Full text of "The environment of vertebrate life in the late Paleozoic in North America; a paleogeographic study"

See other formats

I UNIVE-

RSITY OF.

CALIFORNIA
SANTA CRUZ

Digitized by the Internet Archive

in 2007 with funding from

IVIicrosoft Corporation

http://www.archive.org/details/environmentofverOOcaserich

THE ENVIRONMENT OF VERTEBRATE LIFE IN THE

LATE PALEOZOIC IN NORTH AMERICA; A

PALEOGEOGRAPHIC STUDY

E. C CASE

Professor of Historical Geology and Paleontology in the Umveisity of Michigan

Published by the Carnegie Institution of Washington
Washington, 1919

CARNEGIE INSTITUTION OF WASHINGTON
PUBLICATION NO. 283

(oil

CONTENTS.

PAGE

Introduction v

Chapter I. The Elesients of a Paleogeographic Problem i

A. The nature of the problem 2

B. Outline of points to be considered in any paleogeographic problem 2

I. Stratigraphic limits of the unit to be considered 2

(a) Isolation of the unit by the nature of the deposits 3

Marine deposits 3

Coastal deposits 3

Shallow-water deposits 5

Deep-water and abj-ssal deposits S

Cut-off arms of the sea 6

Non-marine saline deposits 6

Brackish-water deposits 7

Fresh-'water deposits 8

Subaerial deposits (fluviatile, delta, and aeolian) 8

Glacial deposits 11

Metamorphosed sediments 12

Igneous rocks 12

(6) Isolation of unit by limiting planes 12

II. Geographical limits of the unit 14

(a) Mapping of the limits 14

(6) Location of source of material 14

(d) Lateral changes in the beds l6

(e) Positive and negati^'e areas 16

III. Interpretation of the adjacent lands 17

(a) Direct contact of observed surfaces of degradation 17

(6) Phj-sical character of deposits 17

(5) Fossil content 19

IV. The fossil content of the unit 19

(a) The fauna of the unit 19

(6) Origin of the fauna 20

Aquatic invertebrate fauna 20

Terrestrial invertebrate fauna 23

Terrestrial vertebrate fauna 23

(d) Phylogenetic relations of the fauna 24

(e) Peculiarities of the fauna 25

(/) Radiation and depression of life 26

(2) The interrelations of the fauna 27

(A) Faunal elements as time markers 29

(«') The flora of the unit 29

V. Correlation of unit with other beds 3^

VI. Climatology of the past 33

VII. Distribution of the fauna and flora 34

(o) Provincial or cosmopolitan 34

(6) Distribution dependent upon the character of the biota 34

(<i) Migration 36

(e) Autochthony 37

(/) Accidental introduction 37

(g) Extinction of a fauna or flora 38

(A) Survivals and precipitate development 38

(t) Control of distribution 39

00 Enwonment 40

VIII. Checks on the geologist 42

IX. Checks on the biologist 43

(a) Bridges and barriers 43

iii

IV CONTENTS

PAGE

Chapter II. Summary Description of the Different Provinces of North America in Late

Paleozoic Time 47

A. The Eastern Province 49

I. The Northeastern Subprovince 51

(o) The Canadian region — Prince Edward Island and New Brunswick 51

(6) The Joggins section 54

II. The Southern Subprovince 64

(o) Appearance of red beds in Pennsylvania and West Virginia 65

(6) The western part of the Southern Subprovince 76

(d) Conditions in Missouri 82

Chapter III. The Plains Province 85

A. The late Paleozoic in Kansas 85

B. The late Paleozoic in Oklahoma 93

C. The late Paleozoic in Texas and New Mexico 96

D. The late Paleozoic in the northern part of the Plains Province and on the eastern front of

the Rocky Mountains 107

Chapter IV. The Basin Province 113

A. The upper Pennsylvanian in the Basin Province 113

(o) Conditions in Texas 115

(fc) Conditions in New Mexico 117

(d) Conditions in Colorado 120

(e) Conditions in Nevada 125

(/) Conditions in Utah 126

• (g) Conditions in Idaho 130

(ft) Conditions in Wyoming 130

(t) Conditions in Montana 132

(j) Conditions on the Pacific Coast 136

B. Permo-Carboniferous of New Mexico 141

C. Permo-Carboniferous of Arizona 146

D. Permo-Carboniferous of Southwestern Colorado 154

E. Permo-Carboniferous of the northern part of the Basin Province 159

Chapter V. The Late Paleozoic in British Columbia 171

Chapter VI. The Late Paleozoic in Alaska 179

Chapter VII. Interpretation of the Environmental Conditions from the Evidence of the

Deposits 187

A. Permo-Carboniferous conditions verstis Permo-Carboniferous time 187

B. Interpretation of conditions in Allegheny and Lower Conemaugh time 193

C. Interpretation of conditions in Conemaugh and Dunkard time 203

D. Interpretation of conditions in the western part of the Eastern Province 207

E. Interpretation of conditions in the Plains Province 208

F. Interpretation of conditions in the Basin Province 213

G. Interpretation of conditions in British Columbia and Alaska 216

Chapter VIII. Paleobotanical Evidence as to the Equivalence of the Beds in the Eastern

AND the Plains Provinces 221

A. Evidence of fossil insects as to equivalence of the Permo-Carboniferous beds in the Eastern

and the Plains Provinces 229

Chapter IX. Climatology of the Late Paleozoic 231

A. Climate of the late Pennsylvanian 233

B. Climate of the Permo-Carboniferous 244

C. Cause of the climatic change 25 1

Chapter X. Areal Geography of North America in the Late Paleozoic 253

Chapter XL Development and Fate of Vertebrate Life in the Permo-Carboniferous in

Relation to its Environment 263

INTRODUCTION.

The author has been led to attempt this somewhat elaborate discussion
of the environment of Hfe in the latter part of Paleozoic time by many
considerations. Two, however, have been dominant:

(i) He has for many years been concerned with a study of the vertebrate
life of the period here discussed and has, as a preliminary to more extensive
work, recorded the results of his Work on the morphology and surroundings
of that form of life in the publications of the Carnegie Institution of Wash-
ington and elsewhere. This work, dealing more fully with the conditions
under which certain groups had their inception, development, and decadence
is a direct outcome of the previous work, and, it is hoped, the first of a series
which shall embrace a discussion of this period of time wherever its records
occur.

(2) As the author has carried his work to the broader questions which
have developed he has become increasingly aware of the lack of agreement
as to the content of paleogeography and the method of procedure in solving
the various problems which arise. This publication has become, in conse-
quence, in part an attempt to crystallize in some measure our ideas of the
meaning and methods of paleogeography. The chapter upon the elements
of a paleogeographic problem is an effort to set forth in orderly form the
principles upon which such work should be concluded and the matter
that should be treated. It is the result of the direct question which the
author put to himself: What are the things that I should keep in mind
when attacking a paleogeographical problem in the field? The size to
which the answer speedily grew was somewhat appalling, but served as a
very vivid illustration of the need for just such an analysis, primarily for
the guidance of workers in the beginning of their labors. To such workers
the chapter on the elements of a paleogeographic problem is especially
addressed.

It is perhaps needless to state that to the author the term paleogeography
involves a far more complex concept than is usually recognized by writers
upon zoogeographical or paleogeographical (sic) subjects. For the discus-
sion of the content of paleogeography the reader is referred to the chapter
on the elements of a paleogeographical problem; it may be permitted to
quote here a paragraph from that discussion :

"Paleogeography is the geography of past time and is far wider in its scope
than a mere record of the extent of a bed or a formation, or the distribution of
animals or plants in any period of time. It involves all the factors which must
be considered in a modem geographical study, except the economic features appli-

VI INTRODUCTION

cable to human industry, and must take into account every influence, organic or
inorganic, which has had any bearing upon the life of the period or formation."

Of all the definitions of geography which have been offered, the most
satisfactory is that it deals with the response of life to its environment.
The method of evolution is unknown, but the directive effect of environment
is unquestioned; by whatever method new forms arise the environment
largely determines the course they shall run and the ultimate fate of the
race. It is important, then, for this work that the term "environment"
and its contents be understood. To the author's mind environment may
best be defined as the sum of all the contacts which any organism or group
of organisms establishes with the forces, and matter of its surroundings,
either organic or inorganic. The results of this somewhat complex concept
of environment are discussed in the first chapter of the book.

The author is well aware of the intricacy of the problem as here sug-
gested, but he is also most keenly aware of the inadequacy of attacking such
a problem from any more limited viewpoint. The efforts of the geologist
unacquainted with the principles of biology, or disinclined for any reason to
use them in his work, lead to imperfect and erroneous work; the converse
is just as true for the biologist. If the author shall succeed in impressing
the need for adequate training in both subjects and a careful consideration
of the problems from both sides, by all workers on paleogeography, an
important part of this work will have been accomplished.

For the rest, any success which the author may have attained in picturing
the condition under which life developed during one of the critical periods
of the earlier history of the earth is largely due to the support which he has
received from the Carnegie Institution of Washington, and it is a pleasure
to record once more his sense of obligation to that Institution and its officers.

E. C. Case.

THE ENVIRONMENT OF VERTEBRATE LIFE IN THE

LATE PALEOZOIC IN NORTH AMERICA; A

PALEOGEOGRAPHIC STUDY

E. C. CASE

Professor of Historical Geology and Paleontology in the University of Michigan

CHAPTER I.
THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM.

A. THE NATURE OF THE PROBLExM.

Adequate treatment of paleogeographic problems has been delayed by
the lack of a true appreciation of their nature. Much of the literature in
paleogeography is from the pens of workers primarily concerned \\nth the
distribution of land, water, and life, and commonly inadequately prepared
to discuss the complex factors which make up the true geography of any
time or region. Of all the numerous proposed definitions of geography, the
most satisfactory regards it as a discussion of "the response of life to the
conditions surrounding it."

Paleogeography is the geography of past time and far wider in its scope
than a mere record of the extent of a bed or formation or the distribution
of animals and plants in any period of time. It involves all the factors
which must be considered in a modem geographical study, except the
economic features applicable to human industry, and must take into account
every influence, organic or inorganic, which has had any bearing upon the
life of the period or formation. An investigator who concerns himself solely
with the distribution of life is considering the eflFect rather than the cause
and is far from viewing the problem in the broad sense in which it should
be taken up.

Obviously, then, a paleogeographic problem involves not only the deter-
mination of the location and extent of any stratigraphic unit of the earth's
crust and its biota, but all the factors which have in any way determined
the character of the rock or the manner of its deposition and all the factors
which have in any wise influenced the location, movements, and develop-
ment of the fauna and flora. It is equally obvious that such a problem is
enormously complex and may not be solved by attack from any one side
or angle. The biologist must consider the petrographic, structural, and
physiographic features of the rocks so far as they vnll serve to make clear
the physical conditions under which life existed and developed, and the
physical geologist may not neglect the laws of biology or a study of the
effects of environment upon life.'

' Clements (Scof>e and significance of jjaleo-ecology. Bull. Geological Society of America,
Vol. 29, pp. 369-374, 1918) has recently emphasized the necessity of considering and
interpreting the habitat as the causative source of developmental changes. His pajjer,
written largely from the botanical standpoint, stresses the need for the same method
of attack as is urged in this work.
3 1

2 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Paleogeographic studies have been made repeatedly from one side or
the other and maps illustrating the results presented to the scientific world.
The results have almost invariably contained some assumptions or con-
clusions immediately recognized as highly improbable or utterly impossible
by men conversant with other phases of the problem. The biologist builds
bridges of impossible position and extent to accommodate the observed dis-
tribution of plants or animals and the physical geologist describes lands or
seas where the evidence of life denies the possibility of their existence or of
their assumed character.

For this reason the following attempt at a concrete outline of a paleo-
geographic problem is presented to direct attention to the multiplicity of
factors that must be considered. One half of this outline will appear
obvious and unnecessary to the stratigrapher and the other half will appear
equally superfluous to the biologist. It is hoped that in directing attention
to both sides the errors that have formerly crept into such studies may be
in some degree avoided, and especially that students now preparing them-
selves for such work may realize the necessity for a broader training.

The present chapter is intended simply as an outline and few points
have been discussed in detail. Numerous references are given to important
articles which will direct the student to the literature, to extended dis-
cussions, and to fuller information.

B. OUTLINE OF POINTS TO BE CONSIDERED IN ANY
PALEOGEOGRAPHIC PROBLEM.

L STRATIGRAPHIC LIMITS OF THE UNIT TO BE CONSIDERED.

The paleogeography of any geological unit becomes increasingly difficult
of solution as the size of the unit is increased, whether in space or time. As
the size of the unit is increased generalities must necessarily take larger
and larger part in the observations and the conclusions will be proportion-
ately looser and less capable of direct proof. In order to obtain the most
definite results the limits, both geographic and stratigraphic, must be as
confined and as strictly and accurately determined as possible. Such isola-
tion of the unit demands not only a consideration of the stratigraphic and
geographic limits, but of the homogeneity of the deposit in all ways, mineral-
ogic, physical, biologic, etc., which reflect the uniformity of the conditions
under which the bed was formed. Vertical, horizontal, and accidental
changes are all too frequently neglected or undervalued and incongruous
elements included in what should be a carefully restricted unit. Among
the accidental inclusions may be listed, as a suggestion of the many possi-
bilities, creep or slide of material from above onto an exposed surface,
reworked material, residual matter included in cavities of an older formation
(as the Devonian matter included in rifts of Silurian limestones near

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 3

Chicago),^ river channels cut in an older formation,* etc. Such accidents
would bring in extrinsic material, both inorganic and organic.

(a) Isolation of the Unit by the Nature of the Deposits.

The most obvious character by which a bed may be isolated is the source
of the deposits, which is generally revealed in a broad way by the nature of
the materials and the included fauna or flora. Aside from the more common
criteria given in the text-books the following points may be noted :

MARINE DEPOSITS.
COASTAL DEPOSITS.

These may be either conglomerates or sandstones, resulting from the
action of waves in the advance or retreat of a strand-line. Advancing waves
will produce different results, dependent upon the character of the land
over which they make their way. If they are cutting back a high cliff of
hard rock, abundant bowlders and pebbles will be formed at the foot of the
cliff, to be later converted into a conglomerate, and if the cutting down of
the cliff is not completed the fallen blocks may be detected and identified
within the persistent mass at the foot of the former cliff, as in the sandstones
surrounding the Baraboo Range in Wisconsin.' This is a most significant
occurrence, as it locates the position of the strand-line in one interval of
time and determines one edge of the unit under consideration.

A conglomerate formed by a sea encroaching upon a cliff composed of
strata of several geological periods would contain bowlders of different
kinds of rock bearing ver>- different faunae. The age of such a conglomerate
could onh' be determined by its stratigraphic position ; it would not be at all
likely that fossil remains of contemporaneous animals would be preser\-ed
in the conglomerate, because the shells would be triturated by the moving
stones before they were cemented into the conglomerate. Even if the cliff
were formed of rocks of a limited period of time, the resulting conglomerate
would contain fragments from numerous zones of life, so that a careful
analysis would be impossible. If such a condition is suspected the con-
glomerate should be traced, if possible, to its source and a careful study
made of the physiographic conditions when the bowlders of the conglomerate
were formed. The p>ebbles of a conglomerate should alwa^'s be under
suspicion and the fauna of the pebbles never considered as representing
ofie formation or zone unless analj'sis of the source fully justifies such a con-
clusion. An admirable example of the mixture of faunae must occur in

' Weller, Stuart, Journal of Geology, vol. vii, p. 483, 1899.

' Case, E. C, Carnegie Inst. Wash. Pub. No. 207, pp. 77 and 78. 1914.

* Salisbury', R. D.. and \V. A. .\twood, The Geography of the Region .-^bout Devik Lake

and the Dells of Wisconsin. Bull, v., Wisconsin Geological and Natural History

Survey, p. 29, 1900.

4 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

the pebbles and bowlders forming at the base of the cliffs on the Gaspe
Peninsula, where an unbroken series of rocks represents several geological
periods.^

In cases of marine planation, where the waves have advanced for a
long distance either over reduced cliffs or across a country of originally
more subdued topography, similar conglomerate would result, but the
debris would probably be of smaller size, due to prolonged wave-action,
and would be thoroughly mixed and spread over a far larger area, and recog-
nizable fossils of a contemporaneous fauna would be present.

If the waters should advance over a large area where the rocks are
inclined, even at a low angle, and successive outcrops occur in parallel
bands, a mixing of the rock debris and older fauna would result from the
sequent erosion of the successively outcropping beds. We might expect
such a basal conglomerate to have been formed by the encroaching Coman-
chean Sea in Texas or more strikingly as it conquered the southern part of
the Arbuckle uplift. If the encroaching strand were parallel to the strike
of the outcrop, the various elements would be distributed in broad bands
unless disturbed by currents. If the advancing strand were at right angles
or any large angle to the strike of the outcrop, the mixing of the material
and older faunae would be far more complete.

In the case of marine planation of a land of low relief, it would be
expected that much of the material of the conglomerate would be fragments
of loosened material which had lain long upon the surface or buried in the
residual soil, and the bowlders would present a far more weathered appear-
ance than if the waves had cut back a cliff of hard rock.

In the case of marine waters breaking in upon a land of low elevation,
as the relatively rapid flooding of a peneplain or the submergence of a land
below the sea-level by the failure of barriers, the advance of the strand
would be so rapid that little effect would be produced upon any residual
masses of hard rock and there would be no evidence of sea cliffs, shelves, etc.,
except at the highest level of the water, and little or no fresh material
would occur in the conglomerate, but a large amount of weathered material
might be expected. The loosened regolith of the invaded land would be
sorted and the conglomerate would alternate with irregular beds of finer
material where quieter water or greater depths permitted its accumulation.
One would not be surprised to find in such deposits the remains of land
plants and animals. Noble, in his account of the history of the Grand
Canyon of the Colorado, cites one such instance.

* Clark, J. M., New York State Museum Memoir No. 9, 1908.

* Folio 98, Tishomingo, U. S. Geological Survey.

' Noble, L. F., The Shinumo Quadrangle, Bull. 549, U. S. Geological Survey, pp. 42 and
80-83, 1914-

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 5

SHALLOW-WATER DEPOSITS.

Following the conglomerate in the advance of the strand and finally
covering it are the shallow-water depxjsits, the finer sands, sandy muds, and
pure muds formed in the quieter region beyond the zone of wave-action.
Such deposits may result from so many different original conditions and
different forms of transportation that the utmost care is necessary in the
differentiation and interpretation of the beds. The presence of marine
fossils will at once determine the general character of the deposits, and this
may be checked, if necessary, by the adjacent formations, both horizontally
and vertically.

Regularit}^ in the bedding with a high degree of purity in the fauna (lack
of accidental inclusion of foreign forms, as terrestrial or fresh-water animals
or plants) indicate accumulation in large bodies of quiet water. Specimens
of land vegetation floated out, waterlogged, and sunk have been obtained
by dredgings far from land and in relatively deep water; such sporadic
occurrences in beds of great age are not impossible and mean no more than
accidents of distribution, but they may give a hint of the proximity of
powerful streams upon the land and some idea of the vegetation which
covered the land at the time of their deposition.

By far the larger proportion of the sedimentary beds encountered in
geological investigation are marine accumulations in shallow water, and
most of the following remarks are applicable to them.^

DEEP-WATER AND ABYSSAL DEPOSrTS,

These are most easily detected by the peculiarities of the deposits and
fauna and the absence of shore debris. Deep waters are not, however,
always remote from the shore, and the proximity of abysses even to mountain
lands, as the coast of Japan, may permit the remains of shore and land life
and shore debris to be swept out and deposited in unusual surroundings.*
Thus it would not be impossible that on the shores of Japan or the Pacific
coast of North America the impetuous streams might carry mountain forms
or mountain debris so far out that they would ultimately rest in the depths
of the abyss.

It has been pointed out that the abyssal deposits of to-day are not
necessarily the same in character or fauna as those of the remote past.
Neither red muds nor oozes may have been formed in the depths of the

* Concerning shallow-water deposits and subaerial deposits, the student should read with
close attention Barrell, Relative Geological Importance of Continental, Littoral, and
Marine Deposits, Journal of Geology, vol. xiv, Nos. 4, 5, and 6, 1906, and Bulletin
Geological Society of America, Rhythms and the Measurement of Geological Time, pt.
II, p. 776, vol. 28, 1918. While much that the writer points out is not readily seen in
limited exposures, the paper is full of suggestive points of view and working hypotheses.

' White, Da\nd, Value of Floral Evidence in Marine Strata as Indicative of Nearness of
Shore Line, Bull. Geol. Soc. Amer., p. 221, vol. 22, 1911.

6 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Paleozoic seas, and the student of paleogeography may well pause before
he declares that no abyssal deposits occur upon the continental blocks.
Chamberlin has shown that the low temperature of the deep water of the
recent oceans is possibly the result of recent changes, and totally different
conditions may have prevailed in more normal conditions of the earth. ^

CUT-OFF ARMS OF THE SEA.

Areas of water partly or completely separated from bodies of marine or
brackish water, but periodically or continuously supplied with less saline
water, become concentrated by evaporation until they are finally desiccated
or reach the point of saturation and begin to deposit their salts. Deposits
from such bodies of marine water are readily distinguished from salt lakes
or playa lakes by the character of the contained fossils. In a typical de-
velopment, the lower beds should contain a normal marine fauna which
would gradually become pauperized, losing many of its members and having
others dwarfed or developing extreme specializations.

The Gulf of Kara Bagaz, on the eastern side of the Caspian Sea, though
not marine, illustrates one case in which the process is in active development.
The Salina deposits of the upper Silurian in New York and adjacent States
is a good example of such action in past time.

NON-MARINE SALINE DEPOSITS.

Salt lakes. — Lakes without an outlet which have endured some time,
or playa lakes, subject to periodical desiccation, leave evidence of their
former existence in deposits of soluble salts. Gypsum and salt are the
most common, but borax, niter, and other less soluble substances, as the
calcareous tufas of the dying lakes of Nevada, tell the same story. Indica-
tions of such temporary bodies of water are also seen in the presence of non-
marine fossils, which show in their structure the effects of the stress of
adverse conditions. (See below, page 25.) The story of such lakes
may be in part interpreted from the succession of the deposits laid down
in the order of their solubilities; here the worker should turn to Clark's
Data of Geochemistry for invaluable information.^ Gilbert's History of
Lake Bonneville^ may be taken as a type study of such conditions and a
guide for future work.

Salt swamps. — By this is not meant the great salt marshes of the sea-
coast, where the flux of the tides inundates and drains the land, bringing
an abundant marine life and supporting the growth of a typical and luxuriant
vegetation, but rather the great areas where depression of the surface
permits the accumulation of waters from salt springs.

' Chamberlin, T. C, On a Possible Reversal of Deep-sea Circulation and Its Influence on
Geological Climates, Journal of Geology, vol. 14, pp. 363-373, 1906. Dacqu6,
Grundlagen u. Methoden der Palaogeographie, pp. 172 and 213.

' Clark, C. W., Data of Geochemistry, Bull. 616, U. S. Geological Survey, 1916.

' Gilbert, G. K., Lake Bonneville, Monograph i, U. S. Geological Survey, 1890.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 7

Such deposits would be indicated by the presence of large amounts of
water-soluble salts in connection with the peculiar structure of the beds.
The sparse vegetation and the equally sparse animal life, both of a peculiar
kind, would leave the deposits singularly barren of fossils and devoid of the
excess of carbonaceous matter which would produce the black, blue, or
green colors so common in normal swamp deposits. Salt swamps occur
to-day in some arid regions, as in the Salt Plains of Oklahoma or in Australia,
and are apt to be but phases in the life of playa lakes. The periodic desicca-
tion is likely to produce bright red colors by the oxidation of the iron in the
debris swept into the swamp by the winds and waters of violent storms.^

BRACKISH-WATER DEPOSITS.

Brackish-water deposits may accumulate in tidal estuaries and marshes,
in regions subject to periodic flooding by the sea, or near the mouths of
streams. Less common are brackish-water deposits in lakes approaching
salinity, but the fauna of these is so distinct from that of bodies of water
rendered saline by the admixture of marine waters that the distinction
would not be difficult in the deposits of past geological periods.

Estuarine deposits. — ^These are characterized by the mixed fresh-water
and marine fauna and by the rapid alternation and irregular position of the
beds due to the changing composition of the water and the shifting currents
as the tide meets the river. Could the whole length of the estuary be laid
bare, the gradual change from fluviatile deposits and faunae to marine
deposits and faunae would be apparent; river gravels and mud banks
would give place to the mixed and irregular deposits of the region of tidal
influence, and these to the regular deposits of the open sea. Where the
invading tides checked the river current the deposits would be in the nature
of delta deposits, but a delta would not form, because the ebb of the tide,
with its resultant quick outflow of the marine water and the dammed-back
river waters, would scour out the deposited sediments. However, there
would be much material left on the sides of the main channel where the
retarded stream and the inflowing tide would spread over the adjacent
lowland. The deposits here would be delta-like and flood plain-like in char-
acter, but would differ from the subaerial portion of a delta or a flood-plain
in the inclusion of marine or brackish-water fossils. Times of especially
high tide or of great flood in the streams would carry the deposits and life
of one region far into the domain of the other and might leave very puzzling
accumulations for the paleogeographer to interpret.

The upper reaches of such a large estuary as Puget Sound in the early
Tertiary would show all the conditions of a fresh-water swamp. The retarded
rivers far above the reach of salt water would spread abroad and by the filling

' Gould, C. N., Geology and Water Resources of Oklahoma, Water Supply Paper No. 148,
U. S. Geological Survey, 1905.

8 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

of the channel form stretches of swamp and morass, where abundant terrestrial
vegetation would result in the formation of highly carbonaceous beds.^

Tidal marshes, lagoons, and bayous subject to occasional or periodic
inundation by the sea would also present a mingling of brackish-water and
marine forms of life, but the deposits would be more regularly bedded and,
due to the greater intervals between periodic floodings, there would be a
zonal distribution of the fossils with a recurrence of the distinctive faunae.
This theoretical arrangement might be disturbed by the action of storm-
waves in shallow water and perhaps difificult to detect because of the thin-
ness of the successive deposits. In general, however, the deposits would
be more regular and more easily defined geographically than those of an
estuary. Mapping should give a hint as to the nature of the deposit, as the
linear form of the estuary would generally betray its character if considered
with the other features.

FRESH-WATER DEPOSITS.

Lakes of large size reproduce in some degree the form of coastal and
shallow-water deposits of the oceans, but the presence of fossils of fresh-
water life would at once distinguish the two. Due to the smaller size,
lack of tides, strong currents, and powerful wave-action, there would be less
perfect sorting of the material; i. e., the relatively far greater amount of
terrestrial material swept in would always be a noticeable factor, and no
surprise would be felt at the appearance of the remains of land animals
or plants.

SUBAERIAL DEPOSITS.

Fluviatile deposits are most apt to be encountered as seemingly extraneous
accumulations in homogeneous beds or, if making up the bulk of the observed
beds, as irregular masses difificult to resolve into their constituent elements.
When recognized and analyzed into parts, each deposit is linear in general
form if enough remains for the shape to be made out, or if the conditions are
favorable for such an observation. Not uncommonly river deposits are
revealed in cross-section by the dissection of old plains, flats, swamps, or
even marine deposits.^ In such cases the deposits were generally laid
down in more or less sharp valleys formed in an uplifted and eroded marine
deposit. Alluvial fans formed either upon land or in shallow bodies of
water of small size also belong in this group. True deltas are considered
elsewhere.

Old river-channels are generally readily recognized by the shape of the
cross-section of the deposits and the arrangement of the material as a linear,
narrow lens quite sharply marked off from the material on either side.
In such deposits the inorganic material and fossils, other than the aquatic
forms directly referable to the river itself, would be apt to resemble those

* Folios 86, io6, and 139, U. S. Geologcial Survey.

1 Case, E. C, Carnegie Inst. Wash. Pub. No. 207, p. 78, 191 5.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 9

occurring in the adjacent deposits. However, it is very possible and not
uncommon that deposits in such channels would be made up of material
carried from a considerable distance and both the inorganic and organic
contents might be derived from regions remote both in space and character
from the observed banks of the old stream. A stream descending from a
mountain or a high plateau to a lowland might sweep down and deposit
on the flat material of a totally different composition and the remains of a
life belonging to zones of radically different temperatures, altitudes, and
soils. It is possible that remains of animals of very different habitat might
be found embedded together.

The author has in mind a locality in New Mexico where such a fossil
stream-channel yielded the skull of a highly aquatic amphibian in close
association with the vertebra of a land reptile, while but a few feet away
in the red sandstones and clay which were once the margin of the stream
remains of terrestrial and swamp animals occurred in fairly regular beds.

The presence of stream-channels, except in the rare instances of under-
ground streams, is an evidence of subaerial erosion, and they must uniformly
lie below the level of the lower plain of an unconformity. There may be a
bending down of the line of unconformity at this point if the old stream-
valley was a wide one or had cut deeply into the older rocks. In cases where
the river deposits were slight in amount or where the interval of exposure
was long with much erosion, or where marine planation had followed, all
traces of the river might be removed.

Flood-plain deposits are not infrequent.^ The flooding of streams from
whatever cause results in accumulations of material in wide flats which
may attain considerable thickness. The fluctuation of currents in the
flood waters and the very position and duration of the deposition causes
the beds to be extremely irregular in arrangement and composition. Rapid
variations of the strike and dip of outcrops, prevalent discontinuity of
individual beds, cross-bedding, and truncation of older beds by younger, are
common characters. Conglomerates, sandstones, shales, and muds alternate
rapidly. Such deposits yield a pretty full record of their history upon close
examination.

The physical characters of flood-plain deposits formed in arid and
humid regions would be in many ways very similar, though it is probable
that the violence of the floods of arid regions would leave their record in the
coarser material and the evidence of stronger currents.

The chemical and physical character of the material reveal to a large
extent the climatic conditions under which the beds were formed. Flat
or flood-plain deposits of arid regions are marked by the presence of highly
oxidized or carbonated minerals with a lack of hydrous oxides or sulphides,

' For an attempt to describe a flood-plain and subaerial delta region and reveal its history,
see Case, Carnegie Inst. Wash. Pub. No. 207, 1915-

10 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

etc. This is largely due to the normally low water-table, which permits the
penetration of the air deeply into the soil, and the exposure of the mineral
constituents to oxidation or carbonation.^ Also, the lack of vegetation on
an arid flat means a lack of carbon, so that there remains a larger proportion
of ferric oxide, which requires the action of CO2 as an intermediate step
to its solution and removal, and a lack of reduction of the other higher
oxides, sulphates, and carbonates. The common result is the prevalence of
a red color,^ the presence of gypsum associated with the remains of terrestrial
animals, and a lack of plant remains. Wind-ripples, tracks of animals,
sun-cracks, etc., are common but not always conclusive. The author well
remembers his surprise upon examining a wind-rippled surface of desert
sand in Arizona to find many markings which he would have unhesitatingly
pronounced worm-markings upon the wave-rippled surface of a marine
sandstone if the surface had been found fossil, but here the ripples were
formed by the wind — were forming as we watched — and the wormlike mark-
ings were formed by burrowing insects creeping just below the surface of the
heated sands, leaving trails similar to those formed by moles working their
way through a light soil ; nor was he able to detect a single criterion which
would have caused him to reverse his decision if the surface had been an
exposure of ancient sandstone.

In flats formed in humid climates the higher water-table acts by prevent-
ing a free circulation of air and inducing a heavier growth of vegetation.
The excess of carbon and lack of oxygen permits the formation of the lower
oxides, hydrous oxides, sulphides, and even the native metals, such as copper.
The dominant colors are black, blue, green, yellow, from the presence of
compounds of iron low in oxygen. Traces of plants are frequent, and such
marks as footprints, rain-drop impressions, ripple-marks, etc., will be as
common as upon an arid flat.

Alluvial fans and river deposits in general. — A detailed discussion of the
relation of alluvial fans, terraces, and river deposits to climatic conditions
in general has been given by Barrel and need not be repeated here, even
in part; the student is referred to the paper as a type study of such con-
ditions.'

Delta deposits. — So full a discussion of the deposits has been given by
Barrell that only a reference to his articles is necessary.* Dacque has

' A storehouse of information in regard to chemical changes will be found in Clarke, F. W.,
Data of Geochemistry, Bull. No. 616, U. S. Geological Survey, 1916; and Van Hise,
C. R., Metamorphism, Monograph 47, U. S. Geological Survey, 1904.

' Tomlinson, C. W., The Origin of Red Beds, Jour. Geol., vol. 24, pp. 153 and 238, 1916.

' Barrel, Jos., Relations Between Climate and Terrestrial Deposits, Jour. Geol., vol. xvi,
Nos. 2, 3, and 4, 1908.

* Barrell, Jos., Relative Geological Importance, etc.. Jour. Geol., vol. xiv, Nos. 4, 5, and 6,
1906. Criteria for the Determination of Ancient Delta Deposits, Bull. Geol. Soc.
Amer., vol. 23, No. 3, 1912. The Upper Devonian Delta of the Appalachian Geosyncline,
Amer. Jour. Sci., vol. 36, November 1913, and vol. 37, January, March 1914. (This last
is a study of a type area.)

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 11

drawn attention to the similarity which exists between the deposits of deltas
and those of a transgressing sea.^

Aeolian deposits. — Dune sands, loess, and volcanic ash are the most
common wind-carried material.

Dttne sands are generally cross-bedded, but the line of the cross-bedding
is more concave than where it is formed by water-currents.* The lower
part of the line is more nearly parallel to the surface below and then rises
in a sharp curve ; in water-laid beds the lines are straighter and at a sharper
angle. Aeolian sands are generally very pure, clean, quartz grains with
few or no traces of life in them. The character of the grains and the criteria
for distinguishing aeolian and subaqueous sands are given by Sherzer.'

Loess is a light reddish, gritty clay, frequently splitting up in vertical
columns and marked by vertical tubules due to the decay of plant roots.
Not infrequently land and fresh-water shells are present.* Such accumula-
tions are the result of dust-storms or less violent but more persistent trans-
portation of the fine rock debris continued over great distances and in
enormous quantity'.*

Volcanic ash is especially liable to be carried for great distances from its
source, as it is thrown high into the air and may be caught by strong and
persistent winds of high altitudes. Such drifted material frequently forms
large accumulations, especially where it has fallen into bodies of water.
Thick masses accumulated in the Rock>' Mountain region in Tertiary times
were long regarded as lake clays from their fossil content, but the microscope
revealed their origin.*

Some accumulations have been found evidently formed upon land far
from their source, as in Oklahoma, when the nearest source was the north-
eastern Xew Mexican volcanic region.' On the Pacific coast such accumula-
tions carry- marine fossils, showing that volcanoes, perhaps located near the
margin of the continent, had contributed materially to the marine sediments.

