GEOLOGICAL FEATURES OF THE JOHN DAY BASIN.
HISTORY OF GEOLOGICAL INVESTIGATION IN THE
Considerable confusion and uncertainty have existed
regarding geological events and their sequence within this area, due in
large measure to the fact that observations were either hastily made
during brief reconnaissance trips through the region, or were confined
to the vicinity of the richer fossil deposits. Even at the present time
no detailed geological study of this region has been undertaken. By far
the most important contribution to the subject that has thus far been
made is that by Dr. John C. Merriam, in the paper already referred to.a
Before passing to an exposition of his own views Dr. Merriam presents
the following brief summary of previous work:
The first mention of the fossiliferous deposits in
the John Day Basin which appears in the literature was made by Dr.
Joseph Leidy. In October, 1870, Dr. Leidy presented before the
Philadelphia Academy of Sciences b a short paper, in which he
described "A collection of fossils recently received for examination
through the Smithsonian Institution, from Rev. Thomas Condon, of Dalles
City, Oregon." The collection consisted of "remains of mammalia obtained
by Mr. Condon from the valley of Bridge Creek "(and "Big Bottom of John
Day"), "a tributary of John Day's River, Oregon." The collection
included new forms of Paracotylops (Merycocharus),
Rhinoceros, and Anchitherium. New occurrences of
Agriochrus, Leptomeryx, Lophiodon (?), Elotherum, and a
Dicotyles-like form were also noted. Most of the previously known
species, as identified by Leidy, were forms belonging to the White River
fauna, and he probably considered the John Day beds as of nearly the
same age as the White River.
a A contribution to the geology of the John Day Basin:
Univ. Cal., Bull. Dept. Geol., Vol. II, No.9, April, 1901, pp.
b Proc. Phila. Acad. Nat. Sci., Vol. XXII, 1870, pp.
In 1873 Professor Marsh described a several
new fossil mammals obtained by his exploring party in the John Day
country in 1871. He referred two forms to the Miocene and one to the
Pliocene, thus making the first statement regarding the age of the
In his paper on the great lava flood of the West,
Prof. Joseph Le Conte b makes the first mention of the structural
relations of the John Day formations. His statement regarding the
relation of the lava to the John Day beds is in part as follows:
"The lava of this region is * * * underlaid by the
remarkable fossiliferous Miocene lake deposit of the John Day Valley;
erosion has cut through the lava cap into the soft strata beneath."
The earliest general discussion of John Day geology
which appears in literature is the following statement published by
Marsh c in 1875:
"The Blue Mountains formed the eastern and southern
shores of this lake, but its other limits are difficult to ascertain, as
this whole country has since been deeply buried by successive overflows
of volcanic rock. It is only when the latter have been washed away that
the lake deposits can be examined. The discovery and first explorations
in this basin were made by Rev. Thomas Condon, the present State
geologist of Oregon. The typical localities of this Miocene basin are
along the John Day River, and this name may very properly be used to
designate the lake basin. The strata in this basin are more or less
inclined and of great thickness. One section near the John Day River,
examined by the writer in 1871 and again in 1873, seems to indicate a
thickness of not less than 5,000 feet. The upper beds alone of this
series correspond to the deposits in the White River Basin. The lower
portion also is clearly Miocene, as shown by its vertebrate fauna, which
differs in many respects from that above. Beneath these strata are seen,
at a few localities, the Eocene beds containing fossil plants mentioned
above. They are more highly inclined than the Miocene beds, and some of
them show that they have been subjected to heat. The inferior strata
elsewhere are Mesozoic and apparently Cretaceous. Above the Miocene
strata Pliocene beds are seen in a few places, but basalt covers nearly
In this account we find the name "John Day" first
used for the principal fossil beds of the basin. The relation of this
horizon to the great lava beds is also correctly stated, though it is
not quite clear whether he considered the Pliocene as also covered by
the basalt flows. The Pliocene referred to is pretty certainly the
Mascall beds. It is known that Marsh camped near the typical exposure of
this formation and did some collecting in it. To what Marsh referred in
his statements concerning Eocene and Cretaceous it is not certain. He
has, however, correctly described the stratigraphic sequence.
