AUTHOR'S EDITION. .y^'f^'///^//uY.:^

«

DEPAETMENT OF THE I:NTEKI0R.

UNITED STATES GEOLOGICAL AND GEOGRAPHICAL SURVEY.

F. V .• HAYDEN, U. S. Geologist-in-Charge.

FOSSIL FORESTS

OF THE

VOLCANIC TERTIARY FORMATIONS

OF THK

YELLOWSTONE NATIONAL PARK.

BY

A\^. H. holm:es.

EXTRACTED FROM THE BULLETIN OF THE SURVEY. Vol. V, No. 1.

Washington, February 28, 1879.

Art. VII.— Fossil Forests of the Volcaiiac Tertiary For-
ma t^OBis ©r the Yellowstoiae ]¥atioaia] Park.

By \^. II. HolBBies.

The prevalence of extraordinary volcanic activity tlirongliout that
part of the Tertiary age represented by the post-Cretaceous rocks of
the Yellowstone region has given to them a most unique and interest-
ing character. So destitute of animal remains are they, and so uidike
the formations of the age in other j)arts of the Rocky Mountain region,
that, notwithstanding the frequent visits of geologists, no divisions
into sub-groui)s have been made, and no more definite aijpellation for
the whole group has been found than the ^' Volcanic Tertiary^''] this
name, although so general, is singularly appropriate, and, in the absence
of specific determinations, may be used to designate the entire gTOup of
Tertiary strata in the Park district.

It is not my intention in this brief notice to attemi)t the classification
or correlation of these strata, but to give a brief account of some very
remarkable features brought to light by last year's explorations.

In the valley of tlie East Fork of the Yellowstone River, where this
peculiar group of rocks is typically developed, they have a thickness of
uj)wards of 5,000 feet. The prevailing materials which enter into their
composition are fragmentary volcanic products, which have been ai^par-
ently redistributed by water, and now form breccias, conglomerates, and
sandstones. It has been noticed by nearly all visitors that these strata
contain a great abundance of silicitied wood, and in a few cases trunks
of trees in situ have been reported. The lowest observed occurrence of
the strata of this group is in the valley of the main Yellowstone, between
the first and second canons, at an elevation of about 5,000 feet above the
sea. They are also finely developed in the Gallatin Range to the west of
this valley, and about the sources of Canon and Boulder Creeks reach
a thickness of between 3,000 and 4,000 feet. At a number of points
covering this entire thickness, masses of silicified wood occur, and near
the divide at the head of Boulder Creek silicified trunks, many feet in
height, and of gigantic proportions, stand in the identical strata in
which they grew, the crumbling conglomerates having wthered away,
leaving them standing upright along the steep slopes of the mountain.
In general, these strata are horizontal. The bedding is often heavy, and
in places not Avell marked; sub-aerial volcanic deposits apparently alter-
nate to some extent with the sub-aqueous.

Three miles above the mouth of Gardiner's River, in similar strata, there

125

126 BULLETIN UNITED STATES GEOLOGICAL SURVEY. [VolY.

are a number of silicilied trunks in sitUj most of which occur in a stra-
tum of sandstone that lies directly upon the eroded surface of a series
of mica-scliists that belong to the metamori)hic range north of the Yel-
lowstone Eiver. This is at an elevation of G,000 feet.

On the south side of the third caiion, opposite the mouth of Hell Eoar-
ing Creek, is a massive i^romontory comj^osed of conglomerates, in which
are very numerous intercalated beds of sandstones and shales. In the
steeper faces of this promontory, many fine trunks are exposed. In 1872,
Br. Peale obtained some very perfect specimens of fossil leaves from
these beds, on the Elk Creek side, Avhich were determined by Professor
Lesquereux to be of Eocene types. The walls of the caiion in the vicin-
ity of Hell Roaring Creek are formed of the same series of rocks, which
occur also at a number of points on the river between Elk Creek and the
lower falls.

In the valley of the East Fork, the sedimentary formations of the Vol-
canic Tertiary reach their maximum developement. Here they rest u])on
the unevenly eroded surfaces of the palaeozoic and granite rocks, and form
a great part of the mountain ranges that enclose the valley. They are
horizontal and apparently conformable throughout the entire thickness
of 5,500 feet. The greater part of this immense group of strata is filled
with the silicified remains of a multitude of forests.