GLACL\L DEPOSITS.

Aside from the surficial deposits of the last geological period, whose
recognition is a distinct study, the traces of glaciers of the past are hardened
and cemented till (tillite), scratched and soled bowlders, rounded and

* Dacqu6, E., Gnindlagen u. Methoden d. Palaogeographie, p. 147.

* Grabau, A., Principles of Stratigraphy, p. 701.

* Sherzer, W. H., Criteria for the Recognition of Various Tyf>es of Sand Grains, BuU. GeoL

Sec. Amer., vol. 21, p. 625, 1910.

* Shimek, B., Various papers, mostly in the Iowa Academy of Science.

' Keyes, C. R., Deflation and the Relative Efficiencies of Erosional Processes under Condi-
tions of Aridity, Bull. Geol. Soc. Amer., vol. 21, p. 565, 1910. Mid-continental Eolation,
idem, vol. 22, p. 687, 191 1.

•Sinclair, W. J., Volcanic Ash in the Bridger Beds of Wyoming, Bull. Amer. Mus. Nat.
Hist., vol. 22, p. 273, 1906.

' Buttram, Frank, Volcanic Dust in Oklahoma, Oklahoma Geological Survey Bull. 13, 1914.

12 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

scratched rock surfaces, beds of conglomerate composed of angular pebbles,
and the presence of large erratic blocks in marine deposits.

The typical occurrences of such evidence in rocks of Permian and
Cambrian age are obvious enough, but more obscure evidence is difficult to
judge. Erratic blocks may be carried trapped in the roots of drifting trees
and deposited far from shore, or they may be carried by icebergs for enormous
distances. Angular conglomerates may result from landslides, rock glaciers,
or even sudden and violent floods originating on hillsides from cloudbursts.
Their interpretation when discovered must be cautiously approached.
Marks simulating glacial scratches may be produced by the slipping of
masses of rocks or even by the movements of sediments previous to their
cementation into rock.^

METAMORPHOSED SEDIMENTS.

These present so many difficulties in their interpretation which must
be solved by severely technical methods that the problem must be in part
left to the specialist in petrography, but when the original nature of the
rocks is determined the history can be read upon the lines which have been
suggested above. No paleogeographer should neglect the important reve-
lations that may come from metamorphic rocks when completely and cor-
rectly interpreted.

IGNEOUS ROCKS.

Unless secondarily deposited by water or wind, igneous rocks affect
the problem only in an indirect way. The alteration of any sediments by
intrusion or burial by igneous rocks may aid in delimiting a unit, or the
stratigraphic relations to younger or older igneous rocks may be a determin-
ing factor. Changes in color and texture due to local metamorphism can
usually be easily detected.

(b) Isolation of the Unit by Limiting Planes.

A stratigraphic unit may be sharply set off from adjacent units by
structural breaks, by erosional breaks, by changes in the character of the
material, by changes in the bed alone; or it may pass so imperceptibly into
one or other of the adjacent beds that the line of separation is indistinguish-
able or only distinguishable by paleontological evidence.

In the case of overthrusts, where older beds are forced above and across
younger beds, the line between the two is generally very clearly marked
both by the sudden change in the character of the material and contained
fossils (if present) and by the disturbance of the rocks accompanying the

• Woodworth, J. B., Boulder Beds of the Caney Shales at Talahina, Oklahoma, Bull. Geol.
Soc. Amer., vol. 23, p. 462, 1912.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 13

movement. A typical case upon a large scale is the great overthrust on
the eastern side of Glacier National Park.^

Erosional breaks are the most common and most looked for limiting
planes, but would be apt to be present upon only one side of a unit of small
stratigraphic extent. The various t\'pes of unconformity are explained and
illustrated in every textbook and in typical cases are easily recognized, but
increasing attention is being paid to the determination of minor uncon-
formities and unconformities obscure because of the position of the beds.
Where an exposed bed is nearly horizontal and is but little dissected by
erosion, or where long erosion has reduced the exposed surface of horizontal
beds to a near level surface, the succeeding deposit may be so nearly parallel
and conformable as to present the app)earance of uninterrupted deposition.
Schuchert has described and illustrated typical instances of this condition,'
and Ulrich has repeatedly drawn attention to the necessity of determining
even minor unconformities.

Erosional periods followed by the deposits of transgressing and retreating
seas result in peculiarities of the unconformity which have been interpreted
by Grabau.'

The true meaning of an unconformity is not always fully realized by
paleogeographers, either as to the time involved, with all the possibility of
structural and surface changes and changes of environment, or as to the
amount of geological record lost by erosion and by lack of deposition during
the period of exjx)sure. Blackwelder* and Bcurell* have called attention
to the importance and the significance of these erosional intervals.

Changes in bedding, in material, or in color may mean much or little,
dependent upon conditions. In horizontal, persistent beds, evidently de-
posited under uniform conditions in quiet water, such changes would at
once attract attention as indicative of some considerable disturbance either
of surface or climate and hence of the life; but in delta, fluviatile, terrestrial,
or even shore deposits, frequent alterations of bedding, of material, or of
color are entirely consistent with unchanged conditions upon the land.
Small bodies of water may easily receive material from different sources
within a limited period of time; rivers may change their course or alter the
velocity' of their currents in a capricious manner; occasional floods may
sweep together material not deposited in the same place under normal

' Campbell, M. R., The Glacier National Park — A Popular Guide to Its Geology and
Scenery, Bull. 600, U. S. Geological Survey, 1914.

* Pirsson, L. V'., and Chas. Schuchert, Text-book of Geology, pp. 586-587, 1915.
Schuchert, Paleogeography of North America, Bull. Geol. See. Amer., vol. 20, p. 441, 1910.
Barrel, Jos., Rhythms and the Measurement of Geological Time. Illustrations of Rhythms

in Sedimentation. Bull. Geol. Soc. Amer., vol. 28, p. 798, 1918.

* Grabau, A., Types of Sedimentarj- Overlap, Bull. Geol. Soc. Amer., vol. 17, p. 567, 1907.

Principles of Stratigraphy, chap. xvin.

* Blackwelder, Elliot, The Valuation of Unconformities, Jour. Geol., vol. 17, p. 289, 1914.
' Barrel!, Jos., Rhjthms and the Measurements of Geologic Times. Illustrations of RJiythms

in Sedimentation. Bull. Geol. Soc. Amer., vol. 28, p. 798, 1918.

14 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

conditions. Anyone familiar with the sudden appearance of an abundant
vegetation in an arid region can realize how a single season of exceptional
rainfall might furnish enough carbonaceous material to radically change the
color of a goodly thickness of deposits. The subaerial deposits of the
Permian and Triassic are notably lacking in persistence, either as to bedding,
material, or color. Williston and Case have asserted their conviction that
sections made at any point in these beds can not, in most cases, be depended
upon a quarter of a mile away.

II. GEOGRAPHICAL LIMITS OF THE UNIT.

a. Mapping of the limits of the unit, preferably upon a topographic base,
is a primary essential. Most commonly any exact determination of the
outline is difficult or impossible because of the burial of a part of the unit
under younger beds, because of the destruction of a part of the bed by
erosion, or because of its interruption by structural changes. The location
of hypothetical limits is always a most uncertain process and demands
the utmost care and the use of every possible check, such as the physiography
of the underlying beds, recognition of the horizontal changes in the material,
identification of outliers, interpretation of well records, correlation with
other outcrops, etc.

h. Location of the source of the material. — As suggested above, the location
of the inner line of the deposit and the origin of the material is of the utmost
value. If the deposits are largely or even in part clastic in character, a
knowledge of the source will give much information as to the course and direc-
tion of the transporting currents, the distance covered in transportation,
and the weathering or other changes that the material has undergone. At
the same time, an idea may be gained of the physiography of the old land.^

In the case of a transgressing sea with progressive overlap^ of the beds,
new material will be constantly gained by the waves and mingled with
material carried in by rivers. The basal beds will be likely to be conglomer-
ates, especially if the sea is advancing against a resistant coast, and this
conglomerate will be a useful and easily followed criterion in determining
the limits of the beds. In a regressing or stationary sea the upper beds
will be finer as the waves work over the already deposited material or dis-
tribute the material carried in by rivers. Great flats would be formed as
the sea retreated from its shelf and the streams poured out their waste. If,
however, the retreat of the sea were rapid, caused by a sudden uplift of the
land, the quickened streams might carry out much coarse sediment which,
reworked by the tidal flux and by storm waves, would need careful observa-
tion of its character and content, petrographic and fossil, to distinguish it

1 Barrell, Jos., Relation Between Climate and Terrestrial Deposits, Jour. Geol., vol. xvi,

Nos. 2, 3, and 4, 1908.
* Grabau, Types of Sedimentary Overlap, Bull. Geol. Soc. Amer., vol. 17, p. 567, 1906.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 15

from a basal conglomerate. Such regressive or negative movements of the
sea, caused by uplifts of the land or over-deepening of the ocean basin,
is generally more rapid than positive movements and a quickening of the
streams frequently affords much information concerning the adjacent lands.
If the land were still unsubdued, the rivers would run over hard rocks and
coarse deposits would be brought to the sea, while if the land were low and
covered with a mantle of residual soil the river-borne sediments would be
finer and result in clay beds, or insufficient in amount to prevent the growth
of organisms which in favorable localities secreted CaCOs in large quantities,
resulting in beds of limestone. In either case the change of sediments would
not be conspicuous if the change from advance to retreat of the sea were
actually or relatively sudden, but a change from a static strand-line to a
retreat would result in a radical change of the deposits.

Such soft deposits as would follow a retreating strand under the circum-
stances cited above would be especially liable to destruction in any reverse
movement of the strand-line; the loose material, unless cemented with
exceptional rapidity, would be easily torn up and redistributed by the waves
and the coarser material would be sorted out as a basal conglomerate.
However, the ad%ance of the strand over such a flat would be quite rapid
and the pebbles would be little worn by the waves and would be more likely
to retain the character of river pebbles than to assume that of beach pebbles.

Such unconsolidated material also would be subject to rapid subaerial
erosion; the inner limits would soon disappear and the connection of the
unit with the edge of the old land would be effaced and its original continua-
tion recorded, if at all, in occasional outliers, as on the Atlantic and Gulf
Coastal Plains.

If the elevation of the land were of such a character as to form mountains
or steep slopes close to the most distant line of retreat of the strand, such a
cycle of events as suggested above would not occur, for the material from
the land would be carried out into deep w^ater, or at least into the zone of
wave action, where it would at once be distributed in the form of pure
marine sediments.

c. The seaward limits of deposits. — The outer limits of any marine unit
should coincide with the limits of deposition for any position of the strand-
line. In the open sea the approach to such an outward limit would be
indicated by the occurrence of progressively finer material until the clastic
debris was replaced by organic or chemical deposits. In a typical case the
littoral, ben thai, and abyssal zones would be marked by the fossils and the
nature of the material. Such a typical series would extend over so broad an
area that some portion would be almost certainly concealed by later deposits
and only a ver>' broad elevation of the land and only exceptional conditions
of erosion or structural changes would expose the whole histon,-. The
determination of the outer limits of a marine deposit would be one of the

16 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

most difficult parts of a problem to solve, especially if those limits lie in the
open-sea basin rather than in a gulf or bay.

The character of sedimentation in shallow seas has recently been discussed
by Barrel 1.^

d. Lateral changes in the character of the deposits of a marine unit,
either those in sequence from the shore outward or those parallel to the
strand-line, introduce most puzzling elements into the problem. Uncon-
sciously the assumption is always made by the student, especially in his
earlier experience, that a unit must be homogeneous throughout its extent —
alike in all its parts. The tracing of a formation by actual continuity may
easily be rendered a difficult if not impossible task if the possibility of lateral
changes is not kept in mind. Once the direction of the shore is determined,
it is commonly assumed that each zone parallel to this will be made up of
similar material and carry similar fossils in all parts. In the Paleozoic,
before zones of climate were established or sharply differentiated, and when
it is very possible that for a majority of the time the topography of the shores
was far less diversified than now, the chances for the existence of long zones
of similar deposits and fossils were far greater than in later periods, but even
in that time it is to be expected that more rugged coasts gave place to wide
estuaries, and that wide sandy beaches stretched along the coast at intervals ;
lowlands stretching back from the sea or offshore bars and reefs would have
their characteristic effects, all contributing to diversify the deposits of any
definite interval of time. Animal life is not conditioned by temperature
alone; the physiographic conditions suggested above would cause a diversity
of food -supply and habitat which would compel a diversity of life, and
currents controlled by the nature of the coast would influence the distribu-
tion. Then, as now, any zone of reasonable length would have varied
deposits and life in its different parts.

e. Positive and negative areas. — It is obvious from an inspection of paleo-
geographic maps that the submergences of the continents have repeatedly
occurred over well-defined areas. This is clearly seen in the continent of
North America. Schuchert's Paleogeographic maps^ and Ulrich's table of
submergences^ bring out the plan, and it is recognized in the mapping of the
positive and negative areas first suggested by Willis.^

From the Atlantic deep around the northern and southern ends of
Appalachia and, rarely, across it in the vicinity of New Jersey; from the
Arctic deep broadly across the Hudson Bay region, or more commonly from
the northwest down the course of the Rocky Mountain prism and then

1 Barrell, Jos., Rhythms and Measurements of Geologic Time. Rhythms in Sedimentation,

Bull. Geol. Soc. Amer., vol. 28, p. 776, 1918.
' Schuchert, Chas., Paleogeography of North America, Bull. Geol. Soc. Amer., vol. 20, 1910.
' Ulrich, E. C, Revision of the Paleozoic Systems, Bull. Geol. S^c. Amer., vol. 22, pp.

346-347, 191 1.
* Willis, Bailey, A Theory of Continental Structure Applied to North America, Bull.

Geol. Soc. Amer., vol. 18, p. 389, 1907.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 17

southeast across the Dakotas; from the Gulf of Mexico northward on either
side of the Ozarkia; and finally, broadly eastward from the Pacific deep,
the waters of the oceans crept over the lands, at times spreading widely,
at others confined to relatively limited channels. The determination of this
plan of submergences is one of the greatest steps that has been made in
the preparation of a logical geological history of the continent.

III. INTERPRETATION OF THE ADJACENT LANDS.

The paleogeography of any unit is far from completely made out, even
when the constituent rocks and fossils are thoroughly known. The composi-
tion and arrangement of the material in any bed deposited on an ocean
littoral or in a smaller body of water is influenced in large measure by the
nature of the land from which it was derived. The temperature of the
water and the food-supply of aquatic faunae are no less closely influenced
by the condition of the bordering lands. Terrestrial deposits reflect even
more closely the character of the adjacent degrading land. It is obvious
that to understand the paleogeography of any unit of time it is necessary to
know the condition of the land areas which have contributed the sediment
to the observed stratigraphic unit. Such knowledge is gained only with
great difficultv' in many cases. The debris from the land which forms the
observable record has in most cases undergone decided changes and must
be interpreted in the light not only of all possibilities of weathering and
alteration of the original material, but from the method of transportation.

a. Where direct contact can be established the land surface may in part
be made out from the suggestions given above and below. Where, as in
the great majority of cases, the littoral zones have been destroyed, the
source of material may be recognized by the character of the sediments or
inferred from possible sources of supply such as elevated regions. In such
cases the nature of the contact between the eroded surface and the overlying
beds may tell the extent of its degradation and the character of the surface.

b. The physical clmracter of the deposits may reveal the greater or lesser
degree of weathering, erosion, and transportation, and hence the ruggedness,
gentleness, the velocity' of the streams, the amount of protecting vegetation,
climatic variations, etc. The included fossils of a land vegetation may
show something of the nature of the soil and climate. Volcanic ash, loess,
and wind-blown sand all help to restore the condition of the land. A typical
case of the interpretation of wind-drifted sand is to be found in Grabau and
Sherzer's discussion of the Sylvania sandstone of southeastern Michigan.^

Wind-blown sand, volcanic ash, and loess, however, carry far less informa-
tion concerning the adjacent areas than would a water deposit. More easily
carried and far more thoroughly sifted from all foreign substances, these

* Grabau, A., and W. H. Sherzer, The Monroe Formation of Southern Michigan and Ad-
joining Regions. Michigan Geological Survey, series i. Pub. 2, p. 6i, 1910.
3

18 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

materials tell little of the place from which they came. Dunes may form
on the edge of lakes or oceans, or in the most stark desert, and in either
place the information given by the structure and material of the dune would
be essentially the same. The constituent sand might be gathered from
barren mountain peaks, dragged from a normally well-covered region in
time of drought or failing vegetation, or swept from river bars, lake or ocean
beaches.

Loess and volcanic ash tell even less than aeolian sand. These are so
light that they may be transported far from their place of origin and laid
down in the most dissimilar places, in bodies of water, on lands covered
with vegetation, or in deserts.

c. The mineral content is instructive only where conditions are most
favorable. No one would be able to interpret from the deposits of a swamp
in the Archezoic rocks of Canada the nature of its surroundings if the
chemical content of the finer muds were alone observable. A thoroughly
decomposed mass derived from various igneous rocks would give upon
analysis results very similar to those obtained from many shales, but if the
mineral content is still determinable by petrographic methods, or if coarser
deposits on the borders of the swamp or near the inflowing streams are
observable, much might be learned.

An occurrence of scattered local deposits of fine-grained shale rich in
carbonaceous material or plant remains, accompanied by fresh-water fossils,
and perhaps by a quantity of bog iron ore in association with pebbles of
igneous rock in advanced stage of chemical decomposition would strongly
suggest such conditions as now prevail in many parts of Canada. An
abundance of angular fragments of similar igneous rocks with swamp
deposits would suggest the former existence of such swamps as occur in
the higher mountain parks, while, of course, striated pebbles would lead to
the consideration of the possibility that glaciers had had some part in the
formation of the swamp by interfering with established drainage.

A coal swamp in a limestone region would be even less liable to disclose
the nature of the surrounding land, especially if it were of large size. The
fragments of limestone carried into the swamp would disappear by solution
and the infrequent stream-channels would retain such a relatively large
residium of the insoluble material, as quartz sand, as to lead to erroneous
conclusions, unless studied with the utmost circumspection. River deposits
would be far less dependable as indices to the nature of the surrounding
lands than deposits in bodies of quiet water. Large streams carry material
for great distances, and the content of any fossil river-channel might be the
result of accumulations from widely separated sources. The presence of
igneous fragments in the middle or lower courses of the Mississippi would
obviously not be a safe proof that the shores adjacent to the place where the
samples were taken were formed of igneous rocks. Floating ice might

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 19

carr>- such igneous fragments from far north or even farther west, if the
samples were taken below the mouth of the Missouri River, or the fragments
might have been gathered from the glacial drift of the banks. Certainly
the content of a fossil river-channel would present no less difficulty in the
interpretation of its origin. However, in such deposits the banks are
usually observable and much possible error easily avoided.

d. Fossil content. — ^The life of marine and even fresh waters may to some
extent show the nature of the adjacent land as the habits and food-supply
cause it to multiply or disappear upon shores of different character.

Reefs of Bryozoa and corals give a direct suggestion. The animals are
fixed in position; the tentacles and mouths are directed upwards; any
large amount of sediment falling through the water would sjjeedily cause
the extinction of most forms, though some are found living in very muddy
waters. The occurrence of reef corals in lar^e numbers in any place suggests
clear water far from the mouths of great rivers or muddy currents, and
opposed to coasts from which relatively little material is being washed out —
a land with low or wooded slop>es upon which the solvent processes of
degradation are more active than those of mechanical disintegration.

Other animals, as many mollusks, live near muddy coasts, or on sandy
flats, etc. The student must here turn to some treatise on zoology for a
discussion of the life habits of various organisms. (See also Walther,
Einleitung in die Geologie, II Theil, and Grabau, A., Principles of Stratig-
raphy, chapter 2&, Bionomic Characteristics of Plants and Animals.)

IV. THE FOSSIL CONTENT OF THE UNIT,
(a) The Fauna of the Unit.

The fossils of animals are apt to be found in all kinds of deposits; marine
beds carry by far the largest number, but terrestrial beds are frequently
rich in the remains of animal life. All fossils found in water-laid beds are
not aquatic forms; the remains of purely land animals find their greatest
chance of preserv'ation when swept into bodies of w^ater, or buried in the
deltas, sand-bars, mud-flats, etc., of rivers. In the case of floating carcasses,
distended by the gases of decomposition or supported by floating vegetation,
the remains might be swept far beyond the limits of deltas in large bodies
of water and come to rest in the horizontal beds of quiet, deep water.
Agassiz found remains of land vegetation on the bottom of the sea in the
Antilles, i,ooo fathoms down, and the Challenger dredged up plant remains
from as much as 1,400 fathoms in Polynesia.^ Animal remains might go
approximately as far from the lands and sink in as deep water.

Land birds and insects are frequently blown far out to sea, only to p>erish
and sink to the bottom or be devoured by fishes. Such accidental inclusions

* Suess, E., The Face of the Earth, English edition, vol. 11, p. 248.

20 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

should not be permitted to cause an error in the interpretation of the beds.
In swamp deposits and deposits of small bodies of water, either fresh or
marine, the remains are apt to be those of the local fauna, terrestrial or
aquatic.

The bodies of terrestrial animals which find their way into a stream
may be carried long distances with the current and in times of flood may
be carried far outside the normal bed of the stream and laid down on flood-
plains or in places where the streams spread widely over the subaerial
portion of deltas. Even after the cadaver, freed from the distending gases,
has sunk or been torn to pieces, the parts would be swept along until finally
drawn into some eddy or stranded upon a flat. Where streams pass rapidly
itom one physiographic region to another this might result in the mingling
of fossil forms very distinct in their natural habitat, as the remains of purely
mountain or upland forms might to-day be swept out upon the surface of
the Great Plains and mingled with remains of animals of radically different
habitat; or animals entirely inland might be swept out by the floods of the
Mississippi, Nile, Amazon, or other great rivers, and buried in subaqueous
parts of the delta far from shore and intermingled with remains of marine
animals. One would not regard as impossible the occurrence of the bones
of the American antelope or the bison in the muds of the Mississippi delta,
to take an extreme case, or the bones of horses, cows, etc., in muds of the
Louisiana bayous where such remains would not naturally occur. An ex-
ample of the determination of the physiographic habitat of animals is given
by Osborn.^

More difficult is the interpretation of the contents of large bone-beds or
shell-beds in fluviatile deposits. It would be obviously very dangerous
to interpret the surroundings of such a collection from the contents of the
bed until a study of the fossils permits the elimination of foreign forms.
Much-worn bones or shells would naturally indicate long transportation,
but when the cadavers were transported a great distance before being
destroyed or the hard parts subjected to much wear this evidence of trans-
portation would be less noticeable.

The fauna of a bed, aside from the accidental inclusions noted above,
indicates the character of the deposit — marine, brackish, or fresh water,
swamp, or purely terrestrial. But here a new series of factors enters the
problem; the nature of the evidence shifts from the inorganic to the organic.
The time element becomes as important a factor as the space element.

(6) ORIGIN OF THE FAUNA.

Aquatic invertebrate fauna. — ^Any fossil fauna either originated where it is
found by evolution from older types, or it migrated into the region, or it

' Osborn, H. P., Cenozoic Mammal Horizons of Western North America. Bull. U. S.
Geological Survey, No. 361, Age of Mammals, pp. 84-85, 1909.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 21

resulted from a mixture of the two processes. In the first case the group of
animals was isolated for a long time and will bear evidence of its history in the
presence of archaic forms and peculiar specializations such as always arise in
isolated communities. Cases are rare where it can be shown that epiconti-
nental seas existed so long undisturbed that actual evolutionary changes are
apparent. Such a case perhaps is the Gaspe region in northeastern Quebec.
Ulrich asserts that evolution has always taken place (in the Paleozoic inverte-
brates at least) in the ocean basins and not in epicontinental seas; that the
faunal changes noted and attributed to evolutionary processes are due to a
misapprehension of the composite nature of the beds, and that the changes
are due to a retreat of the fauna to the ocean basins and its return after
undergoing an evolution there. ^ If epicontinental beds can be actually
isolated which show the long-continued and uninterrupted presence of the
sea, far-reaching conclusions as to the conditions of adjacent land and sea
are usually forthcoming.

In the second case the evidence is more readily detected. The sudden
appearance of a new fauna or of new tj-pes in an old fauna is almost invariably
due to migration. The source of and the route of the migration become
at once of interest and may be revealed in many ways.

In marine beds the appearance of new forms may perhaps be correlated
with the advance of the strand as in overlaps, etc., or it may be due to
rapid submergences of a land area following the breaking-down of barriers
by the action, slow or rapid, of physiographic forces. It is not improbable
that the sudden diversion of the Colorado River left many fresh-water forms
in the layers of mud it deposited in the Salton Sink. Noble has indicated
a somewhat similar action, but of marine waters, in Paleozoic time, revealed
in the section of the Grand Canyon of the Colorado.*

The partial submergence of the continent of North America in Cambrian
time is a similar case in point.'

The third case, where resident and migrant faunae are mingled, is by far
the most common. Classical examples of this condition occur in the various
troughs of the Appalachian Basin, where new faunae repeatedly penetrated
through channels, the Quebec, Levis, etc., from the North Atlantic, and
across the weak spot in the barrier of Appalachia near Chesapeake Bay, or
around the southern end of Appalachia from the mid-Atlantic*

' Ulrich, E. C, Re\ision of the Paleozoic Systems. Bull. Geol. See. Amer., vol. 22, pp.

495-505. 1911-

* Noble, L. F., The Shinamo Quadrangle, Grand Canyon District, Arizona. The Hotatau

Conglomerate. Bull. 549, U. S. Geological Survey, 1914.

* Walcott, C. D., Abrupt Appearance of the Cambrian Fauna on the North American

Continent. Smiths. Misc. Coll., vol. 57, No. i, pp. 1-16, 1910.

* Weller, Stuart, The Paleozoic Faunas of New Jersey, New Jersey Geological Survey,

vol. 3, 1903.
Ulrich, E. C, and Chas. Schuchert, Paleozoic Seas and Barriers, Bull. 52, N. Y. State
Museum.

22 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Equally good examples occur in the various deposits of Devonian time,
when invasions from the Arctic, across Hudsons Basin and from the north-
west, or from the Gulf of Mexico, successively penetrated to the center of
the continental surface.^

Here arises at once the question of the source and routes of movements
of the migrant forms. The determination of the identity or distinctness of
two faunae, as the resident and the migrant, depends upon their composition
and immediately introduces the question whether a fauna is to be charac-
terized by the similarity of a majority of its species to those of another
region (matching of species) or by the presence of a few unique forms.
These questions have been fully discussed by competent authorities and is
further considered under the subject of correlation below.^

The sudden appearance in a bed of new forms similar to the original
ones, but recognizable as migrants from another region, and the absence
of new types of life implies very similar conditions of water, temperature,
food-supply, etc., in the old and new homes of the migrant forms and in all
intervening places on the route of migration. The only disturbing factors
would be those arising from competition between resident and migrant
faunae. Such an invasion could only arise when the migrations were made
possible by very gentle movements between regions similar in all general
conditions. It is not unlikely that similar temperature conditions prevailed
widely over the earth in Paleozoic times,^ but differences in bottom, food-
supply, and so forth, could easily vary as the waters bordered on different
terranes. These suggestions will be made plainer by a consideration of the
faunae of any part of the continuous seacoast of any continent to-day.

Shore-lines extending across latitude lines have very different faunae,
controlled by temperature, though other factors are also present, and
ocean currents parallel to the coast may extend the range of faunae to the
north or south beyond the effect that would be produced by latitude alone.
The effect of the Greenland Current and the Gulf Stream on the east coast
of North America is a good example. The use of this principle in paleo-
geography is illustrated by Willis's paleogeographic maps, in which an
attempt is made to indicate the course of currents.*

It must be recognized that the migration of invertebrates is largely
accomplished by currents. The free-swimming forms would be borne by

* See Ulrich's Revision of the Paleozoic Systems and Schuchert's Paleogeography of North

America, already referred to, for many instances.
2 Ulrich, E. C, Revision of the Paleozoic Systems, Bull. Geol. Soc. Amer., vol. 22, p. 506,
1911.
Williams, H. S., Bearing of Some New Paleontologic Facts on Nomenclature and Classi-
fication of Sedimentary Formations, Bull. Geol. Soc. Amer., vol. 16, p. 137, 1905.
Smith, G. P., Principles of Paleontologic Correlation, Jour, of Geol., vol. viii, p. 673, 1900.
' White, David, and F. H. Knowlton, Evidences of Paleobotany as to Geological Climate,
Science, vol. 31, p. 760, 1910.

* WiUis, Bailey, and R. H. Salisbury, Outlines of Geologic History. Maps by Willis, 1910.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 23

currents and the free-swimming embryos of sedentary' forms would be
carried in the same way. The relatively sudden app>earance of a fauna
recognizable as originated in some distant region would at least lead to the
consideration of the possibility of ocean currents setting from the old to the
new locality' and may betray the presence of a most important element in
the paleogeography of the time.

Shore-lines extending parallel to lines of latitude would be more apt to
have similar conditions of temperature, unless currents changed the normal
conditions. The course of the Kuro Siwo brings warm water along the
south side of the Aleutian Islands and wsum-water forms of the western side
of the Pacific are found far east.

It is ob\-ious that temperature is but one of the influences brought to
bear on a migrating fauna and great similarity of a migrant sedentary fauna
to its parent fauna must imply similarity in conditions other than tempera-
ture. The known physical conditions of one region may then with some
safety' be applied to a second region in which the fauna is known but the
physical conditions are unknown. Ideal conditions for such similarity of
faunae would be found on an east-and-west coast, such as perhaps existed
on the southern shore of the North Atlantic continent.

Terrestrial invertebrate fauna. — ^The wide distribution of midges, ephem-
erids, ants, etc., overtaken in their nuptial flights by violent \N-inds is well
known. They may be blown far beyond their natural range, or caught in
bodies of water in enormous numbers. The accumulation of insects in the
water-laid £ish-beds of Florissant, Colorado, is the record of such a catas-
trophe or series of catastrophes to insect life. It is very probable that
regions have been reached by insects, normally absent, through such forced
migrations, but it would be a faulty conclusion, if great numbers of fossils
were found suddenly introduced into a horizon from which they were
previously absent, that a great life migration had necessarily taken place;
the bodies may have reached these as the dead debris of some violent storm.
The movements of creeping land invertebrates would be far slower and less
liable to accidental acceleration; they would be far less liable to question
if used as evidence of land connections.

Terrestrial vertebrate fauna. — ^The greater mobility of vertebrate animals
renders them far more fit to cope with minor changes in the environment and
makes them at once better and worse indices of surrounding conditions;
the highly developed power of locomotion and the ability to resist changes
of temperature introduces no small factor of uncertainty. Far more elusive
factors than the obvious ones so often cited above may determine the pres-
ence of such fossils far remote from their proper habitat. The migrations of
birds, for instance, are governed by what we call instinct, and a study of
the distribution of the remains of migratory birds would be a most formidable
problem to future paleontologists. Disregarding in large measure all bcU"-

24 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

riers of sea or land, climate or vegetation, the migratory flight takes them
over wide areas of the earth's surface in a general north-and-south direction.
Should any attempt be made in a later geological period to correlate the
swamp deposits of to-day by the presence of the remains of the wild duck or
the blackbird, for instance, it would be necessary to assume geographical
conditions far from those which really exist. Here the time element would
be correct, but all implications of geographical similarity would be utterly
wrong. The changes of plumage which frequently accompany migratory
movements or the changing seasons are superficial and no traces would
remain in the fossil state.

Similarly, but over less distances, some grazing forms move with the
seasons, following the grass, or water, or temperature changes. Carnivorous
forms always follow the herds. Here the implication both of climate and
geographical similarity might hold true whenever the remains were found
in a determinable natural habitat, but, as suggested above, such forms
might readily be swept far beyond their usual limits by flooded rivers, as
when herds of bison were overcome on the Missouri or Mississippi Rivers
and the cadavers swept away. The inference to be drawn from the occur-
rence of such forms is the work of experts.

The nature of the beds and their surroundings is revealed by the con-
tained fauna in large measure — marine, brackish, or fresh water; fluviatile,
swamp, or terrestrial; arid or humid; plains, plateaus, woodland, or forest,
etc. There facts are revealed by the structure of the animals which inhabit
them. The general facts in such an interpretation are easily recognized,
but the final interpretation depends on minute and exact knowledge. The
best treatment of the matter is found in Abel's Paleobiologie and Lull's
Organic Evolution.^

(d) Phylogenetic Relations of the Fauna.

The genetic relations of fossils are of the utmost significance not only in
placing the beds in their proper position in the geological column, but for
an understanding of the position of the fauna. The place of origin of the
fauna, the connecting links which reveal the route of migration, and the
separation of the constituent elements of a mixed fauna depend upon an
understanding of their phylogeny. The mixed Devonian faunas known at
Rochester, New York, or Milwaukee, Wisconsin, could only have been
separated into their constituent parts by this means and the routes of
migration of the different elements and the movemen.ts of the epicontinental
seas of the time traced out.

* Abel, Grundziige der Paleobiologie der Wirbelthiere, Stuttgart, igt2.
Lull, R. S., Organic Evolution, 191 7.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 25

The stages of development of the various phyla represented frequently
reveal differences in the time of geological horizons, otherwise indistinguish-
able. In the Red Beds (Permo-Carboniferous) of Texas and Oklahoma,
a group of vertebrates occurs in a series of deposits indistinguishable from
a series of beds in north central New Mexico carr>'ing very similar verte-
brates, but the difference in the age of the beds is revealed by the stage of
evolution of the tw^o genetically closely related groups.

The peculiarities of a fauna, either aquatic or terrestrial, are revealed
in the structure of the individual; such peculiarities are generally in close
response to the conditions of life. It would require a long treatise to discuss
the subject with any approach to adequacy, but on the solution of a paleo-
geographic problem where so much must be determined by indirect evidence
it is necessary to exert the keenest observation to detect every suggestion.
All the influence of the organic and inorganic world is reflected in the
armor, mimetic adaptations, weapons, feeding adaptations, modes of pro-
gression, etc. Here the works of Abel, Walther (parts i and ii), Lull, and
Grabau (chap. 28) already cited are most useful.