In 1880 Prof. E. D. Coped published the following
statement concerning the geology of the John Day country:
a Am. Jour. Sci., 3d ser., Vol. V., 1873, p.409.
b Am. Jour. Sci., 3d ser., Vol. VII, 1874, p. 167.
c Am. Jour. Sci., 3d ser., Vol. IX, 1875, p. 52.
d Proc. Am. Philos. Soc., Vol. XIX, 1880, p. 61.
"The regions of the John Day River and Blue Mountains
furnish sections of the formations of central Oregon. Above the Loup
Fork or Upper Miocene there is a lava outflow which has furnished the
materials of a later lacustrine formation, which contains many vegetable
remains. The material is coarse and somewhat gravelly and is found on
the Columbia River, and I think also in the interior basin. Professor
Condon, in his unpublished notes, calls this the Dalles group.
It is in turn overlaid by the beds of the second
great volcanic outflow. Below the Loop Fork follows the Truckee group,
so rich in extinct mammalia and below this a formation of shales. These
are composed of fine material, and vary in color form a white to a pale
brown and reddish-brown. They contain vegetable remains in excellent
preservation, and indeterminable fishes. The Taxodium nearly
resembles that from the shales at Osino, Nevada, and on various grounds
I suspect that these beds form a part of the Amyzon group (American
Naturalist, June, 1880), with the shales of Osino and of the South Park
of Colorado. Below these is a system of fine-grained sometimes shaly,
rocks of delicate gray, buff, and greenish colors containing calamites,a
a which Professor Condon calls the Calamite beds. Their age is
In spite of Cope's assumption that the plant and fish
bearing beds mentioned by him were to be correlated with his Amyzon
group,b Lesquereux c referred the collections from Van
Horn's ranch to the late Miocene. In a later statement regarding the
John Day stratigraphy,d Cope speaks of the calamite beds as doubtless
belonging to to the Triassic or Jurassic. This horizon was determined by
Lesquereux as Eocene.
Following is the geological section of the John Day
region as worked out by Dr. Merriam: e
River terraces, with undisturbed Quaternary
Rattlesnake formation. Gravels, ash, tuff, and
Mascall formation. Ashes, tuffs, and possibly
Columbia [River] lava. Basaltic flows.
John Day series. Ashes, tuffs, and rhyolitic flows.
Sands and gravels near the top. Lower, middle, and upper divisions.
Clarno formation. Ashes, tuffs, and andesitic and
Chico formation. Sandstones and conglomerates.
Knoxville formation. Black shales.
Pre-Cretaceous sedimentaries, serpentines. Granitic
masses of unknown age.
a This is apparently Equisetum oregonense Newberry,
b Cope, Am. Nat. 1879, p. 332, Late Eocene or Early
c Proc. U. S. Nat. Mus., Vol. XI, 1888, p. 13.
d Mon. U. S. Geol. and Geogr. Surv. Terr., Vol. III,
e Merriam, op. cit., p. 278 p. 280.
Although the oldest fossiliferous strata which have
thus far been found in the John Day Basin north of the southern portion
of the Blue Mountains belong to the Cretaceous, there are formations
exposed at a number of points that present the appearance, according to
Merriam, of being much older. Thus, on the Middle Fork of the John Day,
about 5 miles above Ritter, there are certain sedimentary rocks
bordering an area of quartz-diorite which are much more indurated and
deformed than any known Cretaceous within the basin.
At Spanish Gulch, 12 miles southwest of Dayville, the
Chico Cretaceous is seen resting upon serpentine, which has the
appearance of being intruded into it. At the head of the gulch the
serpentine is separated from what was taken to be the Chico conglomerate
by a zone of schist and quartzite. Not far from this locality there is
associated with the serpentine a considerable thickness of quartzite
with quartz veins, which have produced some gold. Limestones quite
different from any seen in the Chico are also exposed here. From the
same neighborhood the writer obtained a specimen of a granitic rock,
said to form one wall of a tunnel.