The section given in the accomi)anying plate occurs in the north face of
Amethyst Mountain, opposite the valley of Soda Butte Creek, and in-
cludes upwards of 2,000 feet of strata. The bed of the river is at an eleva-
tion of 0,700 feet above the sea, and the summit of Amethyst Mountain,
0,400. On the north side of the valley, near the mouth of Soda Butte
Creek, there are between 300 and 400 feet of Carboniferous strata exposed
along the base of the mountain sloi)e. On the south side, occasional
ledges of limestone ai)pear above the detrital deposits. Thin sheets of
basalt cover the flat part of the valley, which is here less than a mile in
width.

The norfh face of Amethyst Mountain does not present as abrupt a
profile as that given in the section, the middle part only being so pre-
cipitous. At the base and toj) there are comparativelj^ gentle slopes j
nevertheless, the actual stratigraphical conditions are truthfully repre-
sented.

As we ride up the trail that meanders the smooth river-bottom, we
have but to turn our attention to the clifl's on the right hand to discover
a multitude of the bleached trunks of the ancient forests. In the steei:)er
middle portion of the mountain face, rows of ujjriglit trunks stand out
on the ledges like the columns of a ruined temple. On the more gentle
slopes farther down, but where it is still too steep to sui^port vegetation,
save a few pines, the petrified trunks fairly cover the surface, and were
at first supposed by us to be the shattered remains of a recent forest.

In. ascending one of the steep spurs that project from the main wall,
the strata were found to consist, toward the base, of shales and fine-

No. 1.] HOLMES ON FOSSIL FORESTS OF YELLOWSTONE PARK. 127

""^oo Vo <^ o/Vo o Vo o'^

Fig. 1.— Section in the north face of Amethyst Mountain.

128 BULLETIN UNITED STATES GEOLOGICAL SURVEY. [YolY.

grained sandstones. Higher up conglomerates occur, and still higher
coarse conglomerates and breccias prevail. Interbedded with the mas-
sive, irregular beds of the latter rocks are always thin layers of sand-
stones and shales. The sandstones are fine-grained, thinly bedded, and
contain more or less tufaceous material. Their prevailing color is green-
ish and greenish-gray. They are apparently composed chiefly of materials
derived directly or indirectly from volcanic sources. In no case are
pebbles of quartz or other granitic constituents found in either the sand-
stones or conglomerates.

The exposures of strata in the first three or four hundred feet at the
base are not good, and but few of the silicified trunks appear above the
covering of vegetation. At the height of 500 feet, the occurrences be-
come very numerous, and the great size and fine state of preservation
of many of the trunks was a matter of much surprise. Prostrate trunks,
50 and 60 feet in length, are of frequent occurrence, and not a few of
these are as much as 5 or 6 feet in diameter.

The standing trunks are generally rather short, the degradation of the
compact enclosing strata being so slow that the brittle trunks break
down almost as fast as they are exi)osed. In many cases the roots are
exposed, and may be seen penetrating the now solid rock with all the
original ramifications One upright trunk, of gigantic proportions, rises
from the enclosing strata to the height of twelve feet. By careful meas-
urement it was found to be ten feet in diameter, and as there is nothing
to indicate to what i)art of the tree the exposed section belongs, the
roots may be far below the surface, and we are free to imagine that
there is buried here a worthy predecessor of the giant Sequoias of Cali-
fornia. Although the trunk is hollow, and partly broken down on one
side, the woody structure is perfectly preserved, the grain is straight,
and the circles of growth distinctly marked. The bark, which still re-
mains on the firmer parts, is four inches thick, and retains perfectly the
original deeply lined outer surface. Specimens of the wood and bark
were collected, but no microscopic examinations have been made. It is
clear, however, that the tree was not a conifer. The strata which en-
close this trunlt are chiefly fine-grained greenish sandstones, indurated
clays, and moderately coarse conglomerates. They have been built
around it as it stood in comi)aratively shallow, but doubtless quiet waters.
As would naturally be expected, these strata contain many vegetable
remains : branches, rootlets, fruits, and leaves are extensively enclosed.
One stratum of sandstone that occupies a horizon nearly on a level Avith
the present top of the giant tree contains a great variety of the most
l^erfectly preserved leaves. Such specimens as we were able to bring
away with us have been submitted to Professor Leo Lesquereux for
identification. They are found by him to belong to the Lower Pliocene
or Upper Miocene, and correspond in a number of their species with the
Chalk Bluffs specimens of Professor Whitney. They include —