It is not alone in response to the environment that peculiarities appear.
As Beecher has shown, the approaching end of the life of a group is heralded
by changes of a marked character in the constituent individuals, as the
assumption of spines, excrescences, etc., and though he drew his illustration
largely from invertebrates, the same thing can be shown for vertebrates.^
Provincial or cosmopolitan character in a fauna is revealed by its constituent
members. Peculiarities of development shown in minor unique variations,
strongly accentuated peculiarities of structure, or even a dominant character
or direction of variation in the whole fauna, as thinness or thickness of shell,
pauperization, etc., all point to a provincial character. The discovery of
such characters would at once direct attention to the examination of all the
other observable facts concerning the unit to test the suggestion of local,
isolated deposits, as in a region of gulfs or bays, lagoons, and small inland
seas, or, for terrestrial forms, isolated peaks, valleys, patches of woodlands,
oases in a desert, etc.^

A greater community of structure in the fauna with less evidence of
peculiarities not (immediately) explainable by use suggests wide areas of
similar conditions where the friction of readily commingled forms from far-
separated areas tends to maintain the mean of life. Such are the marine

' Beecher, C. E., Origin and Significance of Spines: A Study in Evolution. Amer. Jour.

Sd., vol. \i, 4th series, 1898.
Case, E. C, The Permo-Carboniferous Red Beds of North America and Their Vertebrate

fauna. Carnegie Inst. Wash. Pub. No. 207, p. iii, 1915.
* The isolation of the group may be accomplished by more obscure factors than the purely

geographical, as temperature, pressure, strength of waves, etc

26 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

invertebrate faunas of the Niagara limestone, the saurians of the Jurassic
and Cretaceous, or the very similar mammalian faunae^ of certain stages of
the Tertiary.

(/) Radiation and Depression of Life.

Circumstances of variable kinds have determined the abundance or
paucity of life at irregular intervals of the world history. By radiation is
understood the increase not only in number of individuals, but of varieties,
species, genera, and even large groups, reaching out in all directions to find
unoccupied niches in the scheme of existence where food or protection,
breeding-places or homes, might be enjoyed with the minimum of loss.
Osborn has called this, very aptly, "adaptive radiation." ^

Radiations are of two kinds. The greater radiations of the Classes of
animal life, where each successively — fish, amphibian, reptile, and mammal —
asserted the dominance conferred by superior endowments in organization
and for a time reigned as masters of the world. These radiations, which have
given rise to such terms as age of fishes, age of amphibians, age of reptiles,
age of mammals, were the result of the operation of the law of continuous
improvement in life and has a broad but important bearing on the paleogeo-
graphic problem as it was during the periods of expansion of each group
that the closest response to the environment was developed.

Lesser radiations were governed by more evident factors and the dis-
covery of a unit containing an unusual number of individuals, varieties, or
species of any smaller group of life should direct inquiry into the cause,
and this may well be the key to the geographic conditions of the time.
Such radiations may follow the entrance of a migrant fauna into a new
region where enemies do not exist or are not in sufficient numbers to restrain
the natural increase. Classic examples from our own experience are the
remarkable increase of the English sparrow, the cotton-boll weevil, or the San
Jose scale in America, or the rabbits in Australia. Climatic or surface changes,
followed by the inevitable alteration of the vegetation, might give the ad-
vantage to a small or large group and start it upon a career of supremacy.'

Depressions of life are the exact correlatives of the radiations. Untoward
physical conditions, such as the increasing salinity of the remnant seas of
late Silurian time or the unknown conditions which caused the decrease of
the Pelmatozoa in the Permian, illustrate this point. One has but to think
of the effect of great droughts or blizzards on the plains of Argentina,
Patagonia, North America, and Australia to realize what severe climatic
changes may do. The introduction of enemies, as the trypanosomes of the
sleeping sickness in Africa, or the immigration of dominating forms, may

* Osborn, H. P., Age of Mammals, p. 96 and following.

''Osborn, H. P., The Law of Adaptive Radiation, Amer. Nat., vol. 36, p. 353, 1902, and

Age of Mammals, p. 22.
' Osborn, in the article cited above, gives examples of adaptive radiations in the food habits

and mode of locomotion.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 27

cause the complete or nearly complete exhaustion of a group. It would be
unfortunate, however, if an investigator were to be hasty in his conclusions
that the absence or diminished numbers of individuals or varieties of any
group in an observed portion of a unit implied such a condition. The
absence of fossils within the commonly restricted limits of any exposure of a
unit by no means implies the lack of an abundance of life during that interval
of time. The life of the ocean, fresh water, or land is far from uniformly
distributed, even in places where conditions are seemingly identical, and one
can not doubt that similar irregularities of distribution prevailed in past
time. Recognizing the eminently accidental way in which animal remains,
especially terrestrial forms, become preserv^ed as fossils, a depression of
life should be considered as demonstrated only after the most thorough
search. Moreover, certain t>pes of life may be driven out over large areas
and still exist in favorable localities elsewhere, as when the upper Silurian
fauna, depressed in the northern United States, found a "bay of refuge" in
the Gaspe region of Canada; nor can we doubt that, though no great
number of crinoids have been found in Permian deposits, somewhere, as yet
unobserved, this branch of the animal kingdom maintained the stream of
life to reappear as an important factor in the Mesozoic.

(g) The Interrel.\tions of the Fauna.

No small factor in the solution of a paleogeographic problem is the rela-
tion which each separate group or member of the fauna bears to others.
Parasitism renders some forms entirely dependent on the host, and un-
doubtedly the characters and habits of the host are aflFected by the parasites.
How important this is has been amply demonstrated in modem times, where
diseases carried by parasites have depopulated whole areas. The sleeping-
sickness tr>'panosome has practically extinguished human life in parts of
Africa by causing death or migration; the rinderpest wrought havoc with
the game in Africa; the Texas fever killed whole herds of animals in our
0"mi Southwest; parallels to such extreme cases undoubtedly occurred in
past time and may have been the cause of the extinction or change of many
forms of life, both vertebrate and invertebrate. But it is not only in the
extreme cases that parasitism has had its effect. Groups large and small
have changed their habits and form by the assumption of parasitic habits
which were fatal to neither parasite nor host.

Commensalism plays an important part in the modification of structure
and habits and was equally influential in the past. Some forms can only
live in connection with other forms, as in the body-cavities of other animals
or with them in caves, burrows, holes, etc. The little fish Fierasfer lives
in the branchial chamber of the sea-cucumber, and certain sponges grow
only on the backs and legs of certain crabs; the leptoline Hydractina grows
on the shells of hermit crabs, etc. In these cases one form does not prey

28 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

upon the other, but the two live in very constant association, each conferring
some benefit upon the other. Such associations are not uncommon in
fossils. An interesting paper by J. M. Clarke cites several instances of
both parasitism and commensalism among extinct forms. ^

Other relations than the two above should not be neglected. Prominent
among the effects of interrelation are the adaptations of the carnivorous
and herbivorous fauna to each other. Where two such groups have lived
together for a sufficient time for the relations of one to the other to become
well established, extreme and perfect adaptations of structure are to be
expected — the means of defense by armor, thickened shells, mimicry, con-
cealment, etc., will have reached a notable degree of development, while
the carnivorous forms will exhibit equally extreme adaptations to over-
coming the defense. If the fauna is a new one, or if two faunae have been
recently brought together by wide migration or the transgression of a sea
into an inland basin or another sea, the relations will be more simple. The
measure of perfection in the balance between offense and defense will beyond
doubt give some clue to the length of time the whole fauna has been estab-
lished.

It is less probable that forms will be found which have failed to become
adapted more or less accurately to the physical surroundings, as they have
in most cases experienced no sudden change, but in the case of suddenly
increasing salinity of sea-water or the relatively sudden transgression of a
sea over a lowland such conditions might be discovered. Dacque tells us
that changes in facies, organic or inorganic, are gradual; changes between
beds are sudden.

No study of an extinct fauna would be complete were we to neglect to
balance all conclusions drawn from structure against the presence of the
features indicated. For instance, the food of a vertebrate may commonly
be inferred from the character of the teeth — carnivorous (molluscivorous,
durophagous, conchifragous, etc.), herbivorous, or omnivorous. Rodent
teeth, browsing teeth, grazing teeth, etc., all imply definite feeding habits.
Such observations should be checked by a search for the possible food-supply
if it is believed that the remains occur in the original habitat. Frequently
it is possible to determine by a comparison of the armor and the weapons
of offense which animal has been selected as a prey by definite raptorial
forms. It has been suggested that the growing length of the teeth of the
saber-toothed tiger was correlated with the increasing thickness of the cara-
pace of the glyptodonts and that in the Permian vertebrates of North
America the development of strong tusks in Dimetrodon may have been
associated with the development of armor in many of the amphibians and
smaller reptiles.

* Clarke, J. M., The Beginnings of Dependent Life. N. Y. State Museum Bull. 121, p. 146,
1908.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 29

The possibility of error is clearly illustrated in the occurrence in the mid-
Tertiary of forms (the Ancyclopoda) undoubtedly ungulate in most char-
acters, but with strong clawed feet. Such exceptions to the general rules
are rare, but the exceptions serve to compel caution. The inferences drawn
from one portion of the skeleton of these animals — and fragmentary skeletons
are by far the most common remains of vertebrates — would be at total
variance with those drawn from another part.^

Similar examples could be drawn in large numbers from among the

invertebrates, although the specializations and adaptations are not always

so striking.

(A) Faiwal Elements as Time-Markers.

Obvious enough to the trained stratigrapher, it is still important to warn
other workers of the varying value of fossils as indicators of passing time
and changing conditions. Such forms as the brachiopod Lingula and the
star-fishes are classical examples of genera and groups that have remained
unchanged through long periods of time and are valueless as time-markers
unless the most careful specific and varietal determinations are made.
Other more plastic forms, as the ammonites of the Mesozoic, record in their
rapid changes short intervals of time and rapid and slight changes of en-
vironment. Caution is also necessary in accepting as archaic, types which
recall ancient forms of life. The Tuatara lizard Sphenodon, of New Zealand,
long regarded as an example of a survival from Mesozoic or even late Paleo-
zoic times, is now under suspicion as possibly being very specialized, with
but few very archaic characters.^

(»■) The Flora of the Unit.

Much that has been said concerning the fauna of any unit is equally
true of the flora, and similar checks must be used in interpreting the fossil
remains.

One difference between the animal and plant worlds has been given
great weight in all considerations of paleogeography, that is, the compara-
tive immobility of plants. For this reason they have been considered as
especially good measures of climate and climatic fluctuations. This idea
must not, however, remain unchallenged. Attention has often been brought
to the fact that invertebrates, the most common resource of stratigraphers,
are dispersed by forces entirely independent of their own motile powers.
Eggs and free-swimming embryos of fixed forms are dispersed by currents
of water or air; adult individuals of non-sessile kind are equally readily
swept into new regions; their ultimate extinction or preservation in the new
region is entirely independent of the mode of migration. Even vertebrates,

• Scott, W. B., Land Mammals of the Western Hemisphere, pp. 353 and 383, 1913.
2 Ruedemann, R., The Paleontology of Arrested Evolution. N. Y. State Museum Bull.
No. 196, 1918.

30 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

the most mobile of the all, are not infrequently driven by external forces
into new regions. The migration of these animals is thus far equally passive
with that of plants.

Turning to plants and attempting to summarize the powers effecting
their passive dispersal, we find a large and imposing array. Among others,
spores and light seeds are carried by moving currents of air for great dis-
tances, heavier seeds are floated, or carried in the intestines of migrating
animals, or attached to their bodies, to be dropped at great distances. The
chances of survival are neither greater nor less in the new environments
than are those of transported animals. To assume that the flora of a region
has always reached any particular place by the slow process' of self-seeding
under the influence of a slow-shifting climate would be most erroneous.
On the other hand, it would be equally dangerous to assume that a climatic
change comes slowly or suddenly upon a restricted area and remains to
influence it, and that the plants indicate an area of climatic isolation with
unchanging boundaries. It has been repeatedly shown how climatic changes
advance broadly over wide areas in a slow but irresistible march, and the
flora may advance or disappear by self-seeding or death in an equally gradual
manner. An excellent illustration of a relatively sudden climatic change
has been given by Marais.^ Typical illustrations of more slow and regular
changes are given by Huntington.^

The fact that the flora of any period of geological time is frequently in
advance of the fauna in its evolution is well known; the controversy as to
the upper limits of the Cretaceous hangs entirely upon the evaluation of the
floral and faunal evidence.^ Other cases of a similar kind are well known.
May it not be that slowly advancing climatic changes have permitted
plants to advance into new regions, where the greater adaptability of the
vertebrate forms permitted older types to persist for a long time in slightly
changed conditions?

The interpretation of the adaptations of plants to environment is in the
hands of the botanists, but one example of possible confusion may be cited:
Desert plants are protected by a heavy layer of thickened peripheral cells —
palisade cells — and the stomata are set in deep grooves, protected by hairs
or wax, or practically closed. These are adaptations to prevent rapid
evaporation, but similar histologic conditions are found in some plants of
stagnant swamps. It is suggested that this is to prevent rapid evaporation
and so prevent the plant from absorbing too large a quantity of poisonous

' Marais, E. N., Notes on Some Effects of Extreme Drought in Waterberg, South Africa,

Agricultural Journal of South Africa, February 1914. Reprinted in Annual Report

Secretary Smithsonian Institution for 1914, p. 511.
' Huntington, E., The Pulse of Asia.
Some Characteristics of the Glacial Period in Non-Glacial Regions. Bull. Geol. Soc.

Amer., vol. 18, p. 351, 1907.
' A Symposium upon the Cretaceous-Tertiary Boundary Line. Papers by Osborn, Knowlton,

Stanton, Brown, Matthew, in Bull. Geol. Soc. Amer., vol. 25, pp. 321-402, 1914.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 31

water. Succulent plants occur in deserts, but also in purely aquatic habitat,
and when only an impression of the fossil form is preserved the interpretation
is difficult or impossible. The best general-source books are Schimper's
Plant Geography and Clement's Plant Succession.^

Spalding has drawn attention to the fact that a desert environment is a
most complex conception. Ranging from very moist along stream-courses,
where wdllows and arrow-leaves may abound, through less damp soil to
true desert and the high, dry debris slopes of neighboring mountains, with
cactus, greasewood, chaparral, and sagebrush. There is but one common
factor — the hot, dry winds and the intense insolation. All plants of arid
or semiarid regions, even in the damp parts, have coriaceous, heavy, or
otherv\'ise xerophilous leaf-structure. -

Nathorst, in discussing the vegetation of arctic regions as an index of
climate, quotes remarks by Gotham showing that the wood of Cretaceous
trees found in Spitsbergen possesses more definite rings of growth than
those of equal age in Europe, and considers this as an evidence that the
trees were grown in place, in the region of more accentuated climate, and
not drifted in from the south, where the climate was more equable.^

Saporta showed that as general humidity increases the proportion of
monocotyledons increases and of dicotyledons decreases. Lowering the
temperature has the same effect. A dr>^ and warm country has more dicoty-
ledons than a warm and moist or cold and moist country. He also showed
that an abundance of Leguminosae suggest warm and dry conditions, and
that under the same conditions there will be a feeble development of ap-
pendicular organs — coriaceous leaves with frequently spiny margins and a
great complication of the nervation. How uncertain the revelation by
plants may be, however, is shown by Schimper, who, in his Java Flora,
demonstrates that xerophilous adaptations of similar character are found in
xerophiles, halophytes, the Java alpine flora, and evergreen woody plants
of colder climates. Obviously the interpretation from any but the Tertiary
plants must be of the most tentative character.

Bailey and Sinnott* have shown a remarkable relation between the form
of leaves and climatic conditions. In their summary they say;

"There is a clearly marked correlation between leaf margins and en\aronment
in the distribution of dicotyledons in the various regions of the earth. Leaves
and leaflets with entire margins are overwhelmingly predominant in the lowland

* Clements. F. E., Plant Succession: An Analysis of the Development of Vegetation, Car-

negie Inst. Wash. Pub. No. 242, 1916.

* Spalding, V. H., Present Problems in Plant Ecology': Problems of Local Distribution in

Arid Regions, Amer. Nat., vol. 43, 1909. Reprinted in Annual Report Secretary Smith-
sonian Institution for 1909, p. 453, 1910.

* Nathorst, A. G., On the Value of the Forest Floras of the Arctic Regions as Evidence of

Geological Climate, Annual Report Secretary Smithsonian Institution for 191 1, p.
335. In this article many illustrations are given of the use of trees as indices of climate.

* Bailey, I. W., and E. W. Sinnott, The Climatic Distribution of Certain Types of Angio-

sperm Leaves, Amer. Jour. Bot., vol. in, p. 23, 1916.

32 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

tropical regions; those with non-entire margins in mesophytic cold temperate
regions."

They show also that entire leaves occur in tropical and subtropical
regions and in frigid regions, while non-entire leaves occur in cold temperate
regions. A study of Tertiary and Cretaceous dicotyledon leaves leads them
to state that this generalization "affords a simple and rapid means of
gauging the general climatic conditions which existed in regions where these
plants flourished."

Another observation has been made which should be kept constantly
in mind — the eff'ect of altitude upon both animal and plant life. This has
been in part discussed above. AH degrees of humidity are found at the
different levels, the factors of temperature and pressure alone decreasing
constantly as the elevation increases, and the character of the flora with
reference to those, aside from the more variable factors, must be determined
as an isolated fact.^

V. CORRELATION OF THE UNIT WITH OTHER BEDS.

The correlation of the beds is a most necessary step in the solution of
the problem and also one of the most difficult in many cases. The criteria
of correlation as given by Ulrich include diastrophic movements; evidence
of sea-filling and tidal flats; by fossils; by lithologic similarity; by prob-
abilities depending on rhythm of movement; by unconformities, overlaps,
and hiatuses.^ To these should be added the tracing of actual continuity.

When the surface is obscured by vegetation or soil the separation of
exposures even for limited distances may lead to error, especially when the
unit to be traced is variable in character, as a delta or flood-plain deposit.
In many places where the beds by reason of aridity and exposure are laid
bare to the eye for miles it is very difficult to follow a deposit because of the
rapidly changing character of its inorganic content. Individual beds in the
Permo-Carboniferous, Triassic, and Jurassic Red Beds, even when exposed
on the face of a naked cliff or on a bare flat, may not be followed for more
than a short distance in many places. When, as in Kansas, Oklahoma, or
Texas, these exposures are interrupted by areas of grassland or soil, the
difficulty is enormously increased. Many of the older deposits in the
eastern United States are of the same character, and any attempt to strictly
correlate separate exposures in a region of heavy soil, grassland, or thick
woods is sure to lead to very questionable results.

Similarly, but in even higher degree, efforts to correlate beds by samples
from drilled or bored wells are open to question. The persistence of similar
characters in a unit over great areas must be established before isolated

* Seward, A. C, Fossil Plants as Tests of Climate. London, 1892.

' Ulrich, E. 0., Revision of the Paleozoic Systems, Bull. Geol. Soc. Amer., vol. 22, p. 394.
(Index in vol. 24, p. 625, 1913.)

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 33

outcrops can be considered as establishing continuity. For well-established
units of known and persistent character and in regions of continuous exposure
the method leads to incontrovertable results.

Correlation by the fossil content is by far the most commonly applicable,
but the method of applying the evidence is still in dispute. For the corre-
lation of closely adjacent exjxjsures the identity' of faunae or florae is an
unquestioned proof of contemporaneity, but the farther the exposures are
separated the more shrewdly the evidence must be questioned. The ques-
tion of contemporaneity lersus homotaxy has been thoroughly discussed,
and one would hesitate to connect far separate areas cis parts of the same
unit by the simple presence of even closely similar fossils. The character
of the fossils must of course be considered ; floating forms readily dispersed
by currents may spread over a large portion of the earth wathin the time of
the deposition of even a thin unit. So much has been claimed for the
graptolites, and Ulrich has shown how relatively rapidly even moUusks may
be dispersed.^

Another phase of the question of correlation by fossil content is the dis-
cussion as to the relative importance of unique or common species. Is
contemporaneity to be judged by the common occurrence of a large prop)or-
tion of similar sp>ecies, "matching species," or by the common occurrence
of a few peculiar species? The pros and cons of this important question
are taken up in paj)ers by Ulrich, Williams, and Grabau.*

Correlation by inorganic contents — similar mineral or lithological fea-
tures— has proven of value in limited and closely connected exposures, but
for areas separated by any considerable interv'al it has been too frequently
shown of no value to carry any significance of contemporaneity, though
the value of such evidence as suggesting similcU" conditions of deposition is
unquestioned.

Correlations of beds by the size of material, content, and depth of
weathering have been applied to the solution of problems in climatic changes
and terrace formation and destruction; also to the age of glacial deposit.
The arguments suggested by the authors cited may be extended to the inter-
pretation of more ancient deposits in favorable cases. For a discussion of
the value of diastrophic changes in correlation the student is referred to the
citation from Ulrich given above.

VI. CLIMATOLOGY OF THE PAST.

The climatological changes of the past most obviously recorded are those
of the great extremes, as in periods of local or regional glaciation where

' Ulrich, E. 0., Revision of the Paleozoic Systems, pp. 295 and 575.
* Ulrich, E. 0., RcNnsion of Paleozoic Systems.

Williams, H. S., Correlation Problems Suggested by the Eastport Quadrangle, Maine,
Bull. Geol. Soc. Amer., vol. 24, p. 337, 1913.

Grabau, A., Principles of Stratigraphy, chap. 32.
4

34 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

accumulations of tillite and other glacial debris of all kinds are generally
readily recognized. That even such extremes have not left indisputable
records is apparent to all who are familiar with the controversial literature
which has grown up around these phenomena.

It has been demonstrated beyond dispute that in various periods of
the earth's history, from Pre-Cambrian to the Pleistocene, there have been
periods of refrigeration and ice accumulation either as local or continental
glaciers. It is equally obvious, however, that up to at least the close of the
Paleozoic conditions prevailed at times which permitted a uniform distribu-
tion of plants and animals over the surface of the earth under uniform condi-
tions.^

The obvious result of such an apparent conflict of evidence for and
against climatic variability in Paleozoic time is to invalidate to some extent
the evidence on either hand or to enormously increase our conception of the
imperfection of the geological record. The latter alternative would lead
us to increase the estimate of past time by an amount sufficient to permit
repeated revolutions of enormous extent and at the same time to postulate
a total loss of any record of such revolutions. The "geological record,"
while imperfect, shows no hiatuses of this order in the Paleozoic era. We
are permitted to accept with some assurance the general notion that the
climate or climates of any period or smaller division of time were influenced
by local (at least in time) conditions and dismiss in large measure from
consideration, as practical problems of the paleogeography of single units,
the broad problems of climatic revolution, except in certain stages where a
world change is demonstrable, as at the close of the Paleozoic and in the
Pleistocene.^

VII. DISTRIBUTION OF THE FAUNA AND FLORA.

(a) Provincial or Cosmopolitan.

The fauna or flora of any unit may be peculiar in a greater or lesser
degree to that unit, or they may be part of a widely distributed whole.
Such isolation or wide distribution may be due to characters inherent in the
animals or plants themselves, or to the character of the inorganic environ-
ment.

{b) Distribution Dependent on the Character of the Biota.

Animals or plants may become widely distributed, due to some peculiar
resistance or adaptability in themselves which permits them to achieve
success in widely different environments, as the rats and mice, the Canada

- - - ■■■ - — — I ■ - . ■ «

* White, David, and F. H. Knowlton, Evidences of Paleobotany as to Geological Climate,

Science, vol. 31, p. 760, 1910.
^ Schuchert, Charles, Climates of Geological Time, Carnegie Inst. Wash. Pub. 192, pt. 11,
chap. XXI, 1914, with bibliography.
Dacqu6, Grundlagen und Methoden der Palaogeographie, chap. x.
Clements, E. F., Plant Succession, Carnegie Inst. Wash. Pub. No. 242, chap. Xii, 1916.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 35

thistle, or the prickly pear cactus of to-day, which have spread into regions
diflfering notably in climate, soil, and altitude. To infer anything in par-
ticular from the location of such forms would be to strike the mark only
very broadly. One is inclined to believe that the Paleozoic brachiopod
A try pa reticularis^ may have been equally hardy. Such widely distributed
forms can not be considered as good indices of local conditions unless they
possess some known character which has determined their distribution.
On the other hand, sparsely distributed forms may reveal much if they are
correctly understood. Sparseness of a given form may be due to either
rapid evolution or to restricted powers of adaptation.

In the first case, forms which are undergoing rapid change may appear
uncommon because of the really limited numbers of individuals referable
to a given species. A classical example is the large series of ammonites in
the Mesozoic. The discovery of but a few individuals or a single species,
or to find them in a single unit, does not necessarily imply that they were
restricted to any given locality by a definite set of conditions; they may
have had a verv' wide range, but have been only locally preserved under
favorable conditions and have disappeared by actual evolution before such
conditions arose in other places. If, however, the forms are not a part
of a rapidh' changing series, but are highly specialized members of a normally
stable group, their value as indices is high.

Forms may be restricted or dispersed by entirely extrinsic forces, though
this may be in part due to the nature of the animal or plant. Such floating
forms as some aquatic plants, pelagic animals, colonies, as of graptolites,
etc., may be closely confined within certain limits of the temperature and
food-supply of currents, but due to the space covered by such currents and
to the shifting of the currents they might come to be widely dispersed over
the earth and occur in a great variety- of deposits of a very different character.
Could we understand them fully they would tell much concerning the peculiar
environment which favored them, but it would be very erroneous to assume
similarity^ of conditions over broad areas in all places where they are found
fossil. The shifting of the Gulf Stream and the polar currents are well
known, and they bear a life peculiar to themselves; but it is obvious that
they drop the remains of the fauna and flora peculiar to themselves among
widely difi^erent assemblages of more fixed forms.

Other less mobile or movable forms found y^-idely dispersed are clearly
indicative of similar conditions over wide areas, as the fauna of Niagara time.

* See also Ruedemann, Rudolf, The Paleontology of Arrested Evolution, N. Y. State Museum
Bull. 196, 1916.

36 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

(d) Migration.

By migration must be understood the gradual movement of living forms
toward a more favorable environment or their extension within such an
environment. Large accumulations of fossils, especially those of one kind
or a limited number of kinds, are commonly due to the relatively sudden
action of some force catastrophic in nature and results. Frequently this
happens when under the stress of conditions such as a severe storm, unusual
conditions of heat or cold, aridity or drought, shifting currents or introduc-
tion of great quantities of sediment into a body of water, the more mobile
forms are suddenly forced into unfavorable surroundings and perish. Such
catastrophes are not migrations, nor are they to be interpreted as revealing
conditions other than accidental. Great accumulations of fossil material
other than of such forms as habitually grow in masses, beds, or reefs, as
corals or bryozoa, shell-beds, or forest accumulations, are to be looked
upon as unnatural and interpreted with much care.

True migrations are movements in mass of a group of animals or plants
and are induced and checked by extrinsic factors. Even in the case of the
periodic movements of birds and grazing animals, or the more sporadic
movements of the lemmings, locusts, etc., the annual or periodical recur-
rence of the instinct of movement is revived by external factors.

In animals the migration may be active or passive. Increasing numbers
may cause a peripheral pressure which will force individuals ever farther
from the original seat of the group until checked by impassable barriers of
some kind.^ Such a migration is generally in one direction and positive
in character, i. e., there is no advance and retreat, as in the "migration" of
birds, grazing animals, etc. It may be relatively sudden, as when a barrier
is removed from in front of a group experiencing strong peripheral pressure,
as when the Isthmus of Panama was formed or the Behring Straits closed,
or when a land barrier between two bodies of water is broken down; or
it may be slow and regular, as when a climatic change converts a plain
into a forest region or vice versa, or when a sea creeps over the land.

Passive migration occurs where forms sedentary in the adult stage are
free in the egg, embryonic, or young stages and are then borne by water or
other currents to new regions. In this case the animals will go wherever the
current goes and will persist and leave traces wherever the conditions are
favorable. The same principle applies to forms free in the adult stage.
This type of migration is the common thing among the marine invertebrates
and is the form Ulrich has in mind when he announces his belief that little
evolution has taken place within the epicontinental seas, but mostly in the
ocean basins when the seas withdrew and the fauna migrated back upon the
land when the sea returned.

* Scott, W. B., The Isthmus of Panama in Its Relation to the Animal Life of North and
South America, Science, vol. 43, p. 113, 1916.
Matthew, W. D., Climate and Evolution, Annals N. Y. Acad. Sci., vol. xxiv, p. 177, 1915.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 37

IMigration in plants is almost entirely passive; the seeds are carried by
purely external agencies for a greater or less distance, and while the move-
ment may be rapid in some cases, as with animals, it is apt to be very slow.
Seward estimates the average amount of movement in forests by self-
seeding as a yard a year, an amount that is practically negligible.

(«) AUTOCHTHONY (ORIGINATING IN PLACE).

The assumption that particular sp>ots are the original home of certain

forms and that they have migrated in certain definite directions has been

made in a large number of cases. This assumption has, of course, placed

the original home of any form or group at the locality where it is found lowest

in the geological series and has traced the migrations by its appearance at

higher levels in successive spots, but the method is open to objection in

many regards. If Ulrich's assumption that evolution of the invertebrate

forms has taken place in the ocean basins is correct, the first appearance

in epicontinental sea deposits is to some extent accidental and the statement

of direct migration is only, after all, a statement of where we know the fauna

to occur at later dates. Such conclusions should be most tentatively stated.

Indeed, if Ujrich is correct, autochthony in observable regions would be

very rare.

(/) Accidental Introduction.

By accidental introduction is meant the sporadic dispersal of indi-
viduals as opposed to the migration of an entire or a large portion of a
fauna or flora. Such sporadic inclusions of unexpected elements may be
the result of transmission of the living individuals or of the body after death.
In the second case the condition \\-ill be revealed by the fact of the presence
of but a few individuals or a single specimen and may be dismissed as
accidental. Herein lies a grave danger. We have seen how bodies of plant
material may be swept b^^ normal streams, by floods, by wind storms, by
high tides from one locality- to another, and the remains preserved far from
their natural habitat. If such occurrences are carelessly treated they may
involve some erroneous conclusions of the first magnitude, especially if the
observer is inclined to give heed to the presence of unique and jjeculiar forms
in correlation rather than to the " matching of sjjecies." Premature assump-
tions regarding the removal of barriers or of migrations may easily arise.

In the first case, living forms may be carried by such accidents and
find a favorable environment in which they will live and multiply to a
remarkable extent. Cases of this form of dispersal across barriers and
speculations as to other possibilities have multiplied to large numbers and
it is unnecessary to repeat them.^ It may be recalled, as instemces in point,
that eggs and seeds are carried in the mud attached to the feet of aquatic

* Matthew, W. D., Climate and Evolution, Annals of N. Y. Acad. Sd., vol. xxiv, especially
pp. 200-204, 1915.

38 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

birds, that storms may raise eggs or individual animals or plants high in
the air and drop them at great distances ; that seeds ingested by birds and
other animals pass unharmed through the digestive tract; that animals
may be carried upon drifting vegetable material across great bodies of water
and that some seeds and nuts endure long immersion in salt water and may
be carried great distances.

(g) Extinction of a Flora or Fauna.

Extinction of a flora or fauna may be caused by a variety of conditions.
It is especially necessary that the student of stratigraphy recognize that
it is frequently caused by entirely organic conditions, as the introduction
of disease or the advent of powerful enemies which either attack and destroy
the victims or preempt their food-supply or natural habitat. These things
are generally effective upon only a portion of the fauna or flora, but may
attack forms so dominant as to apparently alter the whole biota.

It would be totally unwarranted to assume that the sudden disappearance
of the giant reptiles at the close of the Mesozoic or opening of the Tertiary,
of the horse, mastodon, and elephant from North America in the Pleistocene
was due entirely to inorganic changes, as climate, physiography, etc., which
altered radically the conditions of all life. There is no evidence of a com-
petent change in the inorganic world, and similar catastrophes have been
traced to disease among living forms.

(h) Survivals and Precipitate Development.

Untoward conditions permit archaic forms to survive, as when regions
are long free from disturbances of the inorganic conditions and are pro-
tected by barriers from the advent of destructive or competing forms. The
faunas of the continents of Australia and South America are pertinent
examples.

Other cases of long-lived groups, such as the genera Lingula, Atrypa,
and LeptcBna, are apparently due to a peculiar hardihood inherent in the
group and extraordinary powers of adaptation. It is obvious that such
forms are as little adapted to use in determining the environment as they
are in determining stratigraphic units. ^

Precipitate development is supposed to occur in the youth of a group,
but this is not an invariable principle. Exuberant growth, as has been
demonstrated, is caused by many factors, especially those which in some
manner disturb the phylum. Adverse conditions produce such effects at
times — cross-breeding such as might easily occur in regions crowded with
plants or animals, the approaching extinction of a group— all these and others
induce the appearance of new forms. In paleontology, where the determina-

1 Ruedemann, Rudolf, The Paleontology of Arrested Evolution, N. Y. State Museum Bull.
196, 1916.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 39

tion of species is far less easy and definite than in recent biology, the appear-
ance of new forms whose genesis and relationships are obscure must not be
attributed solely to changes of the inorganic world.

(«) Co>fTROL OF Distribution.

To aquatic invertebrates the presence of land would seem an insur-
mountable barrier to all normal expansion, but it must be remembered that
some of the Crustacea usually aquatic have progressed so far toward a
terrestrial life that they endure a surprising lack of water. The common
crayfish (Asiacus iliiviatUis) has gone so far in this direction that it lives
in very arid regions. The author has found active specimens of crayfish
in little rills formed by recent rains upon the driest part of the plains, and
found one vigorous specimen living among some damp rocks where the
merest trickle of water preserved moisture. It is very probable that this
process has been repeated many times in the past.

The majority of invertebrates found fossil are shallow-water forms,
and to these a deep sea is as impassable a barrier as dry land. Shallow
seas were probably far more common and widespread in the Paleozoic than
in later times and the presence of similar members of shallow-water groups
in remote localities is amply sufficient for the assumption of a shallow-water
connection betv\'een the two places.

The term "barrier," however, involves a most complex conception.
The presence of land and deep water are the simplest types of barriers to
migration. Localized conditions of the water may form impassable barriers.
Currents which have so much to do with the distribution of free-swimming
forms and of larvae would prevent the same forms from crossing them in
anything like a direct path. Warm and cold currents would be as efficient
barriers as equally marked extremes of temperature upon the land.

Alexander Agassiz reported an area almost devoid of life in the deep
ocean waters off the west coast of South America. What the cause of this
is we do not know, but for some reason the life has been barred off.

The waters of the Atlantic Ocean off the west coast of northern Africa
are exceptionally saline, due to excessive evaporation. W^e can not doubt
that in the nice adjustment of life this is an efficient barrier to some forms
of life.