Although no direct proof can be presented, it seems
probable that some of the rocks associated with the serpentine at
Spanish Gulch are older than the Cretaceous. f
f Ibid., p. 280.
At a point 6 miles south of Clarnos Ferry, near the
junction of Muddy and Currant creeks, there are several hundred feet of
black slates. No fossils have been found in these slates, which seem
older than the Knoxville shales.
KNOXVILLE AND CHICO BEDS.
No fossiliferous Knoxville beds have been found
within the basin, and the presence of this formation is based on purely
lithological grounds. At Mitchell there is exposed a section, thought by
Merriam to be hardly less than 3,000 or 4,000 feet in thickness, which
is made up of sandstone, conglomerate, and shale. The lower portion of
this section is composed mainly of shale which, from its resemblance to
the Knoxville so usually developed in California and south-central
Oregon, is assumed to be of this age.
The Chico is exposed at Mitchell and Spanish Gulch.
Only a single fossil has thus far been afforded by the Mitchell
locality. Fossil invertebrates have, however, been obtained at two
localities near Spanish Gulch. These were submitted to Dr. T. W.
Stanton, who reports that they indicate a "horizon at or very near the
base of the Chico formation."
The name Clarno formation has been given by Merriam
to a series of beds some 400 feet in thickness which rests on the Chico
or Knoxville, and which consists almost entirely of eruptive materials
in the form of rhyolite and andesite flows and ash and tuff beds. It is
found in typical exposures at Clarnos Ferry, near the town of Fossil, on
Cherry Creek, and near Burnt ranch.
Where the Clarno has been found in contact with the
John Day there is no apparent angular unconformity of the strata. The
difference in induration and weathering is, however, very noticeable.
The sedimentary parts of the Clarno show a much greater degree of
induration than the John Day beds immediately above, and tend at all
localities to form steep bluffs, ornamented frequently with balanced
rocks or grotesque figures. a
a Merriam, op. cit., p. 286.
Thus far neither vertebrate nor invertebrate remains
have been found in the Clarno, but at most of the localities where
carefully exploited fossil plants have been found, often in abundance.
The celebrated Bridge Creek locality falls within this formation,
occurring at the base of the superimposed John Day beds. I visited this
locality in 1901 and obtained a small collection. The plants occur
abundantly in a reddish shale, which weathers whitish. The other plant
localities in the Clarno will be listed later.
JOHN DAY SERIES.
Resting directly upon and apparently conformable with
the Clarno formation is a thick series of regularly stratified sediments
now widely known as the John Day beds. This series of beds is found
quite generally throughout the basin, and represents what was called by
Marsh the deposits of the John Day Lake. The beds are made up almost
entirely of ashy or tufaceous materials, with occasionally, toward the
top, some 100 or 200 feet of a harder, blocky tuff.
The erosion forms and coloration of the John Day
strata are quite characteristic when compared with those of other
formations in the basin. In general the beds are colored various shades
of red, green, blue, or yellow. In some cases they are white or gray. As
will be shown later, the coloration is an important character in distinguishing
the subdivisions of the system. The beds are usually quite soft
and disintegrate very rapidly, forming a layer of mud several inches
thick over a large part of the exposed surface. A moderately heavy rain
starts the mud almost in streams. a
The thickness of the John Day series north of the
southern portion of the Blue Mountains is placed by Dr. Merriam between
1,500 and 2,000 feet, while to the south, in the vicinity of Logan
Butte, it is estimated to be between 3,000 and 4,000 feet.
