
La Fayette, 43T
Lake Champlain, 42"
Lake Ganiataragachrachat, 43'''
Lake Ontario, mounds on north
shore, 9"

8o

NEW YORK STATE MUSEUM

Leon, 25°

Leroy, 36°

Lewis county, mounds, 2.f

Lewiston, 29", 37-

Lima, 28*, 36°

Linnell's island, 5s, i82

Little Beard's town, 36*

Littleville, 46*

Livingston county, mounds, 28°

Lockport, 29", 37"

Lodge, Lieutenant, map, 41"

Loveland, R. D., mentioned, IQI,

5Q1, 502
Loveland, Mrs R. D., mentioned,

28l

Lowell creek, 53
Lysander, 5o8

Macedon, 49T

Mackay, John, cited, 39; mentioned,

29°, 304
Madison county, mounds, 289;

trails, 45s

Mamakating hollow, 471
Manlius, 34T
Marbletown, 471
Marcellus village, 35*
Marshall, O. H., cited, 46'
Marvin, D. S., cited, 3', 64-76
Massassaga point, 9*
Mattern, J. E., cited, 39; quoted, 49*-

50l

Mayhew, G. F., cited, 4', I93
Mendon, 367

Metallic implements, 48*
Middleburg, 38*
Mohawk castles, 39"; upper, 341;

lower, 379; first, 39'
Mohawk river, 392, 423
Monroe county, mounds, 29l
Moravian journals, 34*
Morgan, L. H., cited, 4\ 33\ 37s
Mt Morris, 287
Mt Toppin, 43'
Munnsville, 397

Napanoch, 471
Napoli, 25*, 453
Neversink river, 471

New Albion, 453

New York city, burial mounds,

30*

Newark Valley, 3I1
Niagara county, mounds, 29"
Niagara Falls, 453
Niagara river, 41*, 46"
Noehnta creek, 36'
North Bridgewater, 39°
North Winfield, 39°
Nuquiage, 357

Oak Orchard creek, 30"

O'Callaghan, E. B., cited, 4-, 40'

Ogdensburg, 4is

Ohegechrage, 363

Clean, 25'

Onaquaga, 437

Oneida, 34°, 39", 40', 4O3, 40", 41',
41°, 4i9, 45s

Oneida Castle, 345, 347, 41°

Oneida county, mounds, 3O5; pot-
tery, so5

Oneida creek, 39T, 459, so9

Oneida lake, 28°, 42l

Oneida Valley, 4i7

Oneidas, 45°

Onekagoncka, 392

Oneonta, 30°

Onnachee, 362

Onogariske creek, 44'

Onondaga, 347, 35*, 4O1, 40% 4O9, 41%
4 17, 4 19, 425, 42°, 448, 45s

Onondaga county, mounds, 30"

Onondaga creek, 40", 4iT

Onondaga lake, 40°

Onondaga Valley, 348, 352, 437

Ontario county, trails, 46*

Oquaga mountain, 4$'

Oriskany, 342

Oriskany creek, 39®

Orleans county, mounds, 30*

Oserigooch, 43"

Osguage, 39s

Ossuary, 27', 29°, 49'

Oswegatchie, 41*, 425

Oswegatchie river, 42l

Oswego, 375, 428

Oswego Falls, 30*

Otsquago, 395

INDEX TO PERCH LAKE MOUNDS

8l

Owasco creek, 352

Owasco lake, 35'"', 41"

Owego, 3I1, 383, 42", 435, 441, 44% 446

Owego creek, 43', 43', 43*

Pebbles, 487

Penfield, 29*

Pittsford, 29*

Pittsford road, 46°

Plates, explanations of, 53-76

Plummet, 4Q1

Pomeroy, Oren, cited, if, i89

Pompey, 4O1, 43T

Poncet, Father, mentioned, 42'

Port Jervis, 47*

Portage, 3i5

Pottery, 5*, 5-, 63, 6°, 8T, 13*, IS7, if,

ig\ i93, 2i\ 2is, so8
Pouchot's map, 367, 41"
Preble, 439, 446
Pride, H. A., mentioned, 48'
Prospect Hill, 43"
Putnam, F. W., cited, 42, 2i7