For land invertebrates and vertebrates the barriers would be of the
same general kind. It is essential to recognize that it is not only the
physiographic features which must be considered. Mountains, lakes, rivers,
climate, deserts, etc., are important and effectual, but a stretch of grassland
or a deep forest is as effectual to forms accustomed to an opposite type of
habitat. Absence of food-supply is equally efficient, and animals may be
restricted by a physiographic or hydrographic barrier which would be
utterly inefficient in itself through its effect upon the vegetation.

40 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Moreover, animals are restricted by barriers as obscure as bacterial
disease. Many forms of life are actually barred from regions in Africa
where the tse-tse fly carries the trypanosome of sleeping sickness, and other
parasites are almost equally effective. Man himself is only slowly winning
past the barriers of tropical diseases to a vigorous health and growth in the
intertropical regions.

Salt water is an effective barrier to amphibians, as it is fatal to the egg
or adult of almost all forms.

(j) Environment.

Environment is the sum of all the contacts which an organism or a group
of organisms establishes with the forces and matter of its surroundings,
either organic or inorganic. With this the concept of isolation becomes
much more complex. Complete isolation is unthinkable, but partial and
effective isolation may be achieved by the acquisition of certain habits,
certain physiological peculiarities or immunities, certain morphological char-
acters, etc., which remove a form from a given number of contacts or neutral-
ize their effect. Isolation is no longer to be thought of as accomplished
solely by the presence of physical barriers. An individual or group which
has developed immunity from a contagious disease may continue and exist
in a state of isolation from a set of contacts which control the development
of individuals or groups around it; physical peculiarities, habits, armor, etc.,
might have the same effect.

Such a state of isolation may amount to very perfect adaptation to the
environment, and, as the author has suggested,^ may lead to extinction.

Any attempt at an analysis of the environment as thus conceived will
at once lead to its separation into two main groups, the organic and the
inorganic, both of which are susceptible to minute subdivision, and all sub-
divisions will show innumerable instances of a most complex interrelation-
ship. At the same time, the environment may be divided in a tripartite
manner — into those contacts which are favorable to the organism or group,
those which are unfavorable, and those which are neutral or have no effect.
The latter class will inevitably be very small, for so intimate are the inter-
relationships of all the forces and matter which surround any unit that the
alteration of even the seemingly most negligible factor may have a far-
reaching effect upon the whole. In tabular form such an analysis may be set
forth as follows:

Organic, favorable (hospitable). Inorganic, favorable (hospitable)

Organic, unfavorable (inhospitable). Inorganic, unfavorable (inhospitable).

Organic, neutral. Inorganic, neutral.

Organic contacts will be with other organisms, dead or alive. Such con-
tacts are susceptible of almost endless subdivision and classification according

* Case, E. C, Carnegie Inst. Wash. Pub. No. 207, p. 115, 1915.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 41

as the problem is approached from different angles. In a previous paper the
author has suggested a scheme which seemed the best for his purposes •}

Favorable, hospitable. Unfavorable, inhospitable.

Active hospitality. Active inhospitality, antagonism.

Passive hospitahty. Passive inhospitahty.

Inorganic contacts may be classified under the same heads. Such con-
tacts will be with the atmosphere, hydrosphere, and lithosphere; they will
almost invariably take place, as Chamberlin has pointed out, at the surface
of one of these or in a narrow zone where two of the spheres meet.

Contacts with the atmosphere will be both dynamic and static. The
dynamic contacts will be with all the movements of the air which in any
wise condition or effect the movements, life, or distribution of organisms.
Temperature and pressure are commonly static factors, but in so far as either
determines the movement of the air they must be reckoned as dynamic.

The static contacts with the atmosphere are the temperature (constant,
annual average constant, annual average range, average seasonal variation,
etc.; if there is a progressive change in any of these it will be an effective
factor), the pressure, and the constitution (including the water-content or
humidity).

Contacts with the hydrosphere will be essentially the same as with the
atmosphere, as it is a mobile sphere; but there must be added the factor
(largely static) of the change of state as the temperature fluctuates across
the point determining solidification or liquefaction. Movements in ice-
masses would afford the same kind of contacts as with the lithosphere.
The constitution of the water will afford a much greater variety of contacts
than that of the air, because the water contains large and various quantities
of material in solution, while the air is, so far as we know, a mechanical
mixture of gases and the variation of its constituents is within a limited range.

Contacts with the lithosphere are predominantly static. Such qualities
as hardness, texture, chemical and mineral composition, water-content,
position and posture of rock layers and masses, depth of soil, etc., occur at
once as important factors in the environment. Sudden movements in the
lithosphere, as landslides, earthquakes, etc., are far too brief and localized to
affect more than a portion of one generation, but may be effective as a
dynamic factor in the extinction of a local fauna or flora. The slow move-
ments resulting in soil accumulation and denudation are certainly effective
in their influence on the evolution of a group, but are normally so slow that
the condition in any unit of time, even a unit of considerable duration, may
be regarded as fixed and the contacts will be static.

One other classification of the environment is valuable in realizing the
effect upon life. An environment may be monotonous or diversified.

* Case, E. C, CEcological Factors of Evolution, Bull. Wis. Soc. Nat. Hist., vol. 3, n.s., pp.
169-180, 1905.

42 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

A monotonous environment is one where there is little change in the
factors, static or dynamic, which form the group of contacts and into which
no new elements are introduced and none are abstracted. Such an environ-
ment permits close adjustment of forms, but tends toward the perpetuation
of archaic forms. The environment may be at the same time complex,
relatively, and monotonous. In such a case there may be an accumulation
of stresses which would cause rapid expansion and evolution of life when
the status was disturbed. The depths of the sea, a desert, or a great plain
would approximate this condition.

A diversified environment contains many shifting factors which con-
tinually introduce new elements into the problem, facilitating or inducing
rapid and radical changes. A region undergoing climatic change, inunda-
tion by the sea, or elevation resulting in greater aridity, the spread or retreat
of vegetation, etc., are illustrations.

With this very brief statement of the very comprehensive conception
of the environment, it is obvious that almost every force or kind of matter
must be reckoned with as a possible agent in the development of any group
of organisms. Because of this complexity and the necessity for the con-
sideration of combinations of factors, especially in the study of extinct
forms of life, usually treated separately and only by specialists in widely
divergent fields, it is obvious that the environment is the dominant element
in any paleogeographic problem.

VIII. CHECKS ON THE GEOLOGIST.

It is obvious from the above that the solution of a paleogeographic prob-
lem involves far more than the discernment of the boundaries of deposits or
the mapping of the occurrences of peculiar forms of life. Nor can the pecul-
iarities of a fauna or flora be explained by the relation of the individuals of
the biota to the inorganic environments alone.

The geologist who would restore the condition of the earth at any definite
interval of time may not limit himself to the interaction and results of
inorganic forces, for his restoration would be incomplete and far from
accurate. Even if he designedly deals with such forces alone and desires to
present only the incomplete picture, he is helpless to delimit the land and
water areas without using indices supplied by the response of organic things.
He is not entirely justified in his criticism of the biologist who would raise
a continent to transfer a toad from one side of the sea to another, for after
all the things are there and their presence must be explained, though the
biologist may have been too enthusiastic in his epeirogenic efforts and too
ready to refer the distribution to geological agencies.

The geologist may be far wrong in his interpretations of structures
unless his knowledge of life is ample. The author again calls to notice
his experience on an area of wind-blown sand in a desert portion of Arizona,

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 43

where he found ripple-marks, thin-leaved stalks of vegetation, obscure
insect tracks, and a series of sinuous convolute markings where some insect
burrowing beneath the burning sand had thrown up a long trail indistin-
guishable from worm tracks at the bottom of a shallow body of water. He
went over much of the area most carefully, certainly over far more than is
normally exposed in a geological outcrop, and utterly failed to find a single
criterion that would have prevented him from pronouncing the exposure
an old sea-bottom or flood-plain if it had been found fossil, and yet the
formation was going on before his very eyes on a sun-stricken bit of desert.
The insect burrows would have unhesitatingly been called worm tracks;
the insect tracks might have been made by any one of many aquatic forms
instead of beetles or grasshoppers; the vegetation once fallen and recorded
only as an imprint could not be told from a bit of aquatic vegetation. The
wind ripples upon most careful analysis might have revealed their origin,
but again, in the author's experience, sand collected in a delta deposit
has been pronounced dunesand which had drifted into the water.

IX. CHECKS ON THE BIOLOGIST.

Perhaps the greatest need by any worker in paleobiology is a compre-
hension of the nature of the movement of the land-masses. It is accepted
by the majority of geologists that certain portions of the earth's surface
have been dominantly land and elevated above the general level and that
others have been peristently depressed and occupied by oceanic waters, but
it is obvious to all that the lands have very frequently been covered by
shallow seas and that portions of the present ocean basin were once dry land.
That there has been law in the development of the present shape of the
continents there can be little doubt, but this law is yet to be discovered and
stated. The attention of paleobiologists is especially invited to the section

following.

(a) Brtoges ano) Barriers.

The term "bridges" must be understood to include all possible means of
normal voluntary' movement by living forms of any kind between distinct
areas. Commonly we think of land connections betw^een bodies of land,
but in the proper use of the term it must be applied to other conditions —
channels between bodies of water, zones of climate, zones of equal altitude,
zones of similar vegetation, etc. — anything which will permit migration or
interchange of life. The term "barriers" must be given the same free inter-
pretation . A " barrier ' ' to some things will obviously be a " bridge ' ' to others
in many cases.

The conflict between the paleobiologist and geologist, at least in the
present stage of both sciences, is largely confined to major questions of
communication, or connection between large masses of land and water.
The evidence of such "bridges" and "barriers" has largely been brought

44 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

forward by the paleobiologists, in the presence of common fossils or the
absence of groups in definite localities, but in some cases, at least, the
physical geologists have given equally important evidence. The continent
which occupied the North Atlantic Ocean basin is vouched for by both and
its presence up to at least mid-Miocene fully accepted. The Mediterranean
Tethys is equally well established. Gondwana Land and the Antarctic
connection between Africa, Australia, and South America depend more
definitely upon biological evidence and await full confirmation.

For a discussion of the general principles of the subject we may follow
the majority of geologists in accepting the permanence of the great conti-
nental blocks and ocean basins and so dismiss for the time all questions of
the movement of life forms upon the blocks or within the basins. Only
the cases where "bridges" have been suggested as existing across permanent
ocean basins need be mentioned.

The condition of the surface of the earth before Mesozoic time is, to say
the least, uncertain. The tetrahedral theory suggests a practical reversal
of the land and water conditions;^ in the Paleozoic the bulk of the land lay
in the southern hemisphere; after the Paleozoic it lay in the northern,
but the great Mediterranean Tethys lay always in an approximately equa-
torial position. If it shall ever be fully demonstrated that this reversal
took place, we shall have a rational explanation for the proposition that
land life of the Paleozoic developed largely in the southern hemisphere and
migrated northward, while in the Mesozoic and Cenozoic the land life
reached its maximum in the north and pressed southward. Somewhere
must be found the bridge across the Tethys. Perhaps the edges of the tetra-
hedroid, which would be in the same position before and after the reversal,
will reveal the clue. The deepening of the ocean basin in the Mesozoic,
suggested by Walther, and the contraction and elevation of the continental
blocks, suggested by Wegener,^ account for the present wide separation of
the land by uncrossable barriers of deep sea; but these are but develop-
ments of previous conditions and it is not proven that the deepening of the
basins took place in the Mesozoic.

The uncertainty as to the mode of origin of geosynclines and their later
elevation into mountains of sedimentary rock introduces another element of
uncertainty. Haug believes that such geosynclines lie between great conti-
nental masses and would use their presence as an evidence for the former
existence of a continent in the Pacific Ocean basin ; others contend only for
their presence upon the edges of continental blocks. However this may be,
such geosynclines to-day, so far as we can now determine, lie upon the edges
of continental blocks and are the origin of bordering mountain ranges,

* Gregory, J. W., The Plan of the Earth and Its Causes, Geog. Jour., vol. xiii, p. 225, 1899.

Reprint in Annual Report Secretary Smithsonian Institution for 1898, p. 363.

* Dacqu6, E., Griindlagen und Methoden der Palaogeographie, chaps, iv, v, and vi.

THE ELEMENTS OF A PALEOGEOGRAPHIC PROBLEM 45

however the upheaval may have been accomplished. There is here a hint
which may guide the paleobiologist in postulating land connections. Such
ridges as are necessary should be drawn or searched for parallel to old lands
or upon their edges. The old edges of the North Atlantic Continent are
still traceable in the seacoast of Great Britain, France, and eastern Canada.
Such evidence does not appear for Gondwana Land.

In examining some paleogeographic maps, as those drawn by Scharf,
we find long bridges drawn parallel to the coast, as from the middle of
western South America to the Galapagos Islands and north. Such sugges-
tions seem at first to exceed the possibilities, as the geologist knows them,
but there is a possible explanation in the conceptions of Wegener that the
continents have decreased in size by constant contraction and elevation,
and such ridges may have existed and parts been left behind as the major
portion receded inward. Again, it is possible that previously exposed areas
parallel to existing continental blocks have been obliterated in whole or in
part by the "suboceanic shove" discussed by Ulrich, which he would
demonstrate by the inward position of repeated uplifts in the geosynclinal
region of the Appalachian Mountains. Opposed to such a conception is the
idea of continental creep elaborated by Chamberlin ; but as one force works
outward from above and the other inward from below there would be a
constant, if intermittent, movement dowTi and in at the edge of the con-
tinental block which would not preclude the existence, temporarily at least,
of parallel lands bordering the present blocks. Such conceptions are far
more feasible than ridges flung boldly across what we know to have been
permanent ocean basins.

Such broad questions are, however, only to be hinted at in a summary
outline. Only extended consideration will permit the true weighing of the
evidence. The author has found Dacque, "Grundlagen and Methoden der
Palaogeographie," and Grabau, " Principles of Stratigraphy," excellent intro-
ductions to the literature of this subject.

CHAPTER II.

SUMMARY DESCRIPTION OF THE DIFFERENT PROVINCES OF
NORTH AMERICA IN LATE PALEOZOIC TIME.

The primary attempt in this part of the work is to isolate as definitely as
may be a distinct interv al of time and give such a description of the deposits
included in that interval that the life and the various factors of the environ-
ment, organic and inorganic, which have influenced the life may be studied.

It is realized at the very outset that such an attempt is destined to only
partial success, for the nature of the geological record is in many places
such as to render the determination of the limits of the interval uncertain.
In places the interval began in a time of terrestrial deposition and ended
with the surface of the earth raised above the possibility of any accumulation
and with the geological record exposed to the obliterating forces of erosion and
the obscuration attendant on later earth-movements. In some other places
the limits are equally uncertain, because of other unfortunate conditions.

Under the very uniform conditions of climate which, despite local abnor-
malities, prevailed over a large part of the earth's surface in the first part, at
least, of the interval, it is necessary to consider very wide areas as units, and
this introduces a new element of uncertainty, for while fairly accurate corre-
lations are possible over limited areas, broader correlations are difficult and
less certain, due to geographical interruptions in the exposure of the deposits.

In the first intention the interval of time proposed for the study of life
was included in that which has been called by authors the Permian or Permo-
Carboniferous, but a ver>^ brief inspection of the stratigraphic data included
in the various papers and reports made it evident that no stratigraphic
limits could be assigned to this accepted interval which would coincide with
the climatological, biological, and erosional evidence. From this arose the
necessity of including in the consideration of the problem a considerable
thickness of strata both above and below the limits originally set and an
effort at correlation of conditions which were perhaps progressive in occur-
rence, rather than absolutely or approximately synchronous.

In any case, or by the application of any methods, the continent of
North America shows three areas at the close of the Paleozoic within which
fairly close approximate correlations may be made by stratigraphic and
biologic evidence and between which correlations can only be made on
climatological and erosional evidence. These areas are:

First, the upper Pennsylvanian and Permo-Carboniferous outcrops of
the eastern half of the United States and Canada, here called the Eastern
Province.

48 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Second, the upper Pennsylvanian and Permo-Carboniferous outcrops of
the central and western portion of the United States east of the Front
Ranges of the Rocky Mountains, defined in Publication 207 of the Carnegie
Institution as the Plains Province.

Third, the upper Pennsylvanian and Permo-Carboniferous outcrops west of
the Front Ranges of the Rocky Mountains, and possibly extending into British
Columbia and Alaska, defined in Publication 207 as the Basin Province.

Between the first and second areas lie the uplift of southern Missouri
and the regions directly north and south of it, which are now largely devoid
of post-Mississippian deposits and probably never had any considerable
amount of such deposits.

Between the second and third areas lie the outcrops of pre-Pennsylvanian
rocks of the crests of the Rocky Mountain Front Ranges. It is not im-
possible that connection was established between these areas either at the
north or the south end, or both ends, of the barrier, but the separation was
sufficient, and sufficiently long sustained, to determine the deposition of very
different material and the location of very different faunae on the two sides,
as is shown elsewhere in this work.

In describing the three provinces, the author has reduced the amount
of descriptive material and the amount of quotations to a minimum con-
sistent with an attempt to place the conditions before the reader. Much
of the material is readily available in the reports of the geological surveys
of the United States, Canada, and the various States, and excellent bibliog-
raphies will introduce the student to a very extensive literature. Older
publications, whose value is largely historical, have been sparingly mentioned
in the discussion, though a large number have been carefully considered for
the many side lights which have borne such an important part in the prepara-
tion of the argument. In general, only the later papers which have sum-
marized the evidence are quoted or discussed. Frequent references have
been made to Publication 207 of the Carnegie Institution, where a part of
the material has been presented, and its repetition seemed needless.

The accompanying correlation tables show the relation of the beds in
the different provinces. It does not purport to be a statement of exact
equivalence, but to show the general relation of the beds within limits
sufficiently exact to support the thesis of this paper — that the conditions of
red-bed deposition appeared at progressively higher levels from east to west.
As is evident, the Pennsylvania and West Virginia sections have been taken
as the standard for the Eastern Province and the Kansas section for the
Western Province. The breaks between various areas render the exact cor-
relation difficult and perhaps impossible, but are not beyond the possibility
of bridging within usable limits. The main breaks in the Eastern Province
are between the New England-Canadian area and the Pennsylvania-West
Virginia area and between the latter and the Illinois- Western Kentucky
area.

Colorado.

Wyoming.

After HesdersoD

After Butters

Up. Wvoming = Up. Wyomiag -

t^fkiBa (In^nde) — Ijrkiiu =

faontain (pars) • CJiugvater
Qnicwuer

Big Horn Mis. Laramie
Chugwater Chugwater

LfOW. Wyoining =
Fomtain (para) =
Badito =
Sangre de Cristo

Low. Wyomiiic —
In^aaHle-

Fountaln =>
Badito =

Sangre de Cristo =
0«uiqr aaadstooe

Bidivh.

ForeOeb.

SMudoi

Sovth
Dakota.

Tiana-Peoos Texas.

MJTiftpjulia

Opeebee

TVnakep

Glass Mountains
Gillian formation

Vidiio

Ord. Mts. and Tieuaty

Shaiter regian

Tran»j*eeos, Texas, and
Pecos Valley. N. M.

Bed beds of
PeeosVidi^

Vidrio

Word

Yellow Is.

Capitan Is.

Sandafcooe and Kini^

Wad

Shale, sandstone and

Leooaid

Caapo-ls. Minaeliiaa

Gaptank

limeatone i
ate and
01 flaesy fimeatone

Rustler h.

Ddaware Is.

Castije gyp-
sum

Tbin-bedded

siaeides

Lower brecei-
ated sone

Tranationbeda

AttaaodCSeai-
eoitabeda

Hoeeo

• C, Omarron; W, WeUington; G, Garrison. Quartermaster, perhaps above the CSmmarron.

New Mexico.

Aiisona- Colorado. Ctah. Idaho.

Soutbem.

Central.

North Central.

SouthveMem.

North and
CentraL

Globe Dis-
trict.

Biriwe Dis-
trict.

Ouray and
Sieo.

Tenmile.

Nartbeastem.

Southeastern.

Pecos Valley red
beds

Ckalile Kypoum

Rustler Is.
Capitan Is.

San Andreas
Ten

Abo

Arroya da AKua

Gym Is.

DeCbeByss.
Moeneoiiie

Globe b.

Naeob.

Coder
R.i<«>

lifaroon

Weber, quarta-
ile*

Park CSty foi^
imUion (Hios-
phoiia forma-
tion)

ParkOty foi^
mationCnios-
pboria forma-
tion)

Chw
EWs

IMawareMt.ls.
Hoeoo

Maedalenals.

Kaibabb. j

Aulroy 1

Hennosa

Wasatch Is.
Weber, shale |

Weber qnaita-
ite=Bingham
quartjdte

Weber quarta-
ite

Xtns

* Weber, quartzite of Tenmile district questionable in poeitioQ.

Correlation Table I. — Eastern and Plains Provinces.

and
A.

Central Texas.

Oklahoma.

Kansas.

Missouri.

Iowa.

Kentucky.

lUinoiB.

Is of
Galley

Absent, or red in north-
ern part of Panhandle

Quartermaster*
Greer

Kiger
Salt Fork

3
08

Undifferentiated
Tarkio Is.

Double Mountain

Woodward

Wellington shale

Is.

Clear Fork

Blaine

1 -

3 2

£ 6

Abilene conglom.
Pearl shale
Herington Is.
Enterprise shale
Luta Is.

3T>-

Wichita

Enid

Winfield Is.
Doyle shale
Fort Riley Is.
Florence flint
Matfield shale
Wreford Is.

Cisco

Ralston (Chandler)

Pawhuska Is.

P
Sspulpa (Hominy)

«
d

_ 1

•Florena shale
Neosho member
Cottonwood la.

Shale and sandstone.
1 coal 3} feet thick.
Union Co., Ky.

New Haven Is.

Shoal Creek Is.
Coal 8

Eskridgc shale
Neva Is.

Elmdale formation
Americas Is.
Admire shales
Emporia Is.
Willard shale
Burlingame Is.

Scranton Shale
Howard Is.
Severy shale
Topeka Is.
Callahan Is.
Deer Creek Is.
Tecumaeh shale
Lecompton Is.
Kanwaka shale

Scranton shale
Kanwaka shale

River channels at
top of the Des

Moines

Coals 13-18, red and
purple shales, sand-
stone and thin lime-
stone = L. Cam-
bridge and Ames.
(Includes Carthage
limestone of Ky.)

AnvileHocks8.= Ma-
honing ss.
Coal 12.

Pink, red and
variegated shale
(local)

CoaI7

1
1

Oread Is.
Lawrence shale
Kickapoo Is.
Le Roy shale

Oread Is.
Weston shale

Stanton formation
Vilas shale
Allen Is.
Lane shale
Tola Is.

Chanute shale
Drum Is.
Cherryvale ahale
Dennis Is.
Galesburg shale
Mound Valley Is.
Ladore shale
Bethany Falls Is.

■s

Stanton form.
Lane shale

Tola Is.
Bethany Falls Is.

Correlation Table II. — Basin Province.

Wyoming.

Montana.

California.

Oregon.

Washington.

Big Horn

Mts.

Yellowstone
Nat. Park.

Owl Creek
Mts.

Western Wy.

Phillipsburg.

Ft. Benton.

Shasta Co.

Klamath Mu.

Sierra
Nevada.

Snoqualmie.

Boundan
Line.

(3hugwater
Embar Is.

Embar Is.

= Park City,
etc.

Ott«r shale

Otter shale
Kibbey ss.

(Pitt forma-
tion) Mc-
Cloud sh.

(Nosoni for-
mation)
McCloud shale
Baird shale

Robinson

(Grizzly

Creek)

Paleozoic re-
sembling
rocks of
California

Peahastin
Hawkins
Eastern achist

Tensleep

Quadrant

Tensleep

Quadrant

Chilliwacl
Hozomeer,
AnarchigU
Attwood 1

Indiana.

Ohio.

Pemuylvania and West

Massachusetts.

Rhode Island.

New Bnmswick.

Prince Edward

—

Virginia.

Island.

Nineveh Is.

Rocks with the Aldrich

-p

Nineveh coal

and Friendsville coal

Claysville Is.

Dunkardf coal, etc.

Prosperity Is.

Parker coal

Tennule coal

■j

Donnely Is.

Cambridge slate

U. Washington Is.

JoUytown coal, etc.

Roxbury conglomerate

Dighton conglomer-; New Glasgow con-

Red sandstone and

3 M. Washin^on la.
i L. Washington Is.
5 Washington coal

ate

glomerate

1. Squantum tillite

Purgatory conglom-

a Little Washington coal and

erate

Merom as. = Inglefield a

5 sandstone

(probably)

Waynesbtirg A & B coal, etc.
Cairaville shale

Unconformity?

Waynesburg coal

Brownstown ss.

2. Dorchester slate

Rhode Island forma-

Shulie formation.

Red and gray sand-

Little Waynesburg cool

tion

(Possibly in part
equal to the New

stone and shale

3 Waynesburg la.

Interral

t
i

i Uniontown sa.
i Uniontown coal
5 Benwoodls.

Glasgow con-
glomerate)

Ditney formation

I Semckley ss.

3 Sewicklcy coal

Cool VIII

3 Fishport (Scwickley) la.
Redstone coal

3. Brookline conglom-
erate

Wamsutta formation
Pondville and Bel-

Somerville formation

Pittsburg sa.

lingham conglom-

Pittsburg coal

erates

Pittsbiu-g U.

U. Pittsburg la.

Interval

Connellsvillc sa.

Little Pittsburg coal
Connellsville ss.
Little Clarksburg coal
Lonaconing coal

Joggins formation

Morgan town as.

Morgantown sa.

J Elk Lick coal

Ames U. J

f Ames Is.

Harlem i =

3 Harlem coal

Round Knob ' i

: 1 Pittsburg red shale

: ' Maynardier coal

Millersburg formation

Barton coal ,

; Saltaburg aa.

Bakerstown coal

Coal VII

[Buffalo 88.

Buffalo ss.

Brush Creek

• L. Cambridge coal
Gallitsin coal
Mahoning Is.

)

1
1

Mahoning ss.

Mahoning ss.
Uffington ah.

British Columbia.

Alaska.

daiy
le.

Phoenix.

Bridge River.

Northwest
B.C.

Pan-handle.

Up. and Cent.
Copper River.

Up. Tanana
Basin.

Mid. and Up.
Yukon.

Lower
Yukon.

Headwaters of 1 ^ bite^ Na-
Gulkana and besna^opper
Susitna Rivers, and Chisana
Rivers.

Heavy la.

Upper (cal-
careous)
portion

■■3

c
ce

a
H

Suslota Is.

Heavy Is.

,-ack= j Brooklyn Is. =
ieen= (Cache Creek)
iist= Uncomfoimity
Dd : FrankUn Is.
1 Gloucester

White Cap Cache Creek
8eries=
Cache Creek

Shale, sand-
stone, lava,
intrusions,
etc.

Lower por-
tion

Nabesnals.

Marine black
shale, thin-
bedded Is.,
volcanics

Nation River
(eriea

Basic lava flows
tuff, tuff con-
glomerate, etc.

Sandstone,
shale. lava,
tuff and in-
truaiona

DIFFERENT PROVINCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 49

The break between the Eastern Province and the Plains Province is
formed by the Missouri Island and between the Plains Province and the
Basin Province by the Rocky Mountain Barrier. Both of these breaks
may be less important from the fact that the beds may be traced around
their northern or southern edges almost or completely to an actual connec-
tion. The northern limits of both the Plains and the Basin Provinces are
not yet known, but it is very possible that both may be traced north of the
United States-Canadian boundary. The break between the Basin Province
deposits and the deposits of the British Columbia-Alaska-Pacific Coast area
is only partially bridged at present, but this is of less significance, as the
deposits of the last-named area are all proven with fair certainty to be below
the level of red-bed deposition and are discussed in this work because of
their bearing up>on the general question and the possible routes of migration.

The correlations given in table i, for the Eastern and the Plains Prov-
inces, show no considerable departure from the published and accepted corre-
lation tables, except possibly in the New England and Canadian regions.

The position of the Carboniferous and Permo-Carboniferous deposits of
the Massachusetts and Rhode Island areas is very uncertain. As shown in
the summar^^ description of the stratigraphy, the Cambridge slate and the
Squantum tillite member of the Roxbur^^ conglomerate are considered to be
post-Pennsylvanian, and the Dighton conglomerate is regarded as of the same
age, but the evidence for this is at best uncertain ; these deposits may be
much earlier. The New Glasgow conglomerate is certainly post-Joggins in
age and in all probability close to the uppermost Massachusetts and Rhode
Island deposits in stratigraphic position. The conglomerate beneath the red
shales and sandstones of Prince Edward Island occupies a similar position.

The position of these four series in the correlation table is therefore only
provisional, and there seems no reason why they might not have been much
lower, for it is very possible that the same disturbances which originated
the deposition of red beds in West Virginia and Pennsylvania might have
caused the glaciation southeast of the Boston Basin and the elevation of the
Cobequid Hills.

A. THE EASTERN PROVINCE.

The coal regions of West Virginia and Pennsylvania may be taken as
the type regions of the province. A list in sequence of the principal layers
of the upper Pennsylvanian and Permo-Carboniferous of these regions is
given in the correlation table opposite page 48. Details of these forma-
tions, additional to those given below and in Publication No. 207 of the
Carnegie Institution of Washington, may readily be found in the excellent
reports of the geological surveys of West Virginia (coal reports), Pennsyl-
vania, and the United States, and in the pages of the bulletin of the
Geological Society of America.^

• Notably Stephenson, J. J., Bull. Geol. See. Amer., vol. 18, pp. 29-178, 1907.

50 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Most of Pennsylvania, western West Virginia, and the adjacent portions
of Ohio and Kentucky were occupied by a basin wherein continuous terres-
trial deposition took place from about middle Conemaugh time on. It is
commonly stated that no marine fossils are found above the Ames limestone,
but I. C. White states that some have been found at slightly higher horizons.
But no doubt it may be accepted that middle Conemaugh time saw the
beginnings of new conditions in the type area of the Eastern Province.

A strong indication of this change of conditions is the presence of heavy
layers of red shales and sandstones, as the Pittsburgh red shales in Penn-
sylvania and the equivalent horizons in West Virginia. It must be recog-
nized that, as I. C. White has so strongly insisted, the change in the sediments
marks a decided change in the environment of life, and though Permo-
Carboniferous (Permian) plant remains do not commonly occur until much
higher horizons, the presence of favorable conditions can be recognized and
the younger flora, and part, at least, of the fauna could now appear, either
by development or migration.

It must be clearly understood that the author proceeds upon the thesis,
right or wrong, that favorable conditions for any type of life or group of
forms must precede the life forms, which follow either as determined by the
direct action of the environment or as permitted by the environment,
evolution being determined by other forces. If this be so, then an interval
of geological time begins when the conditions fitted for the life of that time
appear, not when the first typical fossils of the time appear, which may be
at a somewhat or even considerably later date. From this it follows that
to understand the life conditions of the closing period of the Paleozoic era
it is necessary to start somewhat further back than is ordinarily done.

In order to describe most clearly the conditions of the Eastern Province,
it is desirable to divide it into two subprovinces — a Northeastern Sub-
province, including New England and the Maritime Provinces of Canada,
and a Southern Subprovince, including Pennsylvania, West Virginia, Ohio,
Indiana, Illinois, and Kentucky. To the latter subprovince belongs, per-
haps, Michigan, though deposits of that State have added nothing to the
discussion or solution of the question.

The deposits of these two areas are separated by a considerable interval
occupied by both older and younger rocks, and it is questionable if they
were ever connected, but the similarity of the deposits is such as to permit a
correlation of the conditions, at least, under which they were laid down.
The physiography of the two areas during the time of deposition was so dif-
ferent as to constitute in itself a sufiicient cause for the observed differences in
the deposits. In the Northeastern Subprovince the accumulation took place
in long parallel troughs which were largely if not completely isolated from
each other and which received accretions of material derived from a closely
adjacent source. Moreover, these troughs were in the line of a movement of
earth folding and faulting which was at that time undergoing a constant

DIFFERENT PROVINCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 51

access of intensity as the great Hercynian-Appalachian uplift developed from
the east toward the west. The Southern Subprovince, on the other hand,
was a basin of enormous size, with the bulk of its area far removed from
the source of sedimentary material and subjected to vertical movements
only of a relatively minor character. The folding of the eastern side of the
basin took place at or after the close of the interval under consideration.

I. THE NORTHEASTERN SUBPROVINCE.

The Northeastern Subprovince includes portions of New Brunswick and
Nova Scotia between the Bay of Fundy and Northumberland Strait, Prince
Edward Island, and portions of the United States as far south as Massa-
chusetts and Rhode Island. (Fig. i.)

(a) The Canadian Region.

A summary description of the portion of the subprovince which lies in
Canada has been given by Young :^

"Along the banks of the East River, in the vicinity of New Glasgow.are expos-
ures of a red, coarse conglomerate which has received the name New Glasgow con-
glomerate. This formation is the basal member of a very thick group of strata
which, in a comparatively undisturbed condition, floor the country north and west
of New Glasgow, outcropping along the Nova Scotianand New Brunswick shores
of Northumberland Strait for a distance of about 80 miles (130 km.), and underly-
ing the whole of Prince Edward Island. What have been described as equivalent
measures also occur in the western part of the Joggins section along the Bay
of Fundy coast. The distribution of this group of strata is confined, so far
as known, to the general region lying north of the Cobequid Hills, which
stretch easterly from the Bay of Fundy to not far from New Gla^ow, a
distance of about 100 miles (160 km.). In the portion of Nova Scotia north
of the Cobequid Hills and the adjacent portion of New Brunswick, and in Prince
Edward Island, this thick group of strata, of which the New Gla^ow conglomer-
ate in places forms the base, occurs in four distinct basins or areas. One, the
Prince Edward Island area, occupies the whole of that island and is separated
by the waters of Northumberland Strait from a second which lies on the mainland
fronting Prince Edward Island. The second area stretches westerly to the head
of the Bay of Fundy, lies partly in New Brunswick, partiy in Nova Scotia. It
is separated from the two remaining areas by an anticlinal axis of folding running
eastward from the head of the Bay of Fundy to Northumberland Strjiit and
along which are exposed Carboniferous strata of the age of the Productive Coal
Measures and older. The third area fronts on the Bay of Fundy coast, forms the
western portion of the famous Joggins section, and extends inland along the
north flank of the Cobequid Hills. It is separated from the fourth area by
a.xes of folding along which are exposed older Carboniferous rocks. The fourth
area may be named the New Glasgow area. It stretches from New Glasgow
westward along the north flank of the Cobequids and northward from the foot
of the hills to Northumberland Strait.

"This widely extended and thick group of strata of which, in certain districts,
the New Glasgow conglomerate forms the natural base, appears everywhere to

'Young, G. A., Guide Book No. i, part n, Excursion in Eastern Quebec and the Maritime
Provinces, issued by the Geological Survey, Ottawa, p. 229, 1913.

ENVIRONMENT OF VERTEBRATE LIFE, ETC.

form a considerable series and, in places, even appears conformable with the
Productive Coal Measures. The strata are largely sandstones, and because, in
certain districts, varieties of a red colour predominate, the earliest geological
observers assigned the group, in general, to the Triassic."

75*

70- 6S- 60-

SS'

Sb-

'^\

^^f^2^\

^'^^^^'"^^

\''"

;^"

■^S^nV,/ \ \

80* 75° 70*

6S-

60*

Fig. I. — Map of Northeastern Subprovince, showing the general lie of the late Paleozoic
sedimentary deposits, the present outline of the continental shelf, and the possible outline in late
Paleozoic. The lined areas are pre-Conemaugh (?) in age, the dotted areas are conglomerates
and tillites of Conemaugh (?) or later.

The red rocks of Prince Edward Island and the adjacent portions of the
mainland are now thought to be Permo-Carboniferous, without question.^
The most comprehensive description of the rocks of the island was given
by Ells,'' a portion of which is quoted :

"In New Brunswick a narrow margin of the red sandstones, conglomerates,
and associated shales of the upper series is found at several points along the
shores of the Gulf of St. Lawrence, as far north as Shippegan Island. They are
also well seen in the Tormentine Peninsula, where they pass downward into
underlying gray sandstones, which here are supposed to represent the lowest
portion of the Upper Carboniferous in this direction.

"Near Shediac and along the east coast of New Brunswick, these newer
rocks rest upon gray sandstones and conglomerates which have been regarded
as of millstone-grit age, and the productive Coal Measures have not as yet been
recognized in this part of the province. While the gray beds of the two some-

^ See Carnegie Inst. Wash. Pub. No. 207, p. 86, 1915.

* Ells, R. W., Annual Report Canadian Geological Survey for 1902-3, vol. xv, p. 371.

DIFFERENT PRO^^NCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 53

what widely separated divisions of the Carboniferous rocks present certain points
of similarity, there are some features which render their separation possible.
The sandstones of the upper series can be generally distinguished by being much
softer and less coherent in character than the gray grits and conglomerates of
the millstone-grit series.

"South of Baie Verte, this difference in character can be readily seen on
the road leading across to Aulac. Thus, at the latter place, what is known as
the Aulac Ridge rises near Aulac station on the Intercolonial Railway, and
extends in a northeast direction, in the direction of Pointe de Bute and Tidnish.
The rocks of this ridge are gray grits and quartz-pebble conglomerates, and have
a distinct anticlinal structure.

"About 7 miles south of Baie \'erte the millstone-grit outcrop terminates,
but at Halls Hill, which is about 2 miles further north, a series of gray sand-
stones come in these rocks and have been cut down along the roadway. These
belong to the newer series, and are soon overlaid by the soft red beds which are
so conspicuous along the shores about Baie \'erte, and thence east to Tidnish
and on to Pugsvash in Nova Scotia. In these red beds are bands of conglomerates
in which the pebbles are lai^ely made up of bright red shale, and thin bands of
impure red limestone also occur at several points. The series as a whole is quite
distinct from anything seen in the millstone-grit formation, and precisely re-
sembles the rocks seen along portions of the shore of Prince Exlward Island,
from Cape Egmont to \A'ood Islands, as well as at many other points in that
proWnce. In New Brunswick they are also well exposed at Cape Tormentine,
and along the shores of that peninsula at many places, while at Bayfield Comer
and around Port Elgin they are underlaid by the grayer members of the upper
series, which also show on the road between Shediac and Pointe du Chene.

"These soft red rocks with occasional gray sandstones also appear along
the north side of Nova Scotia in the counties of Cumberland, Colchester, and
Pictou. Here for the most part they overlie directly, in so far as yet knowTi,
rocks of Lower Carboniferous age without the interposition of the millstone-grit
or Productive Coal Measures. This contact appears to be of the nature of an
overlap, since there is no indication of faults between them. It is probable,
therefore, that in this northern portion the true Coal Measures have never been
deposited along this side of Northumberland Strait.

"Further east the rocks of the newer series are exposed along the south
side of Northumberland Strait to a point several miles east of Merigomish Island,
or about 20 miles east of Pictou Harbour. At this place they rest upon sediments
of Silurian and Cambro-Silurian age ^^-ith which are associated granites and
other igneous rocks. E^st of this the red rocks of the upper series cire not exposed,
either along the shores of Nova Scotia proper or on the island of Cape Breton.
There would therefore appear to be a gap of considerable extent in the sequence
of the geological formations in this part of the province.

"The structure of the rocks in Prince Exlward Island indicates the presence
of several lines of anticline which extend across Northumberland Strait from New
Brunswick and Nova Scotia, and traverse the island in a general northeast
direction. * * *

"Apparently the lowest rocks of the island series are dark-red sandstones
with occasioned beds of conglomerate in which pebbles are of soft, bright-red
shale, with irregular beds of impure limestone, generally reddish in colour, but
at several points a gray limestone also occurs. Pebble conglomerates are also

54 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

seen at several places, as at North Cape, and on the shores of Mill River south
of Alberton, the pebbles being of quartz, with, occasionally, pieces of hard meta-
morphic rocks. On the ridge about lo miles north of Wood Island, and on the
road to Cardigan, a deposit of well-rounded pebbles is seen which have evidently
been derived from beds of these conglomerates in the vicinity, and traces of which
can be recognized in place.

"In character most of these red rocks are very similar to the beds seen in
the sections along the Wallace and Waugh Rivers on the north side of the Cobe-
quid Mountains in Nova Scotia. They are sometimes interstratified with beds
of grayish sandstone which are usually thin and irregular, the gray colour appar-
ently due to the elimination of the red colouring matter through the agency of
plant stems which frequently occur in these lowest beds. This character is well
seen at St. Peter Island, near the entrance to Charlottetown Harbour, as well
as on Governor Island near by. Further east similar gray irregular beds are
exposed in parts of the section at Gallas Point.

"On the west coast the nearest approach to this feature was observed on the
shore at Campbelltown, where, underlying the great series of red shales and
sandstones which form the cliff between Big Mimenegash and Wolf Cape, coarse
reddish grits with grayish bands crop out at the base of the bluff. While these
may not be quite so low in the series as some of the lowest beds of Gallas Point,
they apparently indicate the lowest members of the series in this direction.

"These are overlaid by a considerable thickness, probably aggregating several
thousands of feet, of soft red sandstones and shales, occasionally with bands of
impure limestone, which are seen over the greater portion of the surface of the
island. Much of the sandstone is a dark red or red-brown, and these pass up
into red sandstones with shales which continue to the summit of the formation.
Throughout this series there is no very great variety as regards the character of
the rocks themselves, and all may be included in the same general group."

(b) The Joggins Section.

The important section at South Joggins, Nova Scotia, which faces west
from the east shore of Chignecto Bay, has been repeatedly described. The
latest statement by the Canadian geologists arranges the deposits as follows •}

Joggins series :

Late Pennsylvanian,

Shulie formation;
Uplift and renewed erosion.
Middle Pennsylvanian,
Joggins formation.
Early Pennsylvanian,

Boss Point formation;
Disconforrfiity.
Mississippian,

Windsor formation ;
Unconformity.
Cobequid series:

Pre-Mississippian.

* Guide Book No. i, part ii, op. ciL, p. 331, 1913,

DIFFERENT PROVINCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 55

It is only the upper tvro of these that need be considered as being within
the time which would include the evolution of life at the close of the Paleo-
zoic, but they are so clearly a continuum of the lower formations that a clear
understanding of their significance demands a comprehension of the whole
post-Mississippian-Joggins series. The Joggins section reveals the structure
and deposits of the great Cumberland Basin, which extends from the Cobe-
quid Hills on the south to the Minudie Anticlinorum on the north, and
from Chignecto Bay to the Pictou Basin, with only minor interrupting folds.

The Mississippian, Windsor formation, consists of brick-red micaceous
shale (1,024 feet), brick-red sandstone (209 feet), reddish sandstone (36
feet), greenish gray sandstone with comminuted plant remains (156 feet),
and greenish gray lenses of concretionary limestone (88 feet) ; total 1 ,693 feet.

Bell described the Windsor series as follows:^

"The fauna of the marine dolomitic limestones of the Windsor series at the
base of the Joggins section indicates broad, clear-water, shallow, and warm seas.
The succeeding and widely distributed deposits of gypsum were undoubtedly
accumulated in shallow pans of the sea under a subarid and probably warm
climate. The interbedded and overly-ing red shales and marls, barren of life,
with an abundance of mud-cracks and ripple-marks, together with the general
unleached condition expressed by the calcareous concretions and high alkali
content, denote similar climatic conditions and a general retreat of the sea,
followed by estuaries or wholly fresh-water deposition. The enxnronmental
conditions at this time appear to have been especially favourable for the forma-
tion of fresh-water subaqueous delta deposits, adjacent to very shallow seas,
having had, it is thought, the forms of narrow but long basins, situated between
mountain masses that had their origin in Devonian times.

"A complete withdrawal of the sea with consequent relative uplift of the land
prevented further deposition in this area, but possibly an extensive period of
erosion again brought about conditions favourable for flu\-ial deposition early in
Pennsylvanian time — conditions which seemingly persisted to the beginning of
the Permian time, as no truly marine or even estuarine fauna occurs in the Coal
Measures of the Joggins area.

"The sediments of the millstone grit were laid down under more fluvial
conditions, an environment attested by the presence of occasional coal seams,
the increasing importance of dark to black shales, and the lighter coloured, though
still imperfectly leached, sandstones. The interbedded red shales, barren of
fossils, may represent the muds of fluvial flood flats, that subsequently were
oxidized subaerially, while the irregular lenticular beds of concretionary limestone
associated with the grey sandstones apparently add their evidence in favour of
flu\'ial conditions and a warm climate.

"During the early Coal Measures the strata were laid down under more
fluvial and swamp conditions, as expressed by the many thin coal seams, the
predominant dark shales, and the more perfectly leached sandstones.

"In later Coal Measure time there is no evidence for a continued abundance
of water, as the red-shale beds indicate seasons of aridity when all the carbon-

1 Bell, W. G., Summary Report of the Geological Survey Branch of the Department of
Mines for 1911, Ottawa, p. 331, 1912.

56 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

aceous elements were removed through oxidation under subaerial conditions.
There was at this time a return to more arid conditions very similar to those of
the millstone grit."

In a later paper Bell says:^

"The overlying Boss Point rocks are characterized by grey sandstones bearing
abundant drift plant debris, which are like those of the higher beds, i. e., of
Pennsylvanian age, and by the occurrence of basal conglomerates and channels
or lenticular beds of the peculiar limestone conglomerates, referred to above.
These latter, in addition to the nodules of unfossiliferous limestone, contain
pebbles of red sandstone and shale, all of which could have been derived from
the underlying rocks of Windsor age, such as are still seen at Dorchester. . . ."

The Boss Point formation, which is, roughly at least, regarded as
equivalent to the Pottsville, is stated to be —

"Made up of two quite distinct divisions, a lower predominantly red division
and an upper prevailingly grey division. The lower red division consists of
varying proportions of brick-red quartz conglomerates and red argillaceous sand-
stones and shales. The upper division is made up chiefly of greenish grey,
yellow-weathering sandstone, interbedded with brick-red argillaceous shales, and
with subordinate grey shales as well as thin seams of coal or carbonaceous shales,
and thin beds of bituminous fossiliferous limestone. The typical sharp quartz
sandstone of the upper division occurs at Boss Point, which name is accordingly
chosen to designate the formation. In the Joggins section the conglomerates,
aside from the limestone conglomerates already mentioned, are confined to basal
members, but in New Brunswick they are much more prevalent."

According to this author, the Cobequid Mountains were subject to
periodic uplifts in pre-Mississippian times and existed during the Missis-
sippian as highlands or islands in the sea, being subject at this time to active
erosion. At the close of Mississippian there were, apparently, warping
movements parallel to the Appalachian axis somewhere south of the Cobe-
quids which probably aflfected the height of this land and furnished the
material for the Boss Point and Joggins formations.

Bell suggests that the change in sediments was in part due to a climatic
change with increased humidity and more fluviatile conditions, with the
accumulation of flood-plain and delta deposits.

The Joggins formation is separated from the Boss Point, according to
Bell, by 'a disconformity. The red shale and sandstone (2,000 feet) earlier
placed at the top of what is now called the Boss Point, is represented in the
south limit of the anticline by 800 feet of coarse red conglomerate formed by
material derived from the underlying Cobequid group, the red shales of the
north limit being considered as an extension of this erosional and deposi-
tional interval. The various movements are discussed by Bell at some
length in the publication cited, pages 367 and 368.

* Bell, W. G., Summary Report of the Geological Survey Branch of the Department of
Mines for 1912, Ottawa, p. 366, 1914.

DIFFERENT PRO^^NCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 57

Following the red shales and sandstones which initiated the Joggins series
is a considerable thickness of gray shales and an increase in the number of
coal seams —

"A monotonous sequence is quite noticeable of zones of regularly, evenly
bedded shales, thin sandstones, underclays, and coal, in alternation with massive,
uneven beds of cross-bedded sandstone that characteristically channel into the
underlying shale zones. . . . Commonly in association with the coals are thin,
shell-limestones which carrj' abundant Anthracomyas, Spirorhis, and leperditian
ostracods, a fact which may be advanced as an argument in favor of temporary
estuarine invasions, as the fauna is neither a distinctively marine nor a fresh-
water one." ^

It is possible that this same series was deposited south of the Cobequids,
as at Parsboro on the Minas Basin, but the evidence is not conclusive.

The Shulie formation is composed largely of coarse grits and conglomer-
ates, the material of which can be traced to the Cobequid region, with an
increase of the size of the pebbles in that direction. Also there is an ap-
preciable amount of material from the Joggins formation. The beds are
markedly uneven ; showing ripples, or crests and hollows, some of consider-
able size. Drift logs and other vegetation are not uncommon.

The characters cited above show that the Shulie formation, regarded by
Ells as Permo-Carboniferous but placed by Bell in upper Pennsylvanian time,
is almost entirely a subaerial deposit. Bell regards this as due to an elevation
of the Cobequid region at the close of the deposition of the Joggins formation.
It is probably in part the equivalent of the New Glasgow conglomerate.

The whole of the Cumberland Basin was the site of accumulation from
the north, south, and west. In Boss Point time this was probably largely
from the Caledonian upland of New Brunswick, with possibly some contribu-
tion from the Cobequids; in Joggins time it was largely from the Cobequids
as at Styles Brook, 15 miles inland from the Joggins exposure, there are
1 ,000 feet of coarse conglomerate formed of pebbles directly traceable to the
Cobequids. In Shulie time the amount of material from the south is
increased in quantity. In conclusion it is stated —

"(i) That a large proportion of the finer material of the 13,600 feet (Logan's
measurement) of Pennsylvaoian beds of the Joggins section was probably derived
from the pre-Carboniferous highlands to the southwest, west, and northwest;
(2) that the excessive sedimentation in the Cumberland Basin was due to the
establishment of a geosyncline in early Pennsylvanian time and to proximity to
a Cobequid highland to the south; (3) that this Cobequid area was subject to
periodic rejuvenations resulting in renewed activities of erosion; (4) that the
derivation of these terrestrial sediments from the south, west, and northwest
has resulted in an interfingering of synchronous lens-like deposits.

"Furthermore, the establishment of these successive Pennsylvanian periods
of uplift, with their consequent effects on the sedimentation, explains what had

• Bell, U)c. cil., 1912, p. 368, 1914.

58 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

previously been a mystery, viz, how a Cobequid upland of so narrow a breadth,
even of Alpine height, could have furnished any important contribution to the
thousands of feet of Carboniferous sediments." '

The Joggins formation is stated by Moodie to be "very much the same
age as the Linton beds and comes in near the base of the Allegheny River
series." ^ This statement is made on the basis of the report by Bell,^
but this correlation can be regarded only as very provisional. It would
appear more probable that it includes all the upper Pennsylvanian and that
the Shulie is Permo-Carboniferous.

In 1893, Fletcher,* in describing the Permian of the Canadian region,
concluded that the New Glasgow conglomerate was newer than the Coal
Measures and separated from them (the millstone grit) by an unconformity.
The dip of the lower beds is less steep than that of the upper and "gray
sandstone with greenish and reddish tints, dipping 42° to 51°, is overlain by
thick beds of very coarse conglomerate which fills depressions in the lower
beds." The upper conglomerate contains pebbles derived from the upper
Pennsylvanian, millstone grit.

It is impossible to correlate directly the beds of the Northeastern Sub-
province with those of the Southern, but enough has been written concerning
the latter to show that the conditions which influenced the life of the close
of the Paleozoic began as low at least as the middle of the Conemaugh, and
it is possible to show that changes due to similar, if not synchronously
identical, conditions occurred in New England and the Maritime Provinces
of Canada. That the change in regularity, persistence, and color in the
Pennsylvanian and West Virginia beds was due to an elevation to the east
with an alteration in climate can hardly be denied, and the occurrence of
conglomerates, glacial conglomerates, and irregular red beds to the north
and east, with a traceable origin of the material to south and west, points
to the same, or a similar, series of disturbances.

Emerson has recently given a summary of the geology of Massachusetts
and Rhode Island,® in which he states that the Carboniferous of the Boston
basin consists of two series, the Cambridge slate and the Roxbury con-
glomerate.

The Cambridge slate lies unconformably upon the Roxbury conglomer-

1 Bell, loc. cit., 1912, p. 370, 1914.

* Moodie, R. L., The Coal Measures Amphibia of North America, Carnegie Inst. Wash. Pub.

No. 238, p. 19, 1916.
' Bell, W. G., Summary Report Canadian Geological Survey for 1912, p. 360, 1914.

* Fletcher, Hugh, Geological Surveys and Explorations in the Counties of Pictou and

Colchester, Nova Scotia, Annual Report Geological Survey of Canada, new series,
vol. 5, pp. 108-141, 1893.
' Emerson, B. K., Geology of Massachusetts and Rhode Island, Bull. 597, U. S. Geological
Survey, 1917.

DIFFERENT PROVINCES OF NORTH AilERICA IN LATE PALEOZOIC TIME 59

ate; both are folded, faulted, and in places considerably sheared. Meta-
morphism has gone so far that an imperfect cleavage is developed in the
rocks ever>'where in the basin.

In the southern part of the basin, at least, the Roxbury conglomerate is
divisible into three members:

Squantum tillite.
Dorchester slate.
Brookline conglomerate.

The Brookline conglomerate lies upon the Mattapan volcanic complex,
which is in places interstratified with the two lower members. It is from 500
to perhaps 2,000 feet thick and contains some layers or pockets of sandstone
and a few thin lenses of slate.

The Dorchester slate consists of 3,500 feet of slate, shale, and argillite,
with some interbedded sandstone and, at or near the top, 40 feet of greenish
and yellowish quartzite. Here and there occur beds of reworked tuff.
The formation is of a uniform character and appears to have been deposited
in a body of fresh water, possibly a lake at the margin of the ice. According
to Sayles, the Dorchester slate is composed of '' red and purple slates, in part
cross-bedded, interbedded with sandstone, and fine-pebble conglomerate.
The slate is typically rather coarse-grained and consists largely of reworked
volcanic sediments."

The Squantum tillite is made up of conglomerate and tillite with some
interbedded sandstone and slate. It measures in different places from 50
to 600 feet in thickness, but the total thickness is unknown, as the base is
exposed in only one locality. It may be that it is separated from the
Dorchester slate below by an unconformity and it passes into the Cambridge
slate above through 100 feet of transition beds:

"A large part of the Squantum tillite appears to be of glacial conglomerate,
containing striated and facetted pebbles as at Squantum and Hyde Park. * * *
He [Sayles] concludes that the ice probably came from the southeast and that
there were at least three beds of till with two intercalated interglacial beds; a
great piedmont glacier like the Malaspina Glacier must have deposited material
such as is found."

A more detailed account of this important member was given by Sayles.*

"The Age of the Roxbury Series.

"The exact age of the tillite is uncertain. The lithological characters of the
Roxbury series resemble closely those of the Carboniferous and Permian of
the Narragansett and Norfolk Basins. The Roxbury series, which consist of the
Roxbury conglomerate, the Squantum tillite, and the Cambridge slate, is newer
than the Cambrian, as proved by pebbles in it of the granite which cuts the
Cambricm. The Roxbury series lies, without much doubt, on the same granitic

' Sayles, R. W., Bulletin Harvard Museum of Comparative Zoology, vol. Lvi, No. 2, Geo-
logical Series No. x, p. 164, 1914.

60 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

surface of erosion which underlies the Carboniferous in the Narragansett and
Norfolk Basins.

"All that can be said at present is that the tillite is of Permo-Carboniferous
age. The fact that the Permian glaciation was so widespread, and that new
evidence of it is coming in so rapidly, makes it very probable that the tillite is
of Permian age. No fossils of determinative value have been found, although
Burr and Burke did find a fossil tree trunk in the Roxbury conglomerate proper.'

"History of the Tillite.

"A study of the sediments of the Boston Basin gives some idea of the physiog-
raphy of the region, during late Carboniferous or Permian times. The area in
which the sediments were deposited extended far and wide beyond the present
limits of the deposits. That the area of deposition was low relatively to the
surrounding country is certain, but that it was at sea-level is not so easily deter-
mined. Towards the close of deposition the land must have been subsiding as
shown by the thick bed of slate over the tillite. In order for till to be preserved
as a tillite, it must ordinarily be on a surface which is subsiding at or soon after
the time of the retreat of the ice-sheet. * * * Whether the slate above the
tillite is of marine or fresh-water origin it is not possible at present to say. No
clearly marine fossils have been found in it, and so far as this negative evidence
goes it is more probably that this slate is of lacustrine origin. The absence of
fossils, however, does not settle the question. Marine life in the Permian seas
was scarce or wanting altogether in many places, and furthermore, fossils are
not found in the marine clays of Pleistocene age outcrops. If volcanoes were
situated then as now near the continental margins, the sea might not have
been many miles away, for volcanic action was associated with the deposition
of these beds, as shown by melaphyre flows in several places in the basin. Ac-
cording to Bailey Willis (Jour. Geol., vol. 17, 1909, pp. 403-405), land extended
at least 100 miles in a southeasterly direction from Boston and probably much
farther than this. That there was high land to the southeast appears probable
also from a study of the tillite. The evidence so far points to a southeasterly
origin for the ice which formed the tillite. * * *

"The Roxbury conglomerate proper at Atlantic exposes a thickness of about
520 feet. The lowest part shows rather small pebbles averaging about i inch in
diameter. Farther up the pebbles increase in size gradually, while in the transi-
tion beds below the tillite the pebbles are larger, averaging about 4 inches. It
would seem very probable that this gradual increase in the size of the pebbles
heralded the coming ice-sheet by wetter conditions or by a shorter distance from
the source, as the ice drew nearer. If the larger size of the pebbles was due to
more water and greater velocity, the pebbles should be as rounded as formerly,
but if the approach of the ice was the cause of the size, the pebbles should be
more angular as well as larger. The latter appears to be the case.

"Above the Roxbury a sandstone bed was formed, indicating slower stream
action. A bed of conglomerate was then laid down, indicating swifter stream
action. Another sandstone bed was then deposited. At this point a new phe-
nomenon is met with. Above this last-mentioned sandstone comes a conglomer-
atic mass which differs from the Roxbury in having fragments and lenticular
layers of slate. * * *. From a study of this bed I infer that the ice had come

* Burr, H. T., and R. E. Burke, Proceedings Boston Soc. Nat. Hist., vol. 29, pp. 179-184, 1900.

DIFFERENT PROVINCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 61

near when these fragments of clay were deposited. Just above this bed come
about 47 feet of slate and sandstone layers with ripple-mark and some boulderets
from 8 to ID inches in diameter. At this time the ice must have made a temporary
halt or retreat. At least, deeper or slower water conditions prevailed."

Sayles continues his description, arguing for a series of advances and
retreats of the ice.

Opposite page 17 of Bulletin 597, Emerson gives a table of the geological
formations of Massachusetts and Rhode Island, from which the portion
dealing with the upper Carboniferous and Permo-Carboniferous sedimentary
rocks is quoted below:

Berkshire Hills and — Central

Eastern Worcester

Northwestern Massa-

Southern Massa-

Massachu-

County and Merri-

chusetts, including

chusetts and Rhode

setts.

mac Valley.

Boylston.

Island.

Cambridge slate.

Dighton conglomer-

Roxbury conglomer-

ate at north, Purga-

ate.

tory conglomerate

I. Squantum tillite

at south.

member.

Worcester, phyllite

(Pennsylvanian).

Unconformity.

Amherst schist (east !

Dark phyllite.

(?)

Rhode Island forma-

side of Connecticut

Chiastolite schist.

2. Dorchester slate

tion (Pennsylvan-

Valley to Worcester Brimfield

Harvard conglomer-

member.

lan).

County). schist.

ate lentil.

Boylston schist.

Brinfield schist.

Oxford schist.

Wamsutta formation.

Quabin quartzite. Paxton

Oakdale quartzite.

3. Brookline con-

Pondville and Belling-

Erving hornblende ' quartz

Merrimac quartzite.

glomerate member.

ham conglomerates

schist. schist.

(probably the same).

From Emerson w^e have the following comments on the more important
of the various beds shown in the table :^

(Page 66.) The Harvard conglomerate lentil "may be equal in age to the
Squantum."

(Page 72.) The eastern rocks of the Carboniferous area in Massachusetts
are more calcareous than in the western, but "the whole series indicates, when
compared with the more eastern beds described above, that the coal-forming
conditions of the central and eastern parts of the State were disappearing and
that deeper waters existed in the Connecticut region, deep enough for the forma-
tion of limestone and in some places near enough to the shore for the formation
of conglomerate."

(Page 76.) Age of the Worcester, Oakdale and equivalent strata: "In the
Narragansett Basin the coal-bearing Rhode Island formation overlies a series
of coarse-grained strata, largely conglomeratic but including considerable sand-
stone and having at the base a conglomerate which rests unconformably on much
older rocks. The lower formations contain fossil tree trunks, some of which

' Emerson, B. K., Geology of Massachusetts and Rhode Island, U. S. Geological Survey
Bull. 597, 1917.

62 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

belong to the genus Catamites, and the whole series is assigned with little doubt
to the Carboniferous. The similar series in the Worcester district comprises the
Oakdale quartzite below and the Worcester phyllite above. The Worcester
phyllite is Carboniferous, for it contains Lepidodendron and several species of
ferns at the Worcester 'coal mine.' Its substantial equivalence to the Rhode
Island formation is indicated not only by its fossils but by beds of graphitic
anthracite it includes. The lower parts of the series in the two areas also exhibit
many points of resemblance, but in the Narragansett Basin the lower part is
made up chiefly of conglomerate with subordinate sandstone and in the Worcester
district almost wholly of sandstone with only a little conglomerate. It has
generally been maintained that the conglomerates were derived from higher land
lying to the east, and, on the assumption that most of southeastern New England
was once covered by Carboniferous strata and that the rocks of the several basins
were, therefore, originally continuous, this would explain the finer grain of the
Oakdale quartzite lying to the west."

(Page 186.) "After the irruption of the Devonian (?) igneous rocks there
was a long period of quiesence and erosion, during which the region was so greatly
denuded that large areas of those rocks were exposed at the surface and deeply
weathered. Early in Carboniferous time, as nearly as can be determined, another
period of eruptive activity began and lasted, in one form or another, until after
the close of the deposition of the Carboniferous strata."

(Page 51.) "The coarse Dighton conglomerate, spread in great sheets over
the thick coal-bearing shales of the Rhode Island formation in the Narragansett
Basin, presents problems of its own. It is coarser toward the south and the
pebbles of fossiliferous Upper Cambrian quartzite, not known in place, for which
the rock is famous, are also larger and more abundant toward the south. On the
• other hand, pebbles composed of muscovite granite are larger and more abundant
toward the north. To explain such conditions, Mansfield^ assumed the former
existence of mountains of Alpine height on the southeast, which may have been
the source of the floods and glaciers and have supplied the coarse material.
Other mountains on the northwest of the Boston district were assumed as a source
of the muscovite granite, as the nearest known granite of that sort lies in that
direction. It is now known, however, that the muscovite granite northwest of
Boston is younger than the Carboniferous sediments. The Dighton conglomerate
finds its possible equivalent in the conglomerate at Harvard, in the Worcester
district."

(Page 58.) "Correlation and age of the formations: No fossils, except at
one locality a few obscure tree trunks, possibly Cordaites, have been found in the
Roxbury conglomerate and none in the Cambridge slate. The age of the beds
is assumed from what appear to be the most reasonable correlations with the
formations of the Narragansett Basin, on the south. In both basins volcanic
eruptions of similar lavas occurred during the early stages of deposition and
presumably at about the same time. The Roxbury conglomerate is believed to
be equivalent to the formations of the Narragansett Basin as a whole, and if so,
it ranges in age from early Pennsylvanian possibly to Permian."

Emerson believes that the Carboniferous deposits were of continental
formation and that the disconnected areas now forming the several basins

* Mansfield, G. R., The Origin and Structure of the Roxbury Conglomerate, Harvard

College Museum of Comparative Zoology Bulletin, vol. 49, pp. 99-271, 1906.
^ Loc. cil., p. 52.

DIFFERENT PRO^^NCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 63

and troughs were originally continuous over the greater part of southeastern
New England. The outward or northwestern border of the basal conglomer-
ates runs through the Boston Basin, bends southward past Woonsocket, and
thence runs near the west shore of Narragansett Bay. In the region north
and west of this line the basal formation was fine sand, instead of gravel, and
was overlain by fine mud, and the deposits of this kind were probably laid
down in interfluvial plains that were afterw-ard overspread by lagpons
occupied by vegetation. The western margin of the great sheet of de-
posits was somewhere near the east side of the present Connecticut Valley.

A suggestion of the continuation of the Pennsylvanian deposits of the
Boston Basin is found in New Hampshire and Maine.^ The Kittery quartz-
ite is correlated by Katz with the Merrimac quartzite of eastern Worcester
County, Massachusetts, and the Merrimac Valley, and the overlying Casco
group of metamorphosed sediments is probably equivalent to the schists
which lie above the Merrimac quartzite.

In Rhode Island Warren and Powers* have distinguished tuo groups of
Pennsylvanian rocks which they regard as of equal age, the Narragansett
and the Bellingham. The NcU-ragansett consists of four formations, the

Dighton group.
Pawtucket formation.
Wcimsutta red beds.
Pond\'iIle arkose.

The Wamsutta red beds consist of red conglomerate shales and sand-
stones. The Pawtucket formation is "largely shales, sandstones, and some
conglomerates." The age of the Narragansett series has been considered
to be Pottsville-AIlegheny from the paleobotanical evidence. Later dis-
coveries b>- Haynes' of bivalve Crustacea, Estheria sp. and Leaia tricarinata
M. and W., with Cordaiies and Calamites in the Pawtucket suggest Cone-
maugh, but are not definitive.

The Bellingham group consists of lustrous green schists and sheared
conglomerates. "The age of the Bellingham series is supposed to be the
same as that of the Narragansett series. The character of the rock with its
associated amygdaloids places it unquestionably in the Carboniferous."

The uncertainty of the stratigraphic position of the beds in Boston Basin
and adjacent areas, as determined by accepted methods or correlation, is
clearly recognized by the author; for that reason he suggests a test of the
value of correlation by " environmental conditions."

* Katz, Frank J., Stratigraphy in Southwestern Maine and Southeastern New Hampshire,

Professional Paper No. 108, U. S. Geological Surs-ey, p. 165, 1917.

* Warren, Chas. H., and Sidney Powers, Geology of the Diamond Hill-Cumberland District

in Rhode Island-Massachusetts, Bull. Geol. Soc. Amer., vol. 25, p. 447, 1914.

* Haynes, W. P., Science, vol. 37, pp. 191-192, 1913.

64 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

II. THE SOUTHERN SUBPROVINCE.

As stated above, it is impossible to correlate at all exactly the deposits
of the Northeastern Subprovince with those of the Southern Subprovince;
indeed, it is very possible that these two subprovinces may bear very much
the same relation to each other that exists between the Basin and Plains
Provinces. The close similarity in the material and the sequence of deposits
which exists between the Permo-Carboniferous beds in Prince Edward
Island, Nova Scotia-New Brunswick, and Massachusetts-Rhode Island, the
disposition of the beds in local basins between anticlinal elevations, the
location to the east of masses of ancient igneous rock which form the northern
extension of a part, at least, of Appalachia, all seem to indicate the original
separation of these areas of deposition from the Southern Subprovince,
either completely or in large part. The lie of the whole series of basins,
ending with the Narragansett Basin, is such as to suggest a continuation
southward upon the submerged and buried eastern portion of the ancient
Appalachia continent. Such a possibility is entirely in consonance with
the idea of gradual compression of the edges of the continents by suboceanic
shove, as advocated by Ulrich.^ On the other hand, there is a total lack of
paleontological evidence either for or against such a separation.

The recognition of the two areas as distinct provinces must wait for
accumulated evidence, if it is ever to be done. There is much evidence,
however, that both subprovinces were affected by a climatic change during
the same interval of time or during approximately equivalent intervals.

The nature of the record of this change lends support to the suggestion
that the basins of the Northeastern Subprovince are but a portion of a
series which extended farther south, east of the present Piedmont Plateau
and nearer to the area whose elevation initiated the climatic disturbance,
for in these basins the red beds have as their basal members conglomerates
and tillites, while the more abundant red shales and sandstones lie higher
in the series and extend farther to the west. In West Virginia and Pennsyl-
vania it is only the red shale and sandstone which appear in any quantity —
just such a phase of deposition as one would expect if the elevated area
were more distant (to the east) from the aggrading basins.

It is just this evidence of climatic change, though present in slightly
different phases, which serves to bridge the gap in the correlation of the two
subprovinces and also serves to mark the beginning of Permo-Carboniferous
time, as it was not only a major change in itself, but points to a period of
diastrophism, other evidence of which is largely hidden by younger deposits
on the eastern side of Appalachia or did not develop until the close of the
Permo-Carboniferous period.

* Ulrich, E. 0., Revision of the Paleozoic Systems, Bull. Geol. Soc. Amer., vol. 22, pp. 439-
442, 191 1.

DIFFERENT PROVINCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 65

(o) Appearance of Red Beds in Pennsylvania and West Virginia,

The first appearance of red beds in the upper Paleozoic deposits of the
Southern Subprovince is not the conventional line between the Pennsylvanian
and the Permo-Carboniferous. It has for long been drawn at the top of
the Waynesburg, the uppermost bed of the Monongahela series in Pennsyl-
vania and West Virginia, but this location has been contested by I. C. White,
who claims that the alteration in the character of the sediments at a much
lower stratigraphic level indicates a change in climatic and physiographic
conditions which warrant the line being dropped to the level of the Saltsburg
sandstone member of the Conemaugh in West Virginia, approximately the
horizon of the Pittsburgh red shale in western Pennsylvania.

The original location of the line of division at the top of the Monongahela
was determined by the evidence from invertebrate fossils, now somewhat
less significant, owing to later discoveries, and from paleobotanical evidence,
upon both of which David White still maintains the correctness of the
original location.

[He] "is inclined to draw the Westphalien-Stephanien (Mid-Pennsylvanian)
boundary provisionally at or close above the top of the Allegheny, the Mahoning
sandstone being interpreted as showing the beginning of a more pronounced
erogenic movement which seems gradually to have brought about the final
exclusion of the sea." *

In his latest discussion of this point, I. C. White^ says:

"In this connection it should be noted that the writer has for many years
suggested and contended . that the sudden introduction of red sediments into the
Conemaugh series, after their total absence since the close of Mississippian time,
when a long period of erosion supervened, was an event of unusual importance
to geologic history. In fact, so distinctive as to warrant the last chapter of the
Pennsylvanian being closed at that horizon, and the first chapter of the Permian
or Permo-Carboniferous opened with the deposition of the Conemaugh red beds.

"The Permian fauna already described by Case from the horizon 34-40 feet
below the Ames limestone at Pitcairn, Pennsylvania, proves incontestably that
Permian vertebrate life had already arrived in the Appalachian field, just as it
had in the western coal fields at the closing stage of the Illinois Coal Measures,
and hence there can be no valid reason why representatives of Pareiasaurus
may not have been among the arrivals that accompanied the new conditions
producing the Pittsburgh red shales that succeeded the great white sandstone
epoch which began with the Pottsville on top of the Mauch Chunk red beds, and
closed with the deposition of the Mahoning, Buffalo, and Saltsburg sandstones
making up the lower one-third of the Conemaugh series as now delimited. The
marine fauna in the Ames limestone is largely composed of forms common to the
Permian beds, as may be seen from the following list of species identified from
West Virginia localities by Stevenson, Meek, Beede, Price, and others, as com-
piled by Wm. Armstrong Price:

> White, David, in Professional Paper No. 71, U. S. Geological Survey, p. 437, 1912.
'White, L C, Notes on the Paleontology of Braxton and Clay Counties, West Virginia;

Braxton and Clay County Report, Geological Survey of West Virginia, p. 822, 1917.
6

ENVIRONMENT OF VERTEBRATE LIFE, ETC.

[Abbreviations: x, = rare, at one ormore localities; c = common; a = abundant;aa = very abundant.]

Endothyra ? sp x

Serpula ? sp x

Vermes indet. (trails?) x

Crinoidea (plates and stems) a

Rhombopora lepidodendroides? x

Lingula umbonata x

Orbiculoidea missouriensis x

Rhipidomella pecosi x

Derbya crassa a

Derbya robusta x

Chonetes granulifer aa

Productus cora c

Productus semirecticulatus x

Productus pertenuis c

Pustula symmetrica x

Pustula nebraskensis a

Strophalosia sp x

Spirifer cameratus x

Ambocoelia planiconvexa aa

Composita subtilita x

Composita sp x

Solenomya radiata x

Solenomya soleniformis x

Solenomya trapezoides? x

Prothyris elegans c

Solenopsis solenoides x

Eximondia? scutum x

Edmondia ovata var. levis x

Edmondia gibbosa x

Edmondia sp x

Nucula anodontoides a

Nucula ventricosa x

Nucula parva aa

Anthraconeilo taffiana a

Yoldia propinqua x

Yoldia sp x

Leda bellistriata ? c

Leda meekana a

Parallelodon obsoletus c

Aviculipinna americana x

Aviculipinna nebraskensis x

Pseudomonotis hawni x

Myalina subquadrata c

Myalina perniformis x

Schizodus affinis ? x

Schizodus ulrichi ? x

Aviculipecten rectilaterarius c

Acanthopecten carboniferous c

Deltopecten occidentalis a

Pectenoidea (fragments) x

Lima retifera x

AUerisma terminale c

Pleurophorus oblongus? x

Pleurophorus occidentalis a

Pleurophorus cf. obsoletus c

Pleurophorus? sp c

Pleurophorella geinitzi x

Astartella gurleyi x

Astartella concentrica x

Pelecypoda indeterminata (several species) x

Bellerophon crassus var. wewokanus x

Patellostium montfortianum aa

Patellostium kansasensis x

Bucanopsis perlata a

Bucanopsis stevensana? x

Bucanopsis meekiana x

Bucanopsis? sp x

Euphemus carbonarius aa

Pharkidonotus percarinatus x

Pharkidonotus percarinatus var. tricarinatus ... a

Worthenia cf. speciosa a

Worthenia (Orestes) intertexta a

Phanerotrema grayvillense a

Schizostoma catilloides x

Loxonema semicostatum x

Zygopleura plicata a

Zygopleura rugosa x

Zygopleura scitula x

Zygopleura sp x

Soleniscus paludinseformis? x

Bulimorpha chrysalis x

Sphaerodoma ? brevis x

Sphaerodoma ? primigenia x

Sphaerodoma ? primigenia var. intermedia x

Sphaerodoma ? ventricosa x

Aclisina swallovvana x

Aclisina? sp x

Orthonema quadricarinatum x

Orthonema cf. subtaeniatum x

Orthonema ? sp x

Minute, open-spiraled gastropod x

Gastropoda indeterminata (coils) x

Orthoceras sp x

Pseudorthoceras knoxense x

Tainoceras occidentale x

Ostracoda a

Eumalacostracean arthropod fragments x

Boring organism x

"True, this list of Ames limestone fossils contains many that occur in the
Pennsylvanian strata of this and other states, but that is no reason why they
may not have continued to live on into Permian time.

"The Permian or Permo-Carboniferous age of these Conemaugh red beds is
also confirmed by fossil plant remains recently discovered a short distance above
the horizon of the Ames limestone, as related to me in a personal communication
by Mr. David White, chief geologist of the United States Geological Survey, who
states that a species of Callipteris, a genus diagnostic of the Permian beds, has
recently been discovered in the upper half of the Conemaugh. In this connection
it is pertinent to quote here the opinions of the writer concerning the age of these
Conemaugh beds, as published in the reports of the West Virginia Geological
Survey, beginning with his first description of the Conemaugh series as given in
volume II, Coal Report, under date of June 15, 1903, pages 225-227.

DIFFERENT PROVINCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 67

'"The Conemadgh Series.

'"As now limited, it includes all of the strata from the floor of the Pittsburgh
coal down to the top of the Upper Freeport bed, the whole having an average
thickness of 600 feet, though it varies from 400 on the western margin of the
Appalachian field in Ohio to 800 feet near Charleston, West Virginia.

"'The series as thus limited above and below, consists of two widely different
members, lithologically considered, the upper composed largely of soft, red, and
marly shale, the lower of massive, pebbly sandstones. The difference in the
rock type is so marked, and especially in the character of the topography made
by each, that the First Geological Survey of Pennsylvania and Virginia placed
them in two different series, the massive sandstones, at the base of the Conemaugh,
being classed with the underlying Allegheny. That assignment, based primarily
upon difference of rock t>^pe, was more philosophical than the present limitations,
but the fact that no definite boundary (a sandstone always being subject to
sudden and rapid change in both thickness and character) could be assigned to
either the lower limits of the upper one, or the upper limits of the lower one, led
Professors Ste\enson, Lesley, and other Pennsylvania geologists to extend the
limits of the "Lower Barren Measures" of Rogers down to the horizon of the
Upper Freeport Coal, a well-marked and widely persistent stratum. This
arrangement gives definiteness to classification, a great desideratum, but it has
the fault of bringing together rocks of very different type, and hence, while
apparently preferable to the old and indefinite dividing-line between the two
series, is yet not altogether satisfactory. Hence, it is possible that a future and
more detailed study of the series in West \'irginia may reveal some more desirable
di\'iding-plane between the Conemaugh series and the underlying Allegheny than
the present one (Upper Freeport Coal), which will retain all of the desirable features
of the Rogers classification and at the same time relieve it of indefiniteness.

'"Viewed from the standpoint of change in physical conditions, the proper
place for such a dividing-plane between the Conemaugh series and the Allegheny
beds would be the first general appearance of red rocks, near the horizon of the
Bakerstown coal about 100 feet under the Ames or crinoidal limestone horizon.
That a great physical change took place soon after the deposition of the Mahoning
sandstone rocks, the present basal members of the Conemaugh series, must be
conceded, since no red beds whatever are found from the base of the Pottsville
up to the top of the Allegheny, and none worth considering until after the epoch
of the Upper Mahoning sandstone.

'"The sudden appearance or disappearance of red sediments after their
absence from a great thickness of strata is always accompanied by a great change
in life forms, and the present one is no exception. In fact, the invasion of red
sediments succeeding the Mahoning Sandstone epoch of the Conemaugh may
well be considered as the "beginning of the end" of the true Coal Measures,
both from a lithological as well as a biological standpoint, and hence it is possible
that the best classification aside from the conveniences of the geologist, would
leave the Mahoning sandstone in the Coal Measures and place the rest of the
Conemaugh, as well as the Monongahela series above, in the Permo-Carboniferous.
This reference is also confirmed by the character of the fauna and flora, both of
which contain many forms that characterize the Permo-Carboniferous beds of
Kansas and the West, as may be seen in the lists published on a subsequent page
under the detailed description of the principal Conemaugh strata.

"'As already stated, the two types of rock (hard and soft) included in this

68 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

series give rise to two widely distinct varieties of both soil and topography. The
uppermost 400 feet of soft beds, with their included thin limestones, and limy,
red, yellow, and greenish shales, interstratified with two or three rather massive
sandstones, give origin to a beautiful rolling topography often finely adapted to
grazing and agriculture, especially where these beds cover the uplands not deeply
trenched by draining streams. When the hills are high and steep, however, the
red marly shales exhibit a great tendency to landslides, and hence, where such topog-
raphy abounds, grazing rather than agriculture should be the chief occupation for
these Conemaugh soils.

'"A wide band of red marks the crop of this soft portion of the Conemaugh
entirely across the State from the Pennsylvania line on the north to the Big
Sandy River at the Kentucky boundary, 250 miles distant to the southwest.'

"Likewise in volume ll (a). Supplementary Coal Report, West Virginia Geo-
logical Survey, published under date of September 15, 1908, on pages 622 to 624,
inclusive, the writer used the following language in describing the Conemaugh
series :

'"Sediments inherently red make their appearance for the first time in this
series since the close of the Mississippian, with the top of the Mauch Chunk
Red Shale. True, a pink or reddish color in the ferriferous clays, or shales of the
upper portion of the Allegheny Series may sometimes be seen, as near Fort Gay,
on the Big Sandy, and near Coal Grove, above Ironton, Ohio, but these apparent
reds are from oxidation due to weathering, since these sediments were not red
when deposited, and if a bore-hole could be sunk through them a few feet in from
their crops no reds would appear. The genuine red beds of the Conemaugh were
deposited as red muds from an old eroded land surface and are inherently red,
whether at the surface or i ,500 feet below the same, ais is the Pittsburgh red shale
in some portions of Wetzel, Monongalia, and other counties in the center of the
Appalachian basin.

"'The general statements on pages 165, 226, and 227, volume 11, about the
importance of the sudden appearance of red beds after their absence from the
strata for a long period of time, and the possibility that the lowest Conemaugh
reds might mark the dividing-line between important formations, such as the
true Coal Measures and the Permo-Carboniferous, has received strong confirma-
tion during the past year. Dr. Percy E. Raymond, of the Carnegie Museum,
has discovered in these red shales, near Pittsburgh, at 35 feet below the Ames
limestone, an interesting reptilian fauna which is closely related to Permian
types. This fauna, including species of the genera Eryops, Desmatodon, and
Naosaurus, allied closely to what have been regarded as Permian forms in Illinois
and Texas, has been recently figured and described by Professor E. C. Case, in
the Annals of the Carnegie Museum, volume iv, pages 234 to 241, April i, 1908.
It is quite possible that a considerable break in the geologic record occurs at the
close of the great sandstone epoch ending with the Saltsburg horizon just above
Bakerstown coal where the great invasion of red beds begins. Although there is
little or no unconformity in dip at this horizon, there may be a real unconformity
of considerable extent, since the variation in the thickness of the sandstone
deposits at the base of the Conemaugh is very great indeed.

"'In connection with the consideration of these Permian land reptiles dis-
covered at Pitcairn, Pennsylvania, in the Pittsburgh red shales by Dr. Raymond,
it should be mentioned that in 1906 Mr. Ray V. Hennen, assistant geologist,
discovered what appears to be a perfect tibia of a large reptile allied to Pareiasau-

DIFFERENT PROVINCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 69

rus according to Osbom, but who after making a cross-section of the supposed
bone, and finding no bony structure preserved, pronounced it a concretion, the
most remarkable one he had every seen. Many geologists, and other vertebrate
paleontologists who have seen the specimen, declare that its concretionary origin
is not proven, and that it is most probably a sandstone cast, an actual fossil from
which all bony structure and organic material have disappeared before lithifica-
tion in its porous matrix, thus preservang only the outside surface and shape of
the bone to perfection.

'"Mr. Hennen found it lying loose upon the surface, near Salt Lick Bridge,
Braxton County, a few feet above the horizon of the Ames limestone, where it had
evidently weathered out of its original matrix in a greenish, micaceous, fine-
grained sandstone. Of course the testimony of this specimen will remain of
doubtful value until its true nature is determined beyond question by the dis-
covery of other concretions or fossils, as the case may be, in this same region.
In this connection it should also be remembered that Scudder, in his Bulletin
No. 124, U. S. Geological Survey, has described a fossil insect fauna from just
above the Ames limestone near Steubenville, Ohio, in which he finds forms greatly
resembling those in the lower Dyas or Permian of Weissig, Saxony, and hence
it should not be surprising to find Permian reptilian forms in these Conemaugh
red beds, which the writer has for several years insisted were more nearly related
to the Permian than to the Carboniferous proper, and that the introduction of
red sediments after such a long absence marked a change in physical and biological
conditions sufficiently great to warrant a division of the geologic column at or
near the horizon of the Ames limestone. It was formerly suggested that this
division should come just above the Ames horizon, since its deposition marked
the end of marine life in the Appalachian field, but the discoveries of Dr. Raymond
of a Permian reptilian fauna at several feet below the Ames limestone would tend
to show that this division-line should be drawn at the base of the Pittsburgh red
shale, about 100 feet below the Ames horizon, or the top of the Saltsburg sandstone.

'"As these first red deposits were probably laid down upon an eroded land
surface, the great irregularity of their thickness (which varies from 10 to 200 feet)
below the Ames limestone would be thus readily explained.'

"The peculiar type of fossil insects referred to above are described by the
late Professor Samuel Scudder in Bulletin 124, U. S. Geological Survey, and on
page 12 of the same he gives his reasons for regarding not only those found in
the Cassville plant shale as above the horizon of the Pennsylvanian Coal Meas-
ures, but also those found near the Ames limestone near Steubenville, Ohio, in
the following language:

'"The West Virginia locality is at Cassville, Monongalia County, not far
from Morgantown, and the specimens were found in rocks lying above the
Waynesburg coal, in what is termed by Professor I. C. White the Dunkard Creek
series, and referred very positively by him and Professor William M. Fontaine
to the Permian. The blattarian fauna as thus far determined is unquestionably
younger than any known from the Pennsylvania or Illinois rocks, on which we have
hitherto depended largely for our knowledge, and consists of a vast assemblage
of forms, which will undoubtedly be increased by further search. They number
56 species, belonging to 5 genera, the bulk of them (36 species) to Eioblattina.

'"The Ohio locality lies at the edge of the township of Richmond, on Willis
Creek, in the near neighborhood of Steubenville, Jefferson County, and though
far less extensive and less thoroughly worked than Cassville, has already yielded

70 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

22 species belonging to 3 genera, of which the larger number (17) belong to
Etohlattina and others to the genera represented at Cassville by more than a
single species.

" ' It is a curious fact, to which I called partial attention when first describing
some of them, that these species represent for the most part a distinct group of
cockroaches of the genera Etohlattina and Gerablattina, characterized by great
length and slenderness of the internomedian area, by a remarkable openness of
the neuration in the middle of the tegmina, and by their frequently exceptional
length and slenderness. They comprise, indeed, nearly 75 per cent of the species
of these two genera at Richmond, and hardly occur elsewhere excepting at Cass-
ville, where they compose about 25 per cent of the species of these two genera.
The only occurrence of a similar form in Europe is Etohlattina elongata from the
lower Dyas of Weissig, Saxony. The occurrence of this type of cockroaches is
the characteristic feature of Richmond, and must place this fauna high in the
series, as the stratigraphical evidence itself warrants. Its horizon, according to
Mr. Huston, who alone has explored the location, is in the barren Coal Measures,
a little above the Crinoidal limestone.

"*It is remarkable that, notwithstanding the close relationship in general
features of the two rich faunas of Cassville and Richmond, not a single species
has been found common to the two. One species, indeed, I formerly regarded
as found in both, but a closer study convinces me that there are in this case two
nearly allied forms, and they are accordingly separated in this paper. Further
than this, with one or two exceptions, no American species has been found in two
different places, and without exception the American species are completely
distinct from the European.'

"Hence we see that not only the reptilian life, but also the insect and plant
life of the Conemaugh, supports the conclusion that the beginning of red sedi-
ments in the Conemaugh Series marks the dawn of Permian time, while there
is nothing in the marine life of the epoch to contradict the same when properly
interpreted. The presence of the peculiar type of Odontopteris, like Odontopteris
(Lescurites) moorii, in the horizon 20 feet below the great Pittsburgh coal, near
Wheeling, West Virginia, as identified by Fontaine, and also in the roof shales
of the same near Greensburg, Pennsylvania, also confirms the very late age of
the Monongahela series and would thus support the conclusion that the base of
the Rothliegende or Dyas should be brought down from the top of the Waynesburg
coal to near the base of the Conemaugh series or to the zone of the first appearance
of red sediments in that series where there appears to be a true unconformity,
or rather disconformity."

As early as 1880 the significance of the changes shown in the Monongahela
deposits was recognized by Fontaine and White, ^ who say in their report
on the Permian flora:

"We may next inquire whether we have evidence of any considerable change
which would suffice to produce an important effect, and alter the conditions
which prevailed in the lower beds, which all recognize as of Carboniferous age.
For this purpose we must turn to the general geology of the district. From this
we find, after ascending above the Pittsburgh coal and its associated coals, the

* Fontaine, W. M., and I. C. White, The Permian or Upper Carboniferous Flora of West
Virginia and Southwestern Pennsylvania, Second Geological Survey of Pennsylvania,
Report of Progress PP, p. 117, 1880.

DIFFERENT PROVINCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 71

Redstone and Sewickley, two horizons which give evidence of extensive physical
changes.

"The first of these horizons marks the general submergence which produced
the important limestones and calcareous shales which occupy much of the interval
between the Sewickley and the Waynesburg. We find no plants until we reach
the roof shales of the last-named coal. These shales, as we see from our analysis
of the table, contain nearly all the characteristic Carboniferous plants which
pass into the Upper Barrens, mixed with a great number of new forms. The
physical change here was not sufficient to entirely alter the flora.

"The second horizon of changing conditions is found in, and immediately
above, the Waynesburg coal. In the rapid fluctuations in thickness of the clay
parting of this coal we see the first indications of unquiet, and of the approach of
that much greater disturbance which produced the important Waynesburg
sandstone, which in its extent and character gives ample evidence of widespread
change.

"The W^aynesburg sandstone often rivcils the great Conglomerate sandstone,
which forms the base of the Productive Coal Measures in the amount of the
pebbles it contains. It is often 75 feet thick, and in expanse is coextensive
with the Upper Barrens. To form an idea, however, of the amount of change
required to produce this great mass, we must not simply consider the character
of the stratum per se, but must contrast it with the strata which immediately
precede it. Leaving out of \-iew the Waynesburg coal, all the rocks for a con-
siderable distance under it are either limestones or fine-graned shales, which
show that the deposition of sediment must have taken place under conditions of
general quiet. The shale roof of the Waynesburg coal is not always present. We
sometimes find the sandstone lying immediately on the coal, and even descending
into it.

"When, then, in such localities we see the immense sandstone loaded with
pebbles lying immediately upon the coal with its subjacent fine-grained beds, we
are forcibly impressed with the magnitude of the change which has taken place.
The character of the pebbles also is significant. They are not of sandstone, but
of quartz, and hence must have been brought from remote localities.

"Let us now consider what is the evidence from the lithology of the strata
of the Upper Barrens. Leaving out of consideration the finding of a conglomerate
at the base of the series, a feature which it has in common with the Permian of
Europe, we find in it a great deal of red shale, another feature of the Lower
Permian of Europe. These red shales occur in beds 20 feet to 30 feet thick,
sometimes commencing immediately above the Waynesburg sandstone. They
are a pretty constant feature, and are often, as at Bellton, several hundred feet
thick. These features, taken alone, are not entitled to much weight, except as
showing conditions unfavorable for the formation of coal, as they are found in the
barren portions of the Carboniferous formation proper. Besides these character-
istics which mark the Lower Permian of Europe, the Upper Barrens have some in
common with the Zechstein or Upper Permian, in the presence of a large amount
of limestone.

" It is a significant feature that these limestones are devoid of marine fossils,
showing that the sea had access at no time during their formation.

"The evidence from the total disappearance of coal beds in the higher portions
of Upper Barrens, and from the extremely small amount of it found in the lower
portions, is of more value, as indicating a great change from the conditions which
prevailed during the Carboniferous proper. The beds of coal gradually dis-

72 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

appear as we pass upwards, and with the exception of the Washington coal, are
never more than i or 2 feet thick, while the uppermost 200 or 300 feet contain
none at all. This diminution of the coal is accompanied with a great loss in the
amount of plant life."

In the report by Fontaine and White' on the Permian flora of Pennsyl-
vania and West Virginia, there is a statement as to the red beds of the Upper
Barren Measures (Dunkard):

"The upper half of the series is quite variable in the character of its strata.
In some places we find it containing a great deal of massive sandstone, with drab,
argillaceous beds, mainly incoherent shales. At other points, we find on the
same horizon, several hundred feet of red shales, often mottled with green, buff,
or yellow spots, and streaks. Toward the south, the red and variegated shales
increase in thickness and descend lower in the series, sometimes even nearly to
the horizon of the Waynesburg coal. The red shales are quite inconspicuous in
Marshall County, and in the 600 feet of strata shown at Bellton we find about
400 feet of red shales, not in a single bed, but in several beds, from 40 to 60
feet thick, alternating with brown sandstones or drab-colored shales.

"The Waynesburg sandstone, the rock which forms the base of the series,
is an important stratum, since its physical character denotes plainly a great
change in the conditions which had prevailed for a long period previous to the
time of its formation. As has been previously stated, these conditions were
quiet subsidence, and deposition of fine shales, with much limestone. But in
the sandstone now described we find many evidences of strong currents, which
tore up the previously formed coal, and brought in a vast amount of coarse
material. The approach of this unquiet condition of things is indicated in the
structure of the Waynesburg coal itself."

The suggestion made and defended by I. C. White in the articles quoted
above is strengthened by the discovery in these places of vertebrate fossils
closely related to or suggesting the vertebrate fauna of the Texas and
Oklahoma beds:

(i) The discovery of reptiles and amphibians in the Pittsburgh red
shale by Raymond. (2) The discovery of Pareiasaurus (?) henneni W'hite,
200 feet below the Pittsburgh coal and the base of the Monongahela series, in
West Virginia.^ (3) The discovery of an Edaphosaurus spine in the Wash-
ington shales at the base of the Dunkard series, near Elba, Ohio.* This
discovery is from a higher horizon than the others and within the limits of
the Permian as commonly recognized, but at its very base and from red
beds similar in character to those carrying vertebrates in the middle of the

1 Fontaine, W. M., and I. C. White, The Permian or Upper Carboniferous Flora of West

Virginia and Southwestern Pennsylvania, Second Geological Survey of Pennsylvania,

Report of Progress PP, p. 25, 1880.
^ See Carnegie Inst. Wash. Pub. No. 207, p. 84, 1915.
' Case, E. C, Notes on the Possible Evidence of a Pareiasaurus-like Reptile in the Cone-

maugh Series of West Virginia, Braxton and Clay County Report, Geological Survey

of West Virginia, p. 817, 1917.
* Stauffer, C. B., Divisions and Correlations of the Dunkard Series of Ohio, Bull. Geol.

Soc. Amer., vol. 27, p. 88, 191 5.

DIFFERENT PROVINCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 73

Conemaugh series. (4) The reported occurrence of reptiles and amphibians
from the Conemaugh of Ohio was based upon evidence that has not been
verified and seems in itself insufficient.^

According to Scudder, the insects found at Steubenville, Ohio, have a
very decided Permo-Carboniferous (Permian) aspect.

For the reasons advanced by I. C. White and the contributory evidence
of the vertebrate fauna, it seems necessary at the very least to examine
carefully the possibility of faunal equivalence of the middle Conemaugh and
the Permo-Carboniferous of Texas and Oklahoma. Stauffer,^ in the article
cited above, states his belief in the equivalence of the Dunkard of Ohio
and the Wichita formation of Texas.

"In view of this evidence of the vertebrate fossils, there can be no doubt
that the lower portion of the Dunkard series is the equivalent of the lower Texas
beds (Wichita) which overlies the Cisco and that in Jill probability both beds
cire Permian."

There is, however, much doubt that the occurrence of a single spine of
Edcphosaurus is sufficient evidence upon which to base such a conclusion.
A spine tentatively assigned by Case to the same genus was found by Ray-
mond in the Pittsburgh red shale of middle Conemaugh time, which is much
lower than the Texas vertebrate-bearing horizon. It seems to the author
that in view of all facts it is far more probable that the presence of the
vertebrate fauna of Permo-Carboniferous age depends rather upon climatic
and physiographic conditions than upon any single time interval which can
be identified stratigraphically in different parts of the continent. It is well
known that in Pennsylvanian time the continent was gradually rising on
the eastern side and that the conditions which influenced the deposition of
late Pennsylvanian and Permo-Carboniferous beds progressed consistently
to the west. This would produce a series of beds rising obliquely across the
stratigraphic column toward the west and involve a correlation of conditions
by the climatic and faunal elements in no sense synchronous in all places.
This is a thesis which will be defended in another part of this work.

The stratigraphy of the upper Pennsylvanian and Permo-Carboniferous
of Pennsylvania and West \'irginia has been published in detail and need
not be recapitulated. Only those formations which require discussion will
be cited in the course of this summary. The probability of the equivalence
of these beds in the Northeastern and Southern Subprovinces has already
been stated (page 64). David White^ gives the following statement con-
cerning the deposits of the Pennsylvanian in the Southern Subprovince:

"Character of the sediments. — The rock- forming materials are mainly terrig-
enous, brought down by rivers chiefly from eastward lands, which were probably

^ See Carnegie Inst. Wash. Pub. No. 207, p. 80, 1915.

» Stauffer, C. R., Bull. Geol. Soc. .Amer., vol. 27, p. 88, 191 5.

» White, David, in Professional Paper No. 71, U. S. Geological Survey, p. 430, 1912.

74 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

the site of nearly continuous though variable epeirogenic action. Consequently
the formations are in general thicker and more arenaceous toward the east side
of the basin. The greatest thickening is toward the southeast, where at the
edge of the Cretaceous overlap in Alabama, the Pottsville, or lower division,
probably exceeds 7,500 feet. Marine or biackish-water faunas, extending over
wide areas, occur at numerous stages except in the later Pennsylvanian, thus
showing frequent accessibility to marine life, though the conditions of sedimenta-
tion in the Appalachian trough were generally less favorable for open-water
marine mollusca than in the eastern interior arm. The subsidence kept, on the
whole, relatively close pace with loading, so that, though the warping was unequal,
there is slight evidence of the contemporaneous formation of new basins as the
result of the orogenic changes."

The Allegheny series :^

"It embraces the softer sediments of more quiescent waters intervening
between the arenaceous invading Pottsville and the Mahoning and other sand-
stone and shale members of the overlying Conemaugh formation * * *. As
compared with the Pottsville the members of the Allegheny are relatively regular
and continuous, and the occurrence in them of marine mollusca is comparatively
common."

The main strata of the Allegheny beds, according to Orton and I. C.
White, can be traced across Ohio from Columbiana County to Kentucky,
250 miles, beyond the Pennsylvania line. Considering with this the extent
and persistence of the Allegheny series in Pennsylvania and West Virginia,
there is, apparently, a Pennsylvanian base for a Permo-Carboniferous series,
quite similar to the condition which prevails in the western provinces, but
at a considerably lower stratigraphic level.

The Conemaugh, in strong contrast to the Allegheny, has an irregularity
in the beds far exceeding even that of the Pottsville. The different layers
vary in thickness and in many places some are absent, but there are some
which are very persistent. The limits of the Conemaugh have been arbi-
trarily fixed, mostly for convenience in mapping; there is the same shading
from persistent dominantly calcareous beds below into more shaly and irreg-
ular beds above, with a rapid appearance of red beds such as occurs in the
Plains Province and parts of the Basin Province. The Monongahela is,
like the Conemaugh, variable in character, but contains much limestone
and coal. The bulk of the red and green sandstone and shale is in the
southern portion of the Monongahela; towards the north the deposits are
more normal in color.

The Dunkard in its maximum thickness consists of 16 to 18 members,
being alternations of limestone, coal, and sandstone. According to Steven-
son,^ the Dunkard is smaller in extent than the Monongahela and confined
to a limited area in the adjoining portions of Pennsylvania, Ohio, and West

• White, David., loc. cit., p. 434.

' Stevenson, J. J., Carboniferous of the Appalachian Basin, Bull. Geol. Soc. Amer., vol. 18,
p. 160, 1907.

DIFFERENT PROVnNCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 75

Virginia. The area covered was originally much larger, as outliers are
found in Pennsylvania, West Virginia, and Maryland; all evidence of the
original extent to the east has been removed by erosion.

David White^ is quoted by Stevenson to the following effect from a
revision of the original work by I. C. White and Fontaine on the flora of
the Dunkard:

The fauna of the Dunkard is placed in 5 categories:

a. Those characteristic of the Rothliegende or higher 12 species.

b. Those clearly allied to Permian types 12 species,

(but the number might be extended).

c. Those whose habit and fades suggest a late date, all new and unknown elsewhere, suggest

later date than the Coal Measures 14 species.

d. Those of Mesozoic aspect, important as nearest relatives are Mesozoic 9 species.

e. Coal Measure types, widespread forms 29 species.

David White considers the Dunkard flora as transitional between the
Permian and the Coal Measures, the beginning of the former being deter-
mined by the first appearance of the Rothliegende forms. He considers
the beds below the lower Washington limestone as transitional beds; above
this the flora contains an increasing number of Rothliegende forms. The
flora of the upper Dunkard is to be compared with that from Stockheim and
Cusel in Germany and Brives in France.

The persistence of a large number of Coal Measure forms and the well-
known and repeated occurrence of more rapid evolution in floral elements
than in the faunal elements in geological time makes this certainly as good
an argument for Perm9-Carboniferous age as for Permian and in no wise
militates against drawing the Permo-Carboniferous line at a lower level.
(See also I. C. White's note of the discovery of Callipteris in the upper
half of the Conemaugh, as reported by David White, cited on page 66.)

Stevenson continues:^

"None of the characteristic coniferous genera Ullmania, Tylodendron, Walchia,
occurs in Dunkard beds, though all are in Prince Edward Island and Walchia is
reported from Texas [and New Mexico — Case] ; and similarly many genera of ferns
characterizing the Rothliegende of Europe seem to be wholly unrepresented. * * *

"The general physical conditions during Allegheny and Conemaugh were
practically the same; for, while the basin was contracting, there was no material
variation in character of the movements; but with the beginning of Monongahela
the area of greatest subsidence was shifted a hundred miles and the new condition
remained unaltered throughout the Monongahela and Washington, which in
this respect are one as the Allegheny and Conemaugh are one. A notable
change occurred in the Washington, and Mr. Wliite has shown that the strongly
marked lower Rothliegende flora makes its appearance near the bottom of the
Greene formation."

In these remarks concerning the similarity of the Allegheny and Cone-
maugh, Stevenson is not in agreement with other writers, who see a decided
difference between the two series.

* White, David, Permian Elements in the Dunkard Flora, Bull. Geo!. Soc. Amer., Abstract,

vol. XIV, pp. 538-542, 1903.

* Loc. cit., p. 173.

76 ENVIRONMENT OF VERTEBRATE LIFE, ETC

(b) The Western Part of the Southern Subprovince.

The upper Pennsylvanian and Permo-Carboniferous of Pennsylvania
and West Virginia is continued almost without break into Ohio and Ken-
tucky. A detailed summary of the Conemaugh formations in Ohio was
quoted from Condit in Publication No. 207 of the Carnegie Institution,
pages 81 to 84, and need not be repeated here. It is noted, page 83 of that
publication, that it is evident from Condit's description that conditions in
Ohio during Conemaugh time were in many regards very similar to those
obtaining (at a later date?) in the Plains Province. Condit says:^

"The Permian of the West, characterized by bright colors and beds of gypsum,
is a still more striking illustration of this kind [beds deposited as deltas in a
semiarid climate]. While evidence of such pronounced aridity is lacking in the
Permian (Dunkard) beds of the Appalachian basin, still it is evident that condi-
tions were somewhat similar. It is believed that the appearance of the red color
in the Conemaugh marks the beginning of the Permian. In southern Ohio,
where the Monongahela coals and limestones are scantily developed, the red
beds are practically continuous from the Conemaugh through the Monongahela,
uniting with those of the Dunkard."

The Dunkard formation in Ohio according to Stauffer' is represented
by only the thin edge left by the erosion, of the much more extensive bed
which lies to the east.