The John Day series is divided by Dr. Merriam into a
lower, middle, and upper division. The lower division, having an
estimated thickness of 250 or 300 feet, consists of
highly colored shale which breaks down readily,
forming characteristic mud-covered domes. These beds are in the main a
deep red, with occasional alternating strata of buff or white ash. At
Bridge Creek alternating beds of red, white, and green, occurring in a
group of typical hills of this division, form a striking feature of the
landscape, the colored strata snaking sharply-defined rings about the
The middle division, having a thickness of from 500
to possibly 800 or 1,000 feet
consists of drab to bluish-green beds, sometimes
forming rounded hills, but more frequently exposed as steep, pinnacled,
and ribbed bluffs. c
The uppermost beds, showing a thickness of 300 or 400
feet, or in some cases of somewhat more, are
buff, tufaceous, or ashy deposits, sometimes with
sand and gravels near the top. . . . . They are usually harder and are
generally exposed as steeper bluffs than the strata of the lower
The lower division of the John Day series is
practically barren of fossils of all kinds, while the middle and upper
divisions have furnished a very extensive fauna. The only plant remains,
with possibly a single exception, are found in the upper division, and
even here they consist of only four or five species. The locality is
3-1/2 miles south of Lone Rock.
a Merriam, op. cit., p. 292.
b Idem, p. 293.
c Idem, p. 294.
d Idem, p. 205.
COLUMBIA RIVER LAVA.
So far as can now be made out the great Columbia
River lava once covered practically the entire John Day country, with
few, if any, points projecting above it. It consists of a large number
of basalt flows which are sometimes separated by beds of tuff, and it is
estimated by Dr. Merriam to be not less than 1,000 feet in thickness,
and in many cases it seems to be still thicker. This lava sheet has been
comparatively little disturbed, remaining practically flat over the
entire region, being rarely inclined more than 5° or 10°.
At several points within the basin there is a series
of sediments resting upon the Columbia River lava to which Dr. Merriam
has given the name Mascall formation. This series, or portions of it,
has been variously known in literature as the Cottonwood beds, Loup Fork
beds, the Ticholeptus beds, the Amyzon beds in part, and finally the
Protolabis beds. For one reason or another these various terms are
inapplicable. Thus "Cottonwood" is preoccupied by its use for a
Carboniferous formation in Kansas; its correlation with the Loup Fork,
the Amyzon beds, and the Ticholeptus beds is open to doubt, leaving only
Wortman's term, Protolabis beds, which, in Dr. Merriam's opinion, will
cover only a portion of the section. The name Mascall formation was
suggested by the occurrence of the typical section near the Mascall
ranch, 4 miles below Dayville.
At Rattlesnake Creek, near Cottonwood, the Mascall is
not less than 800 to 1,000 feet thick. The beds are made up largely of
ash and tuff, and are generally light colored, though there are some
brownish and reddish strata. Coarse, detrital materials are generally
absent from the typical section. a
a Merriam, op. cit., p. 307.
The Mascall formation has afforded a large and varied
fauna, consisting of mammals, testudinates, and fish, and a large and
interesting flora. The Van Horn's or Belshaw's ranch locality is in this
formation, occurring near the base of the section. The plants are
preserved in a soft, white, fine-grained ash or tuff, which is often 10
feet in thickness, though usually less. This material is so light when
dry that it readily floats for some time on water.
Dr. Merriam has given the name Rattlesnake formation
to a series of coarse gravels, tuffs, and rhyolite flows that rest
unconformably upon the Mascall formation. These beds are very slightly
inclined, showing a dip of only about 5°.
. . . . At one locality on Birch Creek, where a
section of the Rattlesnake was carefully examined, it was found to
comprise 30 to 40 feet of coarse basal gravels, above this about 25 feet
of soft brown tuff, and capping this about 30 feet of rhyolite. At other
localities more than 100 feet of gravel have been seen upon the
rhyolite. The basal gravel beds show a thickness of 200 feet or more in
other localities. They are frequently very coarse and contain many
pebbles, evidently derived from the Columbia [River] lava. a
a Merriam, op. cit., p. 310.
The Rattlesnake formation has therefore not yielded
fossil plants, but contains a considerable vertebrate fauna.
At many places along the John Day and its tributaries
"one or more terraces are to be found not far above the existing floor
of the valley." In several localities they have been found to contain
undisturbed remains of Elephas primigeni.