Aralia Whitneyi,
Magjiolia lanceolata^ ,

2^0.1.] HOLMES ON FOSSIL FORESTS OF YELLOWSTONE PARK. 129

Laurus ccmariensiSj

Tilia (new sp.),

Fraxinus (new sp.),

Diospyros (new sp.),

Cornus (new sp.),

Pteris (new sp.),

Alnus (new sp.), and a

Fern (new sp.).
It will be observed that most of these species are new, which was also
the case with the collection of Professor Whitney. It is also worthy of
remark that none of the genera are identical with those of the Elk
Creek locality previously mentioned. The stratigraphical relations of
the two localities cannot easily be made out, as they are sei)arated by
fifteen miles of broken country in which the strata are obscured by igne-
ous flows and Quaternary drift. The Elk Creek strata are lower by fully
one thousand feet.

As far above the leaf-bearing horizon as I Avas able to ascend, the sili-
cified trunks were very numerous and well preserved, and, by the aid of
a field-glass, others could be detected in all parts of the cliff to the
highest stratum.

At another point, nearly a mile farther east, I climbed the rugged
walls of the mountain for the purpose of examining a number of large
trees that Avere visible from below. Trunks and fragments of trunks
were found in great numbers and in all conceivable positions. In most
cases the woody structure is well preserved, the trunks have a tendency
to break in sections, and on the exposed ends the lines of growth, from
center to circumference, can be counted with ease. In many cases the
wood is quite comi)letely opalized or agatized, and such cavities as ex-
isted in the decayed trunks are filled with beautiful crystals of qnartz
and calcite. Our party was so fortunate as to procure some very hand-
some specimens of amethyst and ferruginous quartz. It is a matter
worthy of observation that nearly all of the beautiful crystals that occur
so i)lentifully in this region have been formed in the hollows of silicified
trees. The same fact has been noticed in regard to similar crystals in
many parts of the West, and notably in the case of the smoky quartz of
the Pike's Peak region in Colorado.

The silicifying agents have been so unusually active in the strata of
the Volcanic Tertiary that not only are all organic remains thoroughly
silicified, but all cavities in the loosely bedded rocks and all fracture-
lines in the strata are filled Avitli chalcedony or other forms of quartz.

On reaching the heavily bedded conglomerates of the upper third of
the cliff, I found the trees still more perfectly preserved. Many of the
trunks are twenty and thirty feet in height. Their roots are in mosfc
cases imbedded in the layers of finer-grained materials, in which they
grew, while the battered and branchless trunks are encased in the coarse
conglomerates and breccias. These latter rocks are comi)osed chiefly of
Bull. V, 1 9

130 BULLETIN UNITED STATES GEOLOGICAL SURVEY. \Yol.Y.

basaltic fragments, iiiany of which are of great size ; there is, however,
always enough tufaceous and other fine-grained material to fill in the
interstices and act as a cement. These beds are massive and irregular,
and seem to have accumulated too fast to be thoroughly redistributed
by the waters. Only the stronger trees of the forest seem to have with-
stood the fierce storms of rocks that must have prevailed at the period of
their entombment, as the smaller trunks and branches are prostrate or
totally destroyed. In most cases where ui)right trunks penetrate tlie
entire thickness of an enclosing bed, the tops may be seen to terminate
with the uiDper surface of that bed, as if causes had acted at the begin-
ning of the deposition of the succeeding stratum to plane down the
irregularities of the old surface. In due course of time, this succeeding
stratum produced its growth of forest, which followed its many prede-
cessors into the subterranean depths, and in its turn was buried by the
rapidly accumulating conglomerates. This remarkable alternation of
events seems, in a general way, to have been kept up from the beginning
to the end of the period.

The very precipitous character of the cliffs prevented me from reach-
ing the upper part of the wall at this point, but I succeeded in making
my way to the summit of the mountain at two other points, and found
that everyv/here the section was i)ractically the same.

On the opposite side of the valley the same conditions were observed:
the fossil trees occur at the liighest point reached, 3,000 feet above the
river. The ranges that form the rim of this valley on the north and east
reach an elevation of 11,500 feet, and as the conglomerates may be seen
reaching and forming the loftiest summits without perceptible break or
change of character, it is probable that they will be found to enclose the
remains of forests throughout.