"At the northern end of the Ohio portion of the Dunkard basin there is no
appreciable break between the Monongahela and the overlying Dunkard series.
From the stratigraphic relations the basal plant beds (Cassville) ought therefore
to continue the same flora that flourished during the formation of the preceding
Waynesburg coal bed, but apparently such is not the case. Over the southern
half of the basin, however, the Waynesburg sandstone usually rests directly on
the Monongahela with marked unconformity, the Cassville, the Waynesburg
coal, and a portion of the underlying shales usually being absent. Uncon-
formities in a series of rocks, such as the Dunkard, probably do not have any
very great significance; in fact, they occur at several horizons within the series;
but the development of the coarse, massive Waynesburg sandstone, often a true
conglomerate, over much of the unconformity between Monongahela and
Dunkard, may be indicative of changed conditions.

"* * * This division of the Dunkard into two formations is very arbitrary,
as the stratigraphical or even the lithological break at the horizon used is not pro-
nounced. It does, however, mark the highest level at which marine or brackish-
water fossils were found and probably represents the approximate close of the
oscillations between land and marine conditions, and introduces the purely land
and fresh-water deposits in the Dunkard basin. * * *

"A large part of the Dunkard of Ohio is to be classed as 'red beds,' although
the Monongahela series and even the Conemaugh are not without their red shales,

* Condit, D. D., The Conemaugh Formation in Ohio, Bulletin Ohio Geological Survey,

vol. 17, p. 259, 1912.
^ Stauffer, C. R., Divisions and Correlations of the Dunkard Series in Ohio, Bull. Geol.

Soc. Amer., vol. 27, p. 86, 1915.

DIFFERENT PRO\lNCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 77

which in the Monongahela are often so like those of the Dunkard as to make them
easily confused if it were not for other well-defined strata associated therewith.
There are but few really red sandstones, and those are usually only coated red
on the outside or weathered surface in the Ohio Dunkard. The red is thus almost
confined to the shales. In the northern part of the Dunkard covered area the
red beds are to be found chiefly in the Greene formation, but to the southward
most of the shale in the whole series is red. In the main, these shales, sandstones,
limestones, and beds of coal represent land and swamp or fresh-water deposits,
but the presence of gypsum in certain of the shales and sandstones, and again
marine or brackish-water fossils in other beds, indicates that these conditions at
times gave place to others of a very different character.

"The Dunkard series as a whole is not very fossiliferous ; in fact, it is almost
as barren of the identifiable traces of life as it is of the workable coal seams, which
originally suggested the term 'Upper Barren Measures' for this deposit. In
addition to the occasional plant fragment that may be found in almost any part
of the series, there are certain rather well defined horizons in the Ohio Dunkard
which have yielded important fossils. Plants are, of course, of first importance.
Their remains are occasionally to be found in the roof shales of any of the coal
seams or even in beds of argillaceous shale and sandstone. Almost any outcrop
of limestone may be found to contain small fresh-water gastropods and ostracods.
The middle and upper Washington limestones often contain fish plates and teeth,
some of which are referable to sharks, and are therefore probably marine. A
Lingula occurs in the shales associated with the Washington coal. The lowest
shales of the series are sometimes a black carbonaceous mass associated with a
hard limestone, and these beds contain scales, teeth, and coprolites, all of which
are probably fish remains. The most important find of the whole fossil collection,
howe\-er, was made in the red shales of the Washington formation in the vicinity
of Elba and Marietta. At the former of these places, near the base of the
Dunkard, amphibian coprolites were found in relative abundance. These are
remarkably similar to those found in the Permian of the Western States. At the
latter place, during the past summer, fragments of a neural spine of Edaphosaurus
were found in the sandstones associated with the red shales just above the Lower
Marietta sandstone. The remains of this reptile have never before been found
in the United States outside of Oklahoma, Texas, and New Mexico.* The
importance of this find must be very evident, since it agrees with the earlier
conclusions drawn from identifications of the Dunkard flora and proves the age
of the Dunkard to be identical with the Permian or Permo-Carboniferous of
Texas. After having seen the whole vertebrate collection. Dr. S. W. Williston
says that 'of the fishes I recognize teeth like those of Diplodus from the Texas
Permian, but this type runs through the Pennsylvanian and is not characteristic.
The Elasmobranch spine is unlike any that I have seen in Texas. The coprolites
can not be distinguished from those commonly found in Texas and New Mexico.
* * * The Edaphosaurus spine is unquestionable, small as it is. The range of the
family in Texas is both Wichita and Clear Fork. It occurs in New Mexico in the
El Cobre beds, which the accumulated evidence now places as the equivalent of the
lower Texas beds (Wichita). * * * In Europe JE<fo/'Ao5a;<rM5 occurs in the upper-
most Carboniferous of Kuono%'a and the Rothliegende of Saxony.' (Personal
letter.)"

• Dr. Stauffer here overlooks the discoveries made by Raymond in the Pittsburgh red
shale of western Pennsylvania.

78 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

As stated above, Stauffer believes that this discovery proves the equiva-
lence of the Dunkard with the Permo-Carboniferous beds of Texas and
Oklahoma.

The Monongahela series is but slightly more extensive in Ohio than the
Dunkard, having lost material from the western edge by erosion.

"The character of the rocks interstratified with the coal beds^ changes greatly
in passing from the Monongahela River southward to the Great Kanawha.
At the northern end of the basin * * * limestone forms about one-half of the
rock material, and the same is true on the western side * * *. Red shale is
unknown in the series at the north, but in passing southward from Harrison
and Lewis Counties [West Virginia] the limestones practically disappear, and
with them all the coals but the Pittsburgh. With their disappearance red shales
come in and apparently replace the limestones, so that on the Great Kanawha
nearly one-fourth of the rock material in this series is red shale, while the thickness
is reduced to 270 [from 413] feet."

The Conemaugh extends beyond the edge of the Monongahela in Ohio
and reaches into eastern Kentucky.

"This series,'' as thus limited above and below by important coal beds, consists
of two very different members — an upper one composed largely of shales, therefore
soft, easily eroded, and always making rounded hills and rolling topography, the
other, or lower, composed largely of massive sandstones which resist erosion. * * *

"The upper portion always contains a large percentage of red and marly
shales, which make a broad band of red soil from Pennsylvania through central
West Virginia, to and beyond the Kentucky line on the one hand, and thence
circling around through eastern Kentucky and southern Ohio, back to Pennsyl-
vania again on the other. * * * .

"The coal beds of this series are, with one or two exceptions, noted for their
variableness and uncertainty. They may be in fair development on one farm
and absent entirely on the adjoining one. They are also usually rich in ash and
poor in carbon, and although they are patchy in their distribution, yet the main
beds appear to maintain the same horizon in the stratigraphy, and can thus be
identified with reasonable certainty over wide areas. The sandstones found with-
in the limits of this group are of more economic importance than the coal beds,
since the former nearly always furnish most excellent building stones. * * *

"The limestones of this series, like the coals, are generally thin and impure,
so they are of more importance in determining the stratigraphy than for economic
purposes."

The deposits of the Western Interior Coal Field are here considered
as a part of the Eastern Province, but it is probable that the major
portion is below the middle Pennsylvanian and so throws little direct light
upon the conditions during late Paleozoic time. Upon the usual basis of
correlation by fossils there is no suggestion that any layers below Coal 6
are higher than the base of the Conemaugh. The uppermost division of the
series. Coal 6 (Grape Creek, Herrin) and above, may be of Conemaugh age.

* White, I. C, Stratigraphy of the Bituminous Field Coal of Pennsylvania, Ohio, and West

Virginia, Bull. 65, U. S. Geological Survey, p. 43, 1891.
> White, I. C, loc. cit., p. 71.

DIFFERENT PROVINCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 79

"WV may safely conclude that the horizon of the Grape Creek flora is not
lower than the Freeport group on the one hand, while granting that it may on
the other hand, eventually be found to be at a somewhat higher stage in the as
yet paleobotanically unknowTi Conemaugh series."

\ATiite also makes the statement that the closest aflfinities with the Grape
Creek flora w ill probably be found in the lower portion of the Missourian
or the uppermost Des Moines:

"The composition of the Grape Creek flora indicates a stage in the lower
Stephanian of the Old World. The latter division of the European Coal Measures
appears, in the present stage of our knowledge of the fossil floras, to correspond
to the Monongahela and Conemaugh series, together, perhaps, with the Freeport
group of the Allegheny series of the eastern United States, and to the Missourian,
with the upper portion of the Des Moines, of the Interior Basin."

In Bulletin 15 of Illinois Coal Mine Investigations, Cady^ states that
David WTiite concludes from floral evidence that Coal No. 6 (Grape Creek,
Herrin) "may be of Freeport age, possibly as high in the stratigraphic
column as the upper Freeport coal, which is the uppermost layer of the
Allegheny formation in the Appalachian region."

In the scheme adopted by the Illinois geologists all the Pennsylvanian
deposits above Coal No. 6 are included in the McLeansboro formation.
The dividing-line between this and the underlying Carbondale formation
is not readily distinguished in many places, but in the light of fossil and
stratigraphic evidence it seems safe to consider the McLeansboro as equiva-
lent to the Conemaugh and higher series in Pennsylvania.

It is to be noted, however, that little or no red shales or sandstones
appear in the McLeansboro formation. If this formation is equivalent to
the Conemaugh and higher, then the necessar>' conditions for the formation
of red beds were either never present in Illinois, or, what is more probable,
the conditions necessary for the formation of red sediments had not reached
Illinois within the time of deposition of any portion of the McLeansboro now
preserved.

In Bulletin 15 of the Illinois Coal Mining Investigations cited above,
Cady gives an account of the McLeansboro formation, from which the
following is abstracted :

"The formation consists of several distinctive beds of shale and a minor
amount of sandstone, limestone, and coal. Although several of the coals above
No. 6 are persistent, none have been found sufl[iciently thick to be of commercial
value. They are significant only as correlation horizons. In its barrenness of
productive coals and in general age, the McLeansboro is similar to the Conemaugh
formation of Pennsylvania."

• White, David, in the Dan\-ille Folio, No. 67, U. S. Geological Survey, 1900.

* Cady, G. H., Illinois Coal Mine Investigations, Bull. 15, Coal Resources of District VI,

p. 26, 1916.

80 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

There follows in plate ii, opposite page 26, a series of graphic representa-
tions of drill-holes in the McLeansboro in district vi and Shelby County
and a detailed record of a single hole, which show the absence of red shale
and sandstone and prevalence of black, blue, and gray colors.

In district vi the following are the most well-marked of the horizons

above Coal 6.

7. New Haven limestone

6. Shoal Creek limestone.

5. Carlinsville limestone.

4. Coal No. 8, 8 inches to 2 feet.

3. A bed of pink, red, or variegated

shale, variable in thickness, seldom

exceeding 15 feet (local).
2. Coal No. 7, generally only a few

inches thick.
I. A hard limestone averaging 7 feet

in thickness, overlying and slightly
above Coal No. 6.

Most of these horizons can be recognized in district vi, but not all, and
there are some in district vi not occurring in vii. Layer 3 is not present in
district vi. The limestone directly above Coal No. 6 bears marine inverte-
brates.

"Of the remaining 400 feet, more or less, of the McLeansboro formation (above
the New Haven limestone) known from drilling in this district, only the lower
300 feet or so has been explored by the drill a sufficient number of times to warrant
generalizations in regard to jt. * * * Most of the material above the New
Haven limestone is shale and sandstone with no characteristic beds."

One bed of coal less than 5 feet thick and lying about 550 feet above
Coal No. 6 is mentioned in a number of well records.

The nature of the beds in the McLeansboro formation appears to the
author a confirmation of his position long held, that the reptilian and
amphibian remains found near Danville, Vermillion County, Illinois, occur
in deposits of a much later date than the beds with which they are associated.

The McLeansboro formation of Illinois is equaled in Indiana by a few
hundred feet of shale and limestone with a few thin coal beds. Ashley^
has divided the upper Pennsylvanian of Indiana, as shown in the correlation
table, opposite page 48.

On page 59 of Ashley's report cited above, it is stated that in a general way
the coals and rocks above coal vii in Indiana belong to the Conemaugh and
higher formations of Pennsylvania. With the exception of limestones above
coal VII there are only shales and sandstone, with clays just above the coals:

"There is no dominant sandstone except one above what may be called
coal IX, which is believed to be the sandstone outcropping at the top of the bluff
at Merom, and from this exposure has been called the Merom sandstone.

* Ashley, G. H., Supplementary Report on the Coal Deposits of Indiana, 33d Annual Report
Department of Geology and Natural Resources of Indiana, 1908.

DIFFERENT PROVINCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 81

"A short distance below the Merom sandstone is commonly found a limestone
which is thought to correlate with what has been called the Somerville limestone
of southern Illinois and southwestern Kentucky, though that correlation is rather
conjectural than demonstrated." (Ashley's report, p. 6i.)

The character and position of the Merom sandstone in Indiana has been
described in Publication 207 of the Carnegie Institution, pages 78 to 80.
The equivalents in Kentucky as taken from Miller^ are shown in the correla-
tion table.

The red and purple shale and sandstone mentioned by Miller is not
described in any of the reports of the Kentucky Geological Survey dealing
with the western field. The only red deposits in western coal field of Ken-
tucky occur in connection with the Madisonville limestone in the Earlington
quadrangle, which lies in western Hopkins and southern Webster Counties.
The Madisonville limestone here lies about 185 feet above the Nebo coal,
which is generally considered to be the equivalent of coal 14.

"The Madisonville limestone* contains two to four divisions, ranging through
a maximum interval of 40 feet. * * * It is hard, brittle, very resistant to weather-
ing agents, weathers to a gray color, and carries an abundance of marine fossils.
Between the beds of limestone are intervals of red clay and shale * * *."

This horizon is well above the red beds of the eastern coal field of Ken-
tucky which are of Conemaugh age.

West from Illinois there is a second break in the outcrop of the upper
Paleozoic caused by the uplift and disturbance of Ozarkia in southern
Missouri. The connection, if any existed between the upper Pennsyl-
vanian beds on either side of this break, was probably through Iowa.

The Missourian, upper Paleozoic of Iowa, is a direct continuation of the
same formation in Missouri and Kansas and does not diflfer materially from
them. There is no indication of red beds or red-bed conditions in this
part of the formation.

In Webster County a small area of red sandstone and shale accompanied
by gypsum lies unconformably upon the Des Moines formation and the
St. Louis limestone where the Des Moines has been eroded away. "An
erosion interval of considerable length thus separates the period of their
deposition from the Des Moines epoch." '

The red rocks and gypsum of this limited area have been tentatively
referred to the Permian upon stratigraphic grounds by Wilder,* but as the

' Miller, Arthur M., Table of Geological Formations for Kentucky, Department of Geology

University of Kentucky, 1917.
' Kentucky Geological Survey, series iv, vol. 11, pt. i, p. 132, 1914.
' Norton, W. H., and others, Underground Water Resources of Iowa, Water Supply

Paper No. 293, U. S. Geological Survey, p. 86, 1912.
* Wilder, F. A., Geology' of Webster County, Geological Survey of Iowa, vol. 12, p. 63, 1902.
7

82 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

author has shown in Publication 207 of the Carnegie Institution, page 77,
the red deposits are not to be directly correlated with the red beds of Texas
and Oklahoma, though they were undoubtedly formed very near the top of
the Pennsylvanian (Missourian) and may even be Permo-Carboniferous.

{d) Conditions in Missouri.

The condition of the area between the extreme western edge of the
Eastern Province (Illinois and Kentucky) and the eastern edge of the Plains
Province in Missouri and Kansas during Pennsylvanian time has been
well described by Hinds and Green. ^

"At the beginning of the Pennsylvanian epoch the area included in the
present boundaries of Missouri was above sea-level. The highest part was a
plateau corresponding roughly with a tongue projecting into the northeastern
part of the State a short distance west of the site of the Mississippi. The region
now occupied by the main body of the Pennsylvanian was lower, though probably
the difference in altitude of the two areas was slight. Meanwhile sediments were
being deposited in shallow seas occupying parts of Oklahoma, Arkansas, and
northern Illinois and the waters were slowly advancing over adjacent land
areas * * *."

[Near the beginning of the Allegheny (Cherokee or Henrietta)] "The land
area had been reduced to an island in southeastern Missouri, with a peninsula
projecting into Pike and neighboring counties and a small part of a northern
land-mass in the extreme northwestern corner of the State. The western sea
continued to advance eastward, while an eastern sea occupying most of Illinois
advanced westward. Probably by the end of Cherokee time the two seas had
joined, submerging practically all of northern Missouri and possibly nearly all
of southern Missouri also. No deposition appears to have taken place at this
time in the extreme northwestern corner of the State, for the Nebraska City
drilling shows less than 100 feet of Des Moines strata, probably of Pleasanton age.

"There is still much doubt as to whether the Pennsylvanian sea finally
covered practically all of southern Missouri and submerged the Ozarks, though
the evidence in hand seems to indicate that a large part of the region was inun-
dated for a comparatively short interval, beginning, probably, near the end
of the Cherokee epoch. In nearly all the Ozark counties there are small outliers
or pockets of shale, sandstone, and coal in sink holes and other protected situa-
tions. Mciny of these, at least, are of Pennsylvanian age, but were probably
deposited before invasion or after the sea receded from the region. The sink
holes themselves were certainly formed while above ground-water level and some
of them seem to have been deepened while being filled with Pennsylvanian coal
and other materials. The remarkably thick pockets of cannel — a coal formed
very slowly from only plant products most resistant to decay— were deposited
in stagnant water that was probably fresh.

"In addition to the pockets, however, sandstone and shale of Pennsylvanian
age are scattered over the Ozarks in small patches capping divides where erosion
has not been active. These outliers may have been deposited at the time when
the sea covered all or most of Missouri. The thinness of the probable marine

* Hinds, Henry, and F. C. Green, The Stratigraphy of the Pennsylvanian Series in Missouri,
Missouri Bureau of Geology and Mines, vol. xiii, ad series, p. 208, 1915.

DIFFERENT PROVINCES OF NORTH AMERICA IN LATE PALEOZOIC TIME 83

Pennsylvanian sediments in all of the Ozarks, however, indicates that the sea
may have retreated again in a comparatively' short time, probably before the
end of the Des Moines epoch. If the Warrensburg and Moberly channels came
into existence late in the Pleasanton eixxrh, as seems probable, a relative uplift
of the Ozark took place at that time. Moreover, the differences in the sediments
laid down in Missouri and Illinois during the Missouri epoch, so far as known
from strata still intact, point toward the presence of a land-mass between the two
areas during that interval. Some of the sands deposited in parts of the Missouri
ep>och are also most easily explained by postulating a land-mass in southern
Missouri. The overlap of Des Moines strata toward the west and the probable
derivation of some early Des Moines sediments from an Ozark land-mass, on
the other hand, seem to show that the Ozarks were above sea until late in the
Cherokee epoch. * * *

"Sedimentation during Missouri Epoch.

"The Missouri group seems to have been deposited under conditions which
alternated between those of quiet watere, which permitted the growth of marine
invertebrates but excluded clastic sediments to a large degree, and those of
unsettled amd disturbed waters in which sandstones and shales were deposited.
From time to time the more unsettled conditions changed during short intervals
in which lenticular coal or limestone beds were formed. While quiet waters
prevailed and calccireous materials were conspicuous Eunong the sediments, condi-
tions were unfavorable for extensive plant growth. Even at other times coal-
forming plants succeeded in establishing themselves only for relatively short
inter\als and, with one or two exceptions, in compzu^atively small swamps. The
intervals of limestone deposition, on the whole, grew shorter as time progressed.

"One of the notable features of the deposition during the Missouri epoch was
the rej>etition of an alternating succession of limestones and thin shales with
thicker shales and sandstones. Almost exactly similar conditions of sedimenta-
tion appccu- to have recurred intermittently over wnde areas. There is a striking
similarity* in the Plattsburg and Stanton, Oread, and Deer Creek limestones and
to a less degree in the Lecompton, Topeka, and Howard, and the Tarkio and
cap-rock limestone of the NvTnan coal. In each case the sections show only
minor variations from the following succession:

1. Limestone, flaggj-; a thin bed (at top). 5. Limestone, dark gray; even-bedded; I or 2 feet.

2. Shale, drab; a few feet. 6. Shale, drab.

3. Limestone, gray, thin-bedded; a thick bed. 7. Limestone, blue (at baise).

4. Shale, black, slat>'.

" In the Plattsburg and Stanton, Oread, and Deer Creek members this succes-
sion is t>-pically shown. In the other cases mentioned the place of the dark-gray,
e\-en-bedded limestone (5) seems to be taken in some areas by coal, and the
limestone (3) is much thinner.

"The clastic members have certain resemblances, which, however, are not
nearly so striking as those just mentioned. Most of them contain sandstones
that vary in apparent stratigraphic position within short distances, and include
limestones that do not maintain uniform thicknesses.

"Deformatioxs.

"From the beginning to the end of Pennsylvanian time in Missouri earth
movements in the r^on now occupied by the series were relatively slow, simple,

84 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

and uniform. In general there was a long-continued subsidence of the region,
broken by periods of stability and with, perhaps, relative uplift of adjacent land
areas during several intervals. The uniformity of the subsidence is shown by
the persistence in thickness, areal extent, and character of most of the members
of most of the formations. The periods of stability culminated in the formation
of the widespread coal beds, after sedimentation had filled the sea and caused its
withdrawal, and ended when a renewal of subsidence again let in the saline waters,
killing the coal plants.

"The relative uplift of neighboring land areas is indicated by the periodic
recurrence of irregular deposition and a comparatively large proportion of arena-
ceous sediments. In most Pennsylvanian formations the strata are remarkably
persistent and regular, but in the Pleasanton, Douglas, and part of the Cher-
okee formations, and in the Lane, Severy, Scranton, and a few other members,
the strata are variable. An influx of sands was usually caused, probably, by
changes in the currents of the shallow sea, in the direction of drainage lines on
neighboring land-masses, or in the derivation of sediments. During the Pleas-
anton and Douglas epochs, however, the phenomena were somewhat more com-
plex. As stated more fully on previous pages, there is evidence that the sea may
have withdrawn from all or part of Missouri in both Pleasanton and Lawrence
time, while long and rather deep channels were formed by subaerial erosion.
These changes appear to have been effected by slight tilting and folding in
northern and western Missouri, as well as by differential uplift of the Ozark region.

"After the close of the Pennsylvanian there were two periods of folding.
The first of these resulted in the blocking-out of the main broad features of the
present structure, namely, the monoclinal dip to the west in north Missouri and
to the northwest in the west-central part of the State. The second period of
folding caused the formation of narrow and comparatively sharp anticlines and
associated synclines trending northwest-southeast and markedly parallel through-
out the State. * * *"

CHAPTER III.

THE PLAINS PROVINCE.

The Plains Province of deposition in Permo-Carboniferous time was
in all probabilit>'^ a continuous whole, as described in Publication 207 of the
Carnegie Institution, with a gradually shrinking body of clear water sur-
rounded by large areas of red-bed deposition traceable from the Black
Hills of South Dakota to New Mexico along the eastern front of the Rocky
Mountains. The red beds deposited, in all probability, on the eastern side
of the shrinking body of water have either been removed by erosion north
of southern Kansas or are covered by younger deposits. The following
summary description is given by States or by convenient units; it is obvious
that the beds frequently extend across the artificial political boundaries.

A. THE LATE PALEOZOIC IN KANSAS.

The series of upper Paleozoic rocks in eastern Kansas is the most com-
plete and illuminating of any section in the western portion of North America
and is taken as the standard with which are compared the various exposures
in the Plains Province. The long-drawn-out controversy as to the age of the
upper Paleozoic rocks of Kansas has now little more than historic value, but
it has, for good or ill, definitely attached to the upper part of the series the
name Permian. This has been, with little doubt, the cause of much of the
difference of opinion and the source of many of the controversial papers
that have been published. Had this difficulty, more than half a historical
matter, not persisted, the effort to find a sharp dividing-line between Penn-
sylvanian and Permian would not have been so vigorous or so long sustciined,
and a recognition of the essential similarity of the beds under the name
Pennsylvanian and Permo-Carboniferous would have been much earlier
recognized. As it is, the line between the "Permian" and Pennsylvanian
has been forced downward by successive stages until now the Kansas lower
"Permian" includes all the rocks from the base of the Elmdale formation
to the top of the Wellington shales. These include series iv and v of the
Kansas Geological Sur\-ey, with stages i, j, the Chase, Marion, and Welling-
ton, as given by Beede in the volume ix of the Kansas University Geological
Survey.

Whatever may be the final outcome of the controversy concerning the
terminology' of these beds, they are very certainly the equivalent of beds
called Permo-Carboniferous elsewhere in the United States, and the author
will consistently regard them as of such age in this work.

ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Permian

Wellington.

Wellington shales.

Marion

Abilene conglomerate.
Pearl shale.
Herington limestone.
Enterprise shales.
Luta limestone.

Chase

■ Winfield limestone.
Doyle shale.
Fort Riley limestone.
Florence flint.
Matfield shales.
Wreford limestone.

1 K°''Tf '^^k""' 1 Garrison formation.
Neosho member /

Cottonwood limestone.

Eskridge shales.
Neva limestone.
Elmdale formation.

A detailed description of these beds is given by Prosser,^ from whose
paper the following descriptions are quoted :

[Elmdale formation.] "It is about 130 feet in thickness, and composed of
yellowish to bluish shales, with thin beds of grayish alternating limestone, includ-
ing two or three thicker ones. About 30 feet above the base of the formation
is a friable limestone with a thickness in some localities of 4 feet, which is com-
posed to a large extent of the tests of Fusulina secalica Say. This stratum
weathers rapidly and leaves great numbers of Fusulina in the soil. About 35
feet higher is another conspicuous yellowish limestone, the center of which
weathers to a rough face, and from 10 to 15 feet below the top is a limestone
stratum from 3 to 5 feet in thickness. * * *

"Neva limestone. — This formation consists of a massive bluish-gray limestone
or of a lower and upper massive limestone, each one a little over 4 feet in thickness,
separated by 2 feet of shales, with a total thickness of about 10 feet. * * *

"Eskridge shales. — * * * a mass of shales, with perhaps some thin limestone
layers, varying from 30 to 40 feet in thickness. The shales are of greenish,
chocolate, and yellowish color, and usually form covered slopes between the two
conspicuous limiting limestones. * * *

[Cottonwood limestone.] — "This is a massive light gray to buff-colored, fora-
miniferal limestone, frequently composed of two layers with a thickness of about
6 feet. It contains very few fossils, with the exception of Fusulina secalica Say,
which is extremely abundant in its upper part, * * *" [called Alma limestone
by Prosser].

"Garrison formation. — This formation is composed of two members, the
yellowish fossiliferous shales at the base, formerly called the Cottonwood shales,
and the upper one, composed of the alternating gray limestones and various
colored shales called the Neosho, with a total thickness of from 140 to 145 feet.
The lower shales have a thickness of 13 feet near Strong, but decrease to 2 or 3
feet in the northern part of the state. * * *

' Prosser, C. S., Revised Classification of the Upper Paleozoic Formations of Kansas, Jour.
Geol., vol. X, p. 708, 1902.

THE PLAINS PROVINCE 87

[Garrison formation, Florena shales.] — "The upper member of the formation
is composed of green, chocolate, and yellowish shales ciltemating with grayish hme-
stones, while in the Big Blue valley a bed of gypsum occurs near the base. * * * "

" Wreford limestone. — This formation is composed of limestone cmd chert,
or flint, as it is popularly termed throughout the Flint Hills region, and varies in
thickness from 35 to 50 feet. In general, it is composed of three strata, a cherty
limestone below and above, separated by a heaivy limestone nearly free from
chert. The rock is buflF color. * * *

" Mat field shales. — The formation is composed principally of variously colored
shales, with some shaly buff, occasionally cherty limestones, and a light-gray
limestone 2 feet or so in thickness, which occurs about 30 feet below its top.
The thickness ranges from 60 to 70 feet, and it generally forms covered slopes
betw-een two massive and conspicuous flint ledges. * * *"

"Florence flint. — This formation is about 20 feet in thickness and consists
of very cherty limestone separated by definite layers of chert, with a band of
shaly or white cellular limestone near the center. * * *"

"Fort Riley limestone. — Overlying the Florence flint is a series of massive
buff limestones, changing to thin-bedded and shaly strata in the upf>er part of
the formarion, which have a total thickness of 40 feet or more. Near the center
of the formation are generally one or two massive layers, which on the weathered
surface form a conspicuous ledge that may be readily followed by the eye for
miles. * * *••

"Doyle shales. — This formation is composed of variously colored shales with
an occasional thin stratum of soft limestone, and has a thickness of 60 feet.
About 20 feet above the base is a thin, grayish limestone which often appears on
the surface, and at the top are yellowish shales containing a few fossils. * * *"

" W infield formation. — This has a thickness of about 25 feet, and is composed
of a cherty limestone at the base wath a massive concretionary one at the top,
the two separated by yellowish shales. * * * The chert is not so uniform in
occurrence as in the Wreford and Florence flints, and at some localities this
horizon is represented simply by a prominent light-gray limestone, nearly free
from chert. * * * The irregular worn upper surface of the concretionary lime-
stone and the appearance of many of the concretions, as though rolled in the mud
on the sea bottom, indicate a shallowing of the sea at this time, followed by a
subsidence of the sea-bottom before the deposition of the succeeding even thin-
bedded limestones. This change of physical condition is indicated in the fauna
by the nearly complete disappearance of the brachiopods and the survival of a
fauna composed mainly of Permian lamellibranchs. * * *"

" Marion formation. — Buff thin-bedded limestones and shales form the prin-
cipal part of this formation * * *. The lower part is composed of rather soft,
porous, thin-bedded limestones and shaly layers to shales, containing near the
base a considerable number of siliceous geodes and occasionally some chert.
Some 50 or 60 feet above the base is a buff limestone containing large numbers of
lamellibranchs. * * *"

"The upper portion of the formation is composed mostly of thin buff lime-
stones similar to those in the lower portion, alternating with a greater thickness of
shales and marls, and in some localities contains beds of g>'psum and salt."
[The top of the formation is a conglomerate, while at various localities and
different levels beds of gypsum of veuying thickness occur.]

88 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

"Wellington shales. — ^This formation consists largely of bluish-gray to slate-
colored shales, but contains some red ones, and in the southern part of the State
beds of impure limestone and calcareous shales, together with occasional beds of
gypsum and dolomite. Limited saline deposits are reported, but no rock salt."

This series has been shown by Beede and Sellards^ to be singularly
persistent through the State. They say:

"From what has preceded it will be seen that the strata of the lower Permian
are remarkably persistent and uniform when the great extent of the outcrop is
considered. The Cottonwood limestone, though only about 6 feet thick, persists
with every detail of structure and fauna over loo miles of strike and several times
as great an outcrop, though it has not been identified with certainty in the
southern part of the State. The Garrison formation extends entirely across the
State, with but slight modifications in the southern part, such as the thickening
of some of its limestones and the possible interpolation of others. The Wreford
limestone is remarkably uniform throughout the entire distance from Nebraska
to the southern line of Kansas, being most highly developed in the central part
of its outcrop in the region of Cottonwood Falls. In the Matfield shales about
the only change worthy of special notice is the thickening of a layer of limestone
and the coming in of an additional one in the southern part of its outcrop. There
are no striking changes in the Florence flint, aside from a slight fluctuation in its
thickness, being somewhat thicker in the central and southern regions."

Toward the southern line of Kansas the limestones shade into sand-
stones and shales. This significant change is described by several authors.

"In tracing the outcrop^ of the limestone formations of the Carboniferous of
Kansas, the writer observed that in going southward there is a gradual transition
in the character of the sediments to those which are more arenaceous, and that
there is a thickening of the shales and sandstones and a thinning and final dis-
appearance of the limestones. * * *

"* * * From what is known of the Permian limestones of Kansas, they will
be found, when followed southward, to diminish in thickness, and this change
will be accompanied by a transition to more sandy beds. * * *"

The "Wellington Shales" are probably represented southwestward by
formations which are red. The approximate limit of the red color is a line
diagonal to the strike of the formations, and is found to correspond in a
general way with a line drawn by Mr. Cummins as separating the Carbon-
iferous and Permian.

"The distinctions which have been thus far outlined in Kansas do not hold
where the rocks are followed southwestward along their strike into Indian Terri-
tory. Approximately along the Arkansas River, or a little south of that stream,
the interstratified limestones disappear from that section, and the formations are
accordingly shales and sandstones. Moreover, the rocks in Indian Territory
gradually assume a red color in the higher portions of the section, the line of
transition to this color being diagonal to the strike."

' Beede, J. W., and E. H. Sellards, Stratigraphy of the Eastern Outcrop of the Kansas

Permian, Amer. Geol., p. 109, 1905.
^ Adams, G. I., Carboniferous and Permian Age of the Red Beds, Amer. Jour. Sci., vol. xn,

P- 383. 1901. A full description of conditions, with maps, is given by Adams in the

Bulletin of the Geological Society of America, vol. 14, p. 191, and in Bull. 211, U. S.

Geological Survey, 1903.

THE PLAINS PROVINCE 89

In 1909, Beede^ gave the following account of the transgression of the
red color into the limestones :

"The limestones do not all continue to the southern limit of Kansas, some
of them pinching out before reaching the Oklahoma line and others soon after
crossing it. Few of them pass beyond the Arkansas River in that State. It
seems that the central part of the Kansas Basin may have been to the north-
westward during later Pennsylvanian time, since the shales frequently become
thinner, and the limestones thicker in that direction, though this can not be
said of the lower part of the section. Above the Americus limestone the succes-
sion of limestones and shales continues for about 700 feet. However, the shales
become more calcareous and marly, the limestones more porous and less crystal-
line; massive gypsum beds are intercalated, and coal in quantities is wanting.
The limestones also weather white. These changes are significant of decided
physical or climatic changes, as the local pools of the lower horizons showed no
tendency to concentrate and form massive gypsum deposits. Probably, also,
the changed aspect of the limestones is indicative of these altered conditions.
The first large deposits of gypsum occur just above the Cottonwood limestone
in the lower part of the Garrison formation (Neosho member). Above these are
the Wreford limestone, Florence flint, Fort Riley and Winfield limestones, heavily
charged with chert, and separated by thick layers of shale. The outcrops of
these formations form the 'Flint Hills' of the eastern part of central Kansas.
Over these strata are two soft limestones with three intervening shale beds and a
variegated, brecciated, thin limestone. These are grouped in the Marion stage,
and end the regular succession of limestones and shales. Over the rocks of the
Marion stage lie the Wellington shales, probably several hundred feet in thickness,
composed of blue, green, and some red shales. Upon these shales lie 1,400 feet
of red beds in Kansas. The upper part of the Red Beds does not occur in Kansas,
but is found in western Oklahoma and the Panhandle of Texas.

"The whole of the lower succession of shales and limestones forming lowlands
and low escarpments divide this section of continuous sedimentation into short
stratigraphic units of great lateral extent convenient for paleontologic study.

"In Oklahoma diff^erent conditions prevailed during much of the time repre-
sented by the Kansas deposits, above the Cherokee shales.

"Passing from Kansas to Oklahoma, the light-colored shales and limestones
of the upper part of the Kansas section grade off into red shales and sandstones.
The lowest horizon in Oklahoma at which the red sediments predominate is
unknown, inasmuch as the strike of the rocks is but little west of south, and the
Red Beds protrude eastward in central Oklahoma as a sort of embayment,
especially north of the Arbuckle Mountains.

"In the region south of the western end of the Arbuckles the Red Beds lie
unconformably upon the tilted and eroded Pennsylvanian rocks. It appears
that the Albany- Wichita sea of northwest Texas transgressed over this region
during a time of slight depression, the waters covering the western end of the
Arbuckle Mountains, swinging eastward on their northern slope as far as the
Seminole country. According to Cummins, there is no unconformity in Texas
between the lighter sediments and the Red Beds, the transition between the

* Beede, J. W., The Bearing of the Stratigraphic History and Invertebrate Fossils on the
Age of the Anthracolithic Rocks of Kansas and Oklahoma, Jour. Geol., vol. xvii, p.
712, 1909.

90 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Albany and the Wichita being a gradual lateral one. The transgression of the
Red Beds in the Arbuckle Mountains may, then, be regarded as a northeastern
or eastern encroachment of the Wichita sea — or conditions of sedimentation,
as all these beds may not be marine. Whether this Arbuckle unconformity
extends northeastward to the easternmost limit of the Red Beds has not yet been
determined, and indeed may be very difficult to determine, where the uncon-
formity would resolve itself to a mere disconformity of layers of shales, and
perhaps accompanied by a greater or less reworking of the lower deposits.
Gould, who has been over this region between the Arbuckles and the Arkansas
River many times, states that he knows of no unconformity. If no unconformity
exists to the north of the Arbuckle Mountains, it seems probable that the first
Permian emergence began here and the deposition of the red beds in the Seminole
country is the first record of it, the later sediments from the Arbuckles reaching
farther north. Regarding the gradation of the upper part of the Kansas section
into the Red Beds in northern Oklahoma, there can be no doubt whatever, and
the same is probably true of the central part of the State.

"The Arbuckle and Wichita mountains are probably the source of much of
the red sediment in which they are partially buried, and the former mountains
are directly responsible for the eastern extension of these beds into central
Oklahoma. The extent to which the lighter-colored sediments of Kansas and
Texas are replaced by red sediments in Oklahoma and near it represents in a
rough way the limits of the influence of these mountains on the deposits of the
time by the spread of their sediments. By the time the deposition of the light-
colored sediments had ceased the conditions had become such that nearly all
the sediments derived from the land surrounding this basin were red.

"In the Oklahoma region the deposition of red sediments began, perhaps,
as low as the Howard or Topeka limestones, and perhaps as high as the Emporia
or Americus limestones. The deposits then seem to be uninterrupted until the
unconformity below the Dockum beds (Triassic) in the Texas Panhandle is
reached. Some of these beds appear to be of subaerial origin, as has been shown
by Case, while others are certainly marine. Careful petrologic study will prob-
ably demonstrate that much of the arenaceous material is windblown sediment,
more or less reworked by currents or waves as the regions were submerged or
flooded. That the sea ever covered the entire area from Kansas to southern
Texas and New Mexico at one time may be questioned. If it did, the sediments
contained were of such a nature and abundance, or the waters so concentrated,
as to preclude the free migration of a normal marine fauna throughout the basin.
That marine conditions prevailed, at least locally, is demonstrated by the White-
horse and Dozier faunas.

"In Texas normal deposits were laid down in higher horizons than in Okla-
homa, and in Kansas there are reasons for believing that the light-colored
sediments were laid down at an even later date than in Texas. These conditions
are illustrated in the subjoined table, showing a vertical section of the Carbonifer-
ous and Permian rocks of the three States.

"The extent of this post- Pennsylvania basin seems to have been very great.
It included much of Kansas (two- thirds), western Oklahoma, much of western
Texas, and all of New Mexico, Colorado, and Wyoming east of the Rocky Moun-
tain axis. In area it probably aggregated 300,000 square miles.

"Together with the varied physical conditions of these three regions went
corresponding faunal peculiarities. In the Albany division of the Texas rocks

THE PLAINS PROVINCE 91

the Pennsylvanian elements of the fauna seemed to persist, while they are largely
wanting in their equivalent beds, the Wichita di\asion. A similar thing occurs
in the clear-water beds of northern Oklahoma and southern Kansas, north of the
Red Beds. Aside from this general fact it should be noted that along the region
of the Red Beds and light sediment (littoral?) contact, some of the Pennsylvanian
elements of the Kansas fauna persisted much longer them in the same rocks to
the northward. The fauna of any given horizon above the Elmdale formation
varies very sensibly as we pass from the Nebraska to the Oklahoma line, both in
abundance of specimens and species, and in the general aspect of the faunules
as well. This is to be expected in the light of the intercalation of massive gypsum
beds as low as the lower part of the Neosho member in the northern region.
From it we would infer that the waters of the northern-main marine part of the
basin were somewhat more concentrated than at its southern shore."

In 1 91 2, Beede^ gave a second account of the same phenomenon:

"In tracing the limestones and shales of the basal Permian beds of Kansas
southward into Oklahoma the relationship of the light-colored sediments to the
red sandstones, red shales, and red limestones of Oklahoma is clearly revealed.
It is shown that some of the heavier ledges of limestone first become sandy along
their outcrops in patches a few rods across. Farther south the sandstone areas
increase in size until the limestone appears only in local areas in the sandstones
and is finally wanting. Traced farther southward, the sandstones become deep
red or brown with local areas of white. The decimation of the fauna sets in as
the limestones diminish and the remains of life are not found far beyond the limits
of the limestones. The shales become red very much farther north than do the
sandstones, and are frequently more deeply colored. Some of the lower lime-
stones become red before they change into Scmdstones. The sandstone ledges
continue for some distance southward as rather even, uniform beds, but farther
on they are found to thicken and thin in a somewhat systematic manner.

"Several ledges of sandstone frequently occur in a single section, and where
one of these ledges is found thickened the others are apt to be thicker than normal.
Likewise they are all found to be thin over certain areas. The r^ons of thicken-
ing and thinning were found to be parallel belts lying north and south at right
angles to the major drainage lines. Two of these belts, together with an inter-
vening region about 8 miles across, were studied. The sandstones thicken at the
expense of the shales, sometimes eliminating them. In one instance a thin lime-
stone was traced southwest into one of these zones. A sandstone 20 feet or more
beneath the limestone thickens and rises above the limestone and practically
unites with the sandstone some distance above it. The limestone seems to die
out a few feet from the sandstone, but farther west the latter shrinks to its
normal thickness and the limestone is present in its proper position with its usucil
characteristics."

A later paper by Beede^ gives a more detailed account of the transition
of the limestone into shale and red beds:

• Beede, J. VV., Origin of the Sediments and Coloring Matter of the Red Beds of Oklahoma,

Science, vol. xxxv, p. 348, 1912.

* Beede, J. VV., The Neva Limestone in Northern Oklahoma, with Remarks upon the

Correlation of the Vertebrate FossU Beds of the State, Oklahoma Geological Survey,
Bull. 21, p. 24, 1914.

92 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

"One of the most interesting features of this whole region is the nature of
the changes from the light-colored limestones and shales to the dark-red sand-
stones and peculiar shales of the Red Beds.

"The shales are red much farther north, as a rule, than are the limestones
and sandstones. The change in color is frequently accompanied by some change
in the character of the shale. The red shales are usually much less compact
and durable and in the immediate region covered by this report seem to become
more or less charged with very fine sand. On account of the fact that the shales
are usually hidden from view, the nature of the transition has not been observed
so carefully as has the transition from limestone to sandstone.

"In the case of some of the higher limestones, Wreford, Fort Riley, etc.,
sand appears in the limestones, which have usually thinned appreciably. The
sand may gradually increase for considerable distances, say from a few rods to
a few miles, and become first a sandy limestone, then a calcareous sandstone.
Followed still farther, the traces of calcium carbonate disappear, sometimes to
reappear as limestone in some areas. Again, as is shown along the Shawnee
branch of the Santa Fe Railroad from Kaw City to Skedee, or the upper Wreford
limestone at Hardy, the first traces of the transition are seen in purple blotches
scattered through the stone. These may enlarge and increase in number until
the whole stratum is practically a purple or red limestone. In other regions the
limestone may turn almost scarlet in a rod or two, as in the case with a lime-
stone in the escarpment south of Gushing. The red limestones of the latter class
usually dissipate quickly into sandstones. They are usually fossiliferous.

"Sometimes a limestone layer will grade into a sandstone layer and then
change back again into limestone in a few rods. Indeed, this is not infrequent in
the region between Kaw. City and Pawnee, and west and northwest of Pawnee.
* * * Sometimes these sandstone replacements may not be more than 3 or 4
rods across. * * * The sandstone in such cases is usually calcareous, but in
some instances it is not.

"At one point a ledge was made up of sandstone and limestone in indis-
criminate masses, which were very irregular in form. The masses were all rather
small, hardly ever over 2 feet in diameter and ranging from that to mere pockets.
Sometimes there were pockets of sandstone in the limestone and sometimes
pockets of limestone in the sandstone. That is, sometimes one or the other forms
the predominating rock. On the whole, the exposure was largely limestone.
In most all cases the transition from the light-colored sandstone to red sandstone
takes place before going a great distance. * * *

"After passing some distance south or southwest of the region of transition
just described, in which the sandstones maintain their usual thickness and relative .
positions, we pass into another zone where they thicken and thin, pinch out, end,
and even cut out intervening beds of shale and limestone. * * * In this region
stratigraphic work becomes more uncertain, the fossils are wanting, and there
seems to be no character of the rocks to tie to. At the bridge at Ripley is a
sandstone about 40 feet in thickness which elsewhere is usually about 4 or 5 feet.
All the sandstones of the section at Vinco are thicker than the average, but appear
to pinch out on the south side of the river between Vinco and Goodnight, so far
as it is possible to determine by surface exposures. At Goodnight they have
more than normal thickness. These belts of thickened sandstones extend nearly
north and south, with the region of very thin sandstones or mere traces of white
sajid and iron concretions marking their horizons between them.

THE PLAINS PROVINCE 93

"These long stretches of sandstone extend from just west of Pawnee, nearly
straight south to the vicinity of Shawnee, a distance of 60 miles on an air-line.
Wherever the region of shales west of this belt was crossed, as near Lela, west of
Stillwater, Goodnight, etc., another belt of thickened sandstones was found.
Another feature of this region that must not be lost sight of is the fact that the
lower horizons traced eastward grade out into normal light-colored beds of
marine origin, at least nearly as far south as Shawnee. Whether these great
masses of sand were thrown up as barriers along the southern tongue of the sea to
the north and northeast, or whether they represent river debouchures from the
mountains to the southward has not yet been determined. For a number of
reasons, some of which will follow, the writer is at present inclined to the opinion
that they are connected with rivers. With further work it appears now that the
question can be settled quite definitely and the origin of the sediments deter-
mined. If they were barriers, it would seem peculiar that the different layers
should thicken and thin so nearly simultaneously, while this is what would be
expected if the sand were brought down to mouths of rivers whose channels at
times extended well out across low fans, coastal plains, and shallow waters.

"In some places the deposition of sandstone is very irregular. Over some
areas a sandstone may be wanting and its place apparently filled with soft shales
that weather and slump very rapidly, forming great amphitheaters. In some
instances the sandstones occupy beds cut in the soft shales by currents of some
kind. * * *

"Many of the peculiarities which have been described occur in the northern
part of the State. Farther south, and especially farther west, they appear to be
more complicated. Another feature that was noted was that some of the beds
became quite coarse by the time the latitude of Shawnee was reached. Our
studies did not extend south of Shawnee.

"The fact that the stratigraphy is more regular in the same horizons in the
eastern part of the region studied than in their western extensions, as well as the
fact that the same formations contained limestones with marine fossils at their
eastern outcrop for some distance south of Pawnee, would seem to indicate that
an arm of the sea at Neva time extended south from the great northern area as
far as the Cimarron Ri\er, or a little beyond, but that its waters were extremely
shallow, if present, on the flats west of the 96° 45' meridian. The disappearance
of the fossils and the irregular and interrupted character of the stratification
seems to indicate the passing from marine conditions on the northeast to shallow
water or even subaerial conditions to the south and west. This would appear
to be the direct result of the influence of the Arbuckle Mountain region upon the
sedimentation of the time. Subaerial conditions continued near the mountains
and marine conditions beyond the influence of its fans."

B. THE LATE PALEOZOIC IN OKLAHOMA.

The relation of the vertebrate-bearing horizons of the Texas and Okla-
homa red beds to the Kansas limestones and shales is not and probably can
not be exactly determined from the very nature of the deposits. Some few
beds have been traced by Adams and Beede directly into Oklahoma, where
they shade off into red shales and sandstones, but they are not vertebrate-

94 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

bearing horizons. Beede^ gives the following statement, which is as near
as we may hope to come at the present time:

The Cowley County, Kansas, vertebrates which are very similar to
those from the Wichita formation in Texas come from 50 feet below the
Wreford limestone in the Neosho division of the Garrison formation.

The vertebrates from the Eddy locality in Kay County, Oklahoma,
which may be equivalent to either Wichita or Clear Fork forms, come from
a horizon as high as the base of the Wellington shales, 460 feet above the
Cowley County horizon.

The direct continuation of the Kansas Permo-Carboniferous beds into
eastern Oklahoma has been emphasized by Gould :^

"* * * while the Flint hills in Kansas consist almost entirely of limestones
and shales, still on the southern line of the State sandstones have already begun
to appear. To the south these conditions obtain more and more until the lime-
stone is entirely replaced by sandstone. * * * South of the State line the sand-
stones from the east and the red beds from the west begin to approach each other,
while the limestone ledges become thinner and thinner, and the flint less pro-
nounced.

"In general * * *, it may be observed that in going eastward from a red-
beds region toward the Carboniferous the sandstones and shales, which have
been of a deep brick-red color, become more and more brownish and grayish,
and finally lose entirely their characteristic hue and take on that of the older
formations. The lithology changes also. * * *

"The Marion and Wellington formations' narrow rapidly in northern Okla-
homa, and their place is taken by the red beds. Perhaps it is more correct to
state that the color of the shales appears to change to the south, and to become
red, while at the same time more of the red sandstone comes in, all tending to
change the formation in lithological appearance to that of typical red beds.

"A section of the Twin Hills, 7 miles east of Ingalls, eastern Oklahoma,
shows three ledges of limestone, the thickest of which is not more than 4 feet,
while all the rest of the rocks are either red shales or sandstones. Above these
limestones are ledges of grayish or red sandstones, which thicken to the south
and west, and, in the region between Stillwater and Orlando, assume the red tint
so common in the red beds. * * * The line of separation between the rocks of
these two ages [Carboniferous and Permian] must finally be drawn far out in
the red beds."

"* * * These formations^ [Marion and Wellington] narrow rapidly in northern
Oklahoma, and their place is taken by the red beds. Perhaps it is more correct
to state that the color of the shales changes to the south, becoming red, while
at the same time more of the red sandstone comes in, so that finally the formation

* Beede, J. VV., The Neva Limestone in Northern Oklahoma, with Remarks upon the

Correlation of the Vertebrate Fossil Beds of the State, Oklahoma Geological Survey
Bull. 21, p. 36, 1914.

' Gould, C. N., Notes on the Geology of Parts of the Seminole, Creek, Cherokee, and
Osage Nations, Amer. Jour. Sci., vol. 11, p. 185, 1901.

' Gould, C. N., General Geology of Oklahoma, Second Biannual Report Oklahoma Geo-
logical and Natural History Survey, p. 27, 1902.

* Gould, C. N., Geology and Water Resources of Oklahoma, U. S. Geological Survey,

Water Supply and Irrigation Paper No. 148, p. 35, 1905.

THE PLAINS PROVINCE 95

changes to typical red beds. On the State line, the distance from the Winfield
formation, the upper conspicuous limestone member, to the eastern outcrop of
the red beds is perhaps 30 miles; on the southern line of Kay County, Oklahoma,
it b not more than 15 miles, while farther south the line of separation can not
be determined, for the reason that the limestone disappears, and its place is
taken by red shales and sandstones. In southern Kansas there are three distinct
kinds of Permian rocks: First, the heavy limestones in eastern Cowley County
and along Walnut River: second, the bluish and gray clays and shales of the
Meirion and Wellington formations from Walnut River to western Sumner
County; and, third, the tj'pical red beds, consisting of red sandstones and clays
extending from this point nearly to the west line of the State. In eastern Okla-
homa, on the other hand, only red beds appear.

"Thus it is seen that the red beds extend farther east in Oklahoma than in
Kansas, and that the eastern limit of the red beds does not coincide with the line
of separation between the Pennsylvanian and the Permian. In other words,
the red color of the rocks, which has been thought characteristic of only the
Permian of the region, in fact transgresses far into the region of the Pennsylvanian
rocks. This means, of course, that the line of separation between the rocks of
these two epochs must finally be drawn far out in the red beds, and this the writer
has attempted to do.

"The horizon of the bone beds of Texas is an extended one, and probably does
not correspond to any one horizon in Kansas or Oklahoma, but to several of them.
Dumble's correlation of the Phacoceras dumbeli zone of the Wichita formation
with the Fort Riley limestond is probably about as near correct as we can state
at the present time."

Gould further states that the vertebrates from the Pittsburgh red shale
in Pennsylvania are from a horizon equal to the Oread limestone, i ,000 feet
above the Cowley Count>' horizon, and that the vertebrates from near
Danville, Illinois, are from a horizon which it is certain "that there is no
reason for supposing that the surrounding shales are as high stratigraphically
as the basal Permian of Kansas."

In a discussion of the upper Pennsylvanian rocks of eastern Oklahoma,
Gould^ and others have included deposits as high as the equivedent of the
Garrison formation. They show^ that the limestones and shales below the
Wreford become sandy toward the south and many of them disappear before
the Arkansas River is reached. Of the 10,000 to 12,000 feet of shales or
sandstones reaching from the Mississippian to the Permian (well into the
Permo-Carboniferous — Case) the shales greatly predominate. WTiile it is
apparent that the limestone thins out and disappears to the south, there
are present some limestones as far north as Bartlesville and Tulsa which
thin out to the north, having all the appearance of detached lenses. The
shales and sandstones constantly increase in thickness and frequently
coalesce. The two upper groups recognized in eastern Oklahoma are the:
Ralston, from the base of the Pawhuska to the base of the Wreford ; Sapulpa,

* Gould, C. H., D. VV. Ohern, and L. I. Hutchinson, Proposed Groups of Pennsylvanian of
Eastern Oklahoma, Research Bulletin State University of Oklahpma No. 3, 1 910.

96 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

from the base of the Lenapah to the Pawhuska, which equals the Deer
Creek and Hartford of Kansas. The details of the stratigraphy of the Red
Beds in Oklahoma are summarized in Publication 207 of the Carnegie
Institution, pages 51 and 56, and need not be repeated here.

C. THE LATE PALEOZOIC IN TEXAS AND NEW MEXICO.

A very full description of the stratigraphy of the Permo-Carboniferous
red beds of north central Texas was given in Publication 207 of the Carnegie
Institution, pages 19 to 41, and need not be repeated, but some repetition
of the accounts of the shading of the red beds into the limestones is necessary.
Cummins in 1897 described the gradual change from red shale to limestone
on the south and southeast side of the red beds:'

"By walking along the outcrop every foot of the way we were enabled to
note the gradual change in the lithological character of the bed. [Following a
prominent bed of the Albany northeastward we found] the limestone * * *
gradually changed in composition to a calcareous sandy clay, entirely destitute
of fossils * * *. North of the Brazos River, in the area heretofore designated
as the Wichita division in previous reports, the strata of the escarpment became
more and more composed of red clay, and the limestone beds less conspicuous.
The limestone gradually loses its limy nature."

Gordon records the same observations:^

"The red sandy shales and red standsones so conspicuous in the Wichita
Valley region were replaced southward in large part by blue shales, light-colored
sandstones, and limestones. In some places the transition from a sandstone to a
limestone was plainly seen. * * * It is the conclusion of the author that the
red beds of this region are the near-shore representatives of the Albany and the
decision as to their age will rest upon that of the latter."

In 1911, Gordon,' discussing the relation of the Albany to the Wichita,
says:

"When traced northward, the limestones of both the 'Albany' and the
Cisco formations diminish in thickness, while there is a corresponding increase in
the intervening beds of shale. In the case of the 'Albany,' the limestones show
also a change, becoming more earthy and irregular in their texture, and some of
the beds passing into gray indurated clays. The few limestones in the upper
part of the Cisco formation disappear entirely in the northern part of Young
County. Along with this change there is an increasing development of red clay,
alternating with blue. * * *

"At Fane Mountain, a low elevation in the southeastern corner of Throck-
morton County, is an outcropping of limestone characterized by an abundance
of Myalina permiana. These beds occur at intervals northward in eastern
Throckmorton County, and at Spring Creek in the northwestern corner of Young

' Cummins, W. P., The Texas Permian, Trans. Texas Acad. Sci., vol. 11, No. i, p. 95, 1897.
^ Gordon, C. H., The Red Beds of the Wichita-Brazos Region of North Texas (Abstract),

Science, vol. 29, p. 752, 1909.
' Gordon, C. H., The Wichita Formation of Northern Texas, Jour. Gaol., vol. 19, p. 118, 1911.

THE PLAINS PRO\aNCE 97

County they outcrop in the bank of the river about a mile from the post-office.
Here the beds show a local gradation into sandstone, suggesting near-shore condi-
tions of sedimentation. * * *

"Nowhere in the southern area, so far as observed, are there any indications
of unconformity. Notwithstanding the Hthological and faunal characteristics
which distinguish the 'Albany,' these beds apf)ear perfectly conformable with
the Cisco below and the Clear Fork above, nor is there within the formation any
indication of stratigraphic discordance. The change in the Hthological character
of the beds toward the north is evidently the result of differences in the conditions
of sedimentation. The character of this part of the formation suggests very
strongly its origin on a coastal plain, or river delta, to the south and west of which
lay the sea, in which were deposited the marine 'Albany' sediments. The inter-
relations of the two kinds of sediments suggest oscillation of the shore-line upon
a relatively wide coastal plain. These changes may be expljiined as the result
of oscillations of the land surface, or, possibly better, by the slow, but inter-
mittent, sinking of the coastal region."

In a later paper Gordon^ further discussed this point:

"A feature of importance in the Cisco formation, and one which it shares
with the next succeeding formation, is the series of changes observed as the
formation is traced northward along the strike. These changes relate both to
variation in lithologic character and to thickness of beds. In the Colorado
Valley, interstratified with the sandstones, clays, and conglomerates, are six or
more beds of limestone, each from 5 to 25 feet thick and all aggregating a thick-
ness of 100 to 150 feet. In the southern part of the Brazos Valley the calcareous
divisions are only about half as thick as they are farther south, jmd the clays show
a corresponding increase in thickness. In Young County the calcareous material
diminishes northward at an increased rate until, at the northern boundary of the
count>', the limestones have practically disappeared, and beyond that point they
are represented apparently by irregular nodular masses of earthy limestone in a
matrix of clay. With the thinning out of the limestones the shales and sand-
stones increase in thickness. In Stephens County, and farther south, the shales
are prevailingly blue and sandstones gray. Red Beds are dispersed sparingly
through the formation. The blues gradually give place to reds until in the
vicinity of Red Ri\er the red color dominates. In this part of the region the
rocks consist, for the most part, of red sandstones, clays, and sandy shales, with
a few beds of blue shale and bluish to grayish-white sandstones. Limestones
are conspicuously absent. * * *

"Beds of red clay make their app>earance south of Young County, but they
increase notably to the north, especially in the upp>er part of the formation, along
unth the diminution of the limestones, and they constitute the dominant feature
of the formation in eastern Clay and western Montague Counties."

On the western side of the Red Beds areas of Kansas, Oklahoma, and
Texas the sandstones and shales pass unchanged beneath the Mesozoic and
Tertiary' deposits of the Staked Plains. The western border of the Plains
Province of deposition lies close to Front Ranges of the Rocky Mountains.

The stratigraphy of the western border of the Red Beds in the States

' Gordon, C. H., U. S. Geological Sur%'ey, Water Supply and Irrigation Paper No. 317,

pp. 18-20, 1913.
8

98 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

mentioned is still unsettled. Fragments of vertebrates of Wichita or Clear
Fork age were collected by the author near Buffalo Gap, a few miles south
of Abilene, Texas, and west of the region between Abilene and San Angelo
no beds were encountered that could be correlated with the Double Moun-
tain. Ten miles east of Big Springs, Texas, the red shales and sandstones
carry Triassic fossils. From Mitchell County to as far north as Briscoe
County in Texas the uppermost Permo-Carboniferous beds are of Double
Mountain age; beyond that to the north, the position of the uppermost red
beds is less certain. In the Panhandle of Texas the exposures of red beds
in the Canadian River are probably the equivalent of the Greer and Quarter-
master (Whitehorse) of Oklahoma; the latter, at least, Gould considers as
entirely above the Cimmaron of Kansas. There is a strong suggestion of a
longer continuation of red-beds conditions in this region or an excessive
supply of material. Around the southern end of the Staked Plains the
exposure of the red beds is interrupted by the overlying Tertiary and
Mesozoic deposits. Between Big Springs and Midland the narrow strip
of red beds is apparently all Triassac, but there has been no conclusive
evidence of their age, either stratigraphic or paleontologic, reported. From
Midland to the Pecos River the exposed material is all Tertiary, but at the
bridge across the Pecos, on the Fort Stockton road, 20 miles east of Grand
Falls, and in the bed of the river near Grand Falls, there are exposures
of red shale and sandstone very similar in lithologic character to certain
phases of the Double Mountain formation of north central Texas. From
Grand Falls to Pecos only red sands and disintegrated red shale are exposed,
but on the east side of the Pecos River red beds again appear. The author
has been unable to obtain any evidence for the Permo-Carboniferous age of
these red beds other than their red color and assumed stratigraphic position.
They were first called Permian by Marcou in 1852, and the designation seems
to have clung for lack of any definite evidence to the contrary. Cummins,^
in 1891, said:

"We found no fossils in the beds west of the Plains, but as we had traced the
formation on both the eastern and northern sides there was no doubt as to its
being the same when we found it upon the west. * * * The strata lie un-
conformably on the Carboniferous, dipping at a small angle to the southeast."

The red beds are seen only in isolated patches from Pecos north nearly
to Roswell, as they crop out from below the Tertiary covering. The best
exposure is just east of Roswell. The following description of the section
is given by Fisher:^

* Carnegie Inst. Wash., Year Book, p. 374, 1916.

' Cummins, W. F., Notes on the Geology of the Country West of the Plains, Third Annual
Report Geological Survey of Texas, p. 212, 1891.

• Fisher, C. A., Preliminary Report on the Geology and Underground Waters of the Roswell

Artesian Area, New Mexico, Water Supply and Irrigation Paper No. 158, U. S.
Geological Survey, p. 6, 1906.

THE PLAINS PROVINCE 99

"The rocks of the [Roswell] district comprise limestone, sandstone, clay, and
gypsum which are believed to be of Permian age. * * * The so-called Permian
series of this district consists of an upper red bed member of gypsum, red sand,
limestone, and clay 600 feet thick, forming the high bluffs along the east side of
Pecos River and underlying the recent deposits of Pecos Valley, and a lower
member of massive limestone, clay, and gypsum of undetermined thickness, which
constitutes high rugged slopes to the west. * * *

"Permian (?) Series.

"Red-bed division. — These rocks consist of alternating beds of gypsum, red
sand, and clay, with an occasional layer of dark-gray, compact limestone. The
gypsum predominates and usually occurs in beds about 10 feet thick. It is
often found, however, in thinner layers, interbedded with clay and limestone.
The red beds are provisionally placed in the Permian, although no fossils have
been found in them. * * * The upper part of the beds is well exposed in the
bluffs along the east side of Pecos River, where a number of sections have been
measured. * * *"

A typical section of this bluff is as follows:

East of Roswell: Feet.

Alternating layers of gypsum and red sand, with an occasional layer of limestone ... 50

White gypsum 6

Red sand 6

WTiite, thin-bedded gypsum 10

Red sandstone containing thin layers of limestone 24

WTiite gypsum 5

Red sand 13

Gypsum 10

Red sand 3

Gypsum 8

Red sand 8

Gypsum 4

Greenish-gray sandstone 25

Gypsum 6

Total 178

"Limestone division. — The massive limestone beds underlying the so-called
Permian red beds of this region consist mainly of gray, compact limestone, with
layers of soft sandstone, clay, and gypsum. In the upper part the limestone is
more or less thin-bedded and porous, and contains many sandy layers. * * *
Limestone outcrops along the west side of the district, and farther to the west
forms high, rugged plateaus, extending toward the mountains. Fossils are not
abundant in the formation, but in one locality northwest of Roswell a number were
collected, which consisted mainly of Schizodus and Pleurophorus, preserved as casts.
According to Doctor Girty, the fauna and lithology of these specimens suggest the
highest Cai boniferous beds or the Permian of the Mississippi Valley in Texas."

Beede^ has maintained that the red beds of the eastern side of the
Pecos Valley are equivalent to, or a continuation of, the upper red beds of
Texas and Oklahoma. He shows that the Capitan and Delaware limestones
shade north and east into red sandstones and shales, which he regards as a

* Beede, J. W., The Correlation of the Guadalupian and the Kansas Sections, Amer. Jour.
Sci., vol. XXX, p. 131, 1910.

100

ENVIRONMENT OF VERTEBRATE LIFE, ETC.

GUADAUUPE
MTS.

Whitehorse Beds

re la-ware
Mountain

Hueco

PANHANDLE-
KATJSAS

Triassic TriassicC —

STAKED
PLAINS

Vlfoodward

Enid

Wellin|lon
Marion
3enesIV

Greer

Series in

geriea 11

Series 1

Mississippia n

continuation of those to the east, and maintains that the conditions which
determined this change separated the fauna of Capitan Hmestone from that
of the Quartermaster (Whitehorse) of Texas and Oklahoma. He says :

" If the conclusions reached above are correct, it leads at once to the correla-
tion of the Kansas and Guadalupian sections. If we use the Whitehorse sand-
stone, probably the equivalent of the beds in contact with the Guadalupian
limestone near Carlsbad, as a common basis of correlation of the two sections,
we attain the result shown in the accompanying dia-
gram. [Fig. 2.] Disregarding their actual faunal
relationships and comparing them as to their thick-
ness, the strata of the two sections compare as fol-
lows, the figures of the Guadalupian rocks being
approximations :

"In southern New Mexico we have some 4,500
feet of the Guadalupian series, composed of 2,100
feet of Capitan and overlying limestones, and 2,400
feet of the Delaware Mountain formation, composed
of limestones and sandstones overlying 5,000 feet
of Hueco limestones. Beginning at the same horizon
in Kansas, we have the remainder of the Red Beds,
the lighter Permian and the Pennsylvanian, aggre-
gating about 4,500 feet of strata, composed of lime-
stone shales and sandstone. So far as mere thick-
ness is concerned, it leaves the base of the Delaware
Mountain formation about on the level with the
Cherokee shales (as exhibited in Kansas). The
horizon of the base of the Delaware Mountain for-
mation in the Kansas section, interpreted upon its
fauna, or actual time equivalency, may be a very
different matter. The base of the Capitan falls
near the bottom of the Elmdale formation strati-
graphically, which is probably not far from its cor-
rect faunal correlation as well. The paleontological
comparisons are yet to be worked out. The unconformity above the Capitan
limestone, and locally even in the Delaware Mountain formation, the Capitan
having been carried away, is not taken into account in making these comparisons.
It is probable that it diminishes rapidly to the northward, where it is of less
consequence.

"One of the most interesting features of the Guadalupian fauna is its isolation.
As has been stated by Girty, the fauna is a unique one, and, as a unit, is now
known from no other part of the western hemisphere. At first thought it seems
peculiar that more of its members were not distributed over the adjacent regions
where contemporaneous strata occur. Their absence in such rocks has been a
serious difficulty in any attempt to correlate them with other American faunas.

"In the first place, the lower red beds lying to the eastward, with which the
Guadalupian limestones are probably contemporaneous, are believed by some
to be to a considerable extent of subaerial origin, while the temporary seas that
occupied portions of it from time to time were too concentrated in salt content
for normal marine faunas. So far as my collecting in the typical Capitan limestone

Fig. 2. — Diagram from Beede,
showing his idea of the rela-
tions of the beds in the Gua-
dalupe Mountains to those
in Texas and Kansas.

THE PLAINS PROVINXE 101

goes, the fossils were abundant only in the purer limestones, and were very rare,
or wanting in what appeared to be the dolomitic portions of it. These limestones
occur in the Apache Mountains and at Guadalupe Point, but appear to be want-
ing, as does the fauna, north of the Texas line; the only exceptions noted were
Fusulina elongata and one or two other species in Dog Canyon and Sitting Bull
Cannon. From this it will be seen that the fauna was closed off on the north by
untoward conditions and on the east by the red-bed sedimentation, which con-
stituted a barrier. No other bcirrier is known.

"Two other considerations must be taken into account. First, that the
Permian facies of this fauna may be an abnormally early precursor of the Permian
faunas de\eloped in an isolated basin. Such an occurrence of Permian forms is
known in Kansas well down in deposits of Pennsylvanian age. However, the
variety' and richness of the Guadalupian fauna, which possess such a young
appearance, seem to me to argue against this hypothesis. Second, the other
possibility is that the fauna is no older than it apf>ears, and that it developed
normally' with little outside connection, as did the Kansas Permian fauna. The
same features as before would have controlled its isolation. Much of the red
beds being almost a land surface a considerable part of the time — if we accept
the subaerial origin of a large part of the deposits — aggradation may have but
slightly overbalanced degradation, and they may have accumulated slowly for
that class of sediments. Thus, though disturbances raised the southern pju-t of
the Guadalupe limestones above sea-level, and permitted their partial removal
and the subsequent deposition of the upf>er red beds up>on the eroded surface,
the fauna ma}' well have been an early Permian fauna. Until further data