Quinte, Bay of, mounds on shores
of, 9s

Randolph, 253

Relations des Jesmtes, cited, 43, 228-

234, 4i"

Rensselaerville, 451
Rings, 302
Rochester, 46°
Rome, 405, 423
Corner, Colonel, cited, 34°; map, 4O3,

405

Rondont creek, 471
Royal Blockhouse, 4i7
Royalton, 371
Rutledge, 25s

St Lawrence river, burial mounds,

27s

St Regis island, 277
Salmon river, 41', 42*, 42'
Sandstone, 492
Sangerfield, 39°
Saugerties, 46'
Sauthiers map of 1779, 41*

Savannah, 31% 3i3, 49"
Sayles farm, mounds, i59-i61
Schenectady trail, 335, 45''
Schoharie county, mounds, 26";

trails, 46"

Schoharie creek, 37°, 38"
Schoharie valley, 451
Schuyler, David, mentioned, 4O3
Scottsburg, 287
Scottsville, 29'
Senatsycrosy, 393
Seneca castle, 35"
Seneca Falls, 35"
Seneca lake, 35", 35°, 41", 48', 48"
Seneca river, 358, 41'-, 45', 50°
Seneca towns, 4O1
Sennett, 35"
Shell ornaments, 3O3
Sherman, George W., mounds on

farm, 8°

Simms, Jeptha R., cited, 46°
Skaneateles lake, 355
Skeletons, io8, 25', 25', 28', 29"
Skinner, Alanson B., cited, 43, 51"-

52; mentioned, 50°, sr
Slate knife, 491

Smith, Harlan I., cited, 43, I98
Sohanidisse, 39s
Sorel river, 425
South Rochester, 467
Spangenberg, Bishop, cited, 431, 43*,

438

Spears, 483
Spraker's, 39*
Spring lake, 25s
Squier, cited, 52
Stafford, 36"
Staten Island, 5i2; aboriginal sites,

5iG-52

Stone implements, 87
Stone pipes, 81
Susquehanna river, 37", 382, 41*, 43',

456

Susquehanna valley, 42*
Sword blades, 3o2

Tehondaloga, 34'
Tenotoge, 39'
Thiohero, 45*

82

NEW YORK STATE MUSEUM

Thomas, Cyrus, cited, 4, /"'. 2O+-2i\

22l

Three Rivers, 41'

Timmerman farm, mounds on, 13°,

I44-I58

Tioga county, mounds, 3I1
Tioga point, 37V
Tioughnioga, 43", 44*
Tonawanda, 371
Tonawanda creek, 27s, 36°
Tonawanda island, 29"
Trails, 33-52

Trent river, mounds on, 9"'
Tueyahdassoo, 44°
Tully, 44l

Tuscarora village, 37% 37°
Tuscaroras, 45°
Twining, J. S., cited, 4*, 5?, 6'

Ulster county, trails, 46°
Unadilla, 30°, 31°
Unadilla river, 37",- 39"'
Utica, 34\ 41°

Van Curler, Arent, mentioned, 34*;

cited, 38'
Van De Walker, Alonzo, mounds

on farm, 89, 13s
Verona, 34'"
Victor, 367
Virgil, 43°

Wallbridge, Thomas C., cited, 4*.
9s, gi', io3, if'

Wampum belt, 5i3

Waowowanoonk, quoted, 47'

Watertown, 5O1

\\"aterville, 39"

\\"atkins, 48'

Waverly, 43"

Wayne county, mounds, 31*; ossu-
ary, 497

Weiser, Conrad, mentioned, 42''

West Bloomfield, 36"

West Canada creek, 42*

West creek, 43"', 43"

Whitesboro, 34"

Williamsville, 373

Wilson, James Grant, cited, 4*

Wilson, Peter, quoted, 473

Wilson, 29"

Windsor, 45"

Wood creek, 41*

Woodworth, Henry, cited, 4", 6",
7'-83, so3

Wynkoop creek, 433

Wyoming county, mounds, 31*

Yates county, mounds, 3i3

Zeisberger, David, cited, 4", 444;

mentioned, 432
Zonesschio, 363

* 05

APR 2 6

R

f

-n

REC'D L.D

MAP, 21

LD21-lOOm

.1/54(1887816)476