On some of the higher summits to the east of Yellowstone Lake, simi-
lar stratified conglomerates contain silicified wood in a very fragmentary
state. These conglomerates are composed mainly of basaltic and tra-
chytic materials, but contain large quantities of fragments of sand-
stones and quartzites, which leads to the conclusion that jiortions of the
earlier Tertiary strata have been broken ui> and ejected with the igneous
l)roducts. It is quite probable that these strata were among the later
products of the Volcanic Tertiary age proper. They are generally found
abutting against masses of unstratified igneous materials that probably
mark the sites of islands Avliich were doubtless volcanic centers. I find
that as we recede from these centers of eruption the strata diminish very
perceptibly in thickness and coarseness of materials, and have at the
same time a very i)erceptible dip toward the surrounding valleys. One
is at times led to susi)ect that portions, at least, of tliese beds are of sub-
aerial formation, as is the case with extensive strata about the cones of
modern volcanoes, but tliere are a multitude of facts that go to i^rove
that the greater part of the formations of this age were rearranged or
sedimented in water.

No. 1.] HOLMES ON FOSSIL FORESTS OF YELLOWSTONE PARK. 131

As to the character of the seas or lakes in which the Volcanic Tertiary
beds were laid down, it is clear that their waters were fresh, but as to
their extent or distribution little is known. The formations cover or
have covered an area of not less than 10,000 square miles, but they lie
at a much greater elevation above the sea than the formations of syn-
chronous lakes of neighboring proviDces, and, so far as is known, have
no actual connection with them.

It has been suggested by some one that these coarse volcanic strata
may have been formed in very restricted bodies of water held high
amongst the mountain ranges 5 but lakes cannot exist without barriers,
and as has already been shown, the conglomerates, although naturally
disintegrating more rapidly than any of the older rocks, now form the
summits of many of the highest peaks that face the eastern plains, and
the basins in which they were formed must have had free communication
with the lowlands to the west, from the beginning to the end of the
period.

For my present purpose it is sufficient to know that the bodies of
water of this period were of sufficient extent not to be greatly affected
in level by the filling-in of volcanic products or by the oscillations of
the district under discussion, since we can have no correct measure of
those oscillations of the surface which define the thickness and decide
the character of strata without the barometer-like records of a sea-level.

The change of level produced by the great oscillation that preceded
the Volcanic Tertiary period, and brought the lofty ranges of this region
into existence, cannot fall far short of 20,000 feet. In order to reach these
figures, we have but to add to the full thickness of the palaeozoic and
mesozoic strata the present elevation of the granitic ranges above the
lowest observed stratum of the Tertiary rocks. At the beginning of the
deposition of the Volcanic Tertiary rocks, however, the upward movement
had ceased. The land had undergone enormous erosion, and subsidence
had commenced. The great ranges that had lifted their crests to such
lofty heights were again sinking beneath the sea. This subsidence did
not cease until all, or nearly all, of the mountain peaks were submerged.
It is in the strata deposited during this great subsidence that we must
look for evidences of conditions and events that made the entombment
and preservation of a vertical mile of forests possible.

The Yellowstone Valley, from the head of East Fork to the Lower
Canon, is carved out of strata which were formed along the west and
south bases of the main eastern range of mountains. In many places
the river has penetrated the full thickness of Tertiary strata, and has
cut down into bodies of metamorj)hic rocks that at the beginning of the
age were promontories or islands. It is plain, therefore, that those
parts of the tree-bearing strata examined, were deposited along a shore-
line, or, at least, near the borders of the Tertiary lake. Over large dis-
tricts there must have been, during the period of general sitbsidence, a
frequent alternation of land and sea. Land would have to exist while

132 BULLETIN UNITED STATES GEOLOGICAL SURVEY. [Yol.Y,

the forests grew and matured ; water would have to cover the same area
to deposit the succeeding stratum ; and again this stratum would have
to rise above the water before a second forest could grow. There are
two ways in which this result could be brought about. In a district
subject to such intense volcanic action as this must have been, a succes-
sion of minor oscillations might have been associated with the general
subsidence, so that large areas of the lake border districts would be
alternately above and beneath the sea, or, as was doubtless often the
case, the shallow portions of the sea became filled up with the rapidly
accumulating ejecta, and sub-aerial deposits of sufficient depth were laid
down to allow the growth of forests, which, in time, were depressed
by the general subsidence, to be buried by a succeeding stratum of the
volcanic debris. But this latter method was not the ordinary one, as is
attested by the fact that many of the forests have grown in beds of fine-
grained material that must have been formed beneath the surface of the
water.

I shall, however, not attempt to pursue this matter farther until all
the data and materials collected have been examined. A thorough study
of the various volcanic rocks will iDrobably throw much light upon this
very interesting group of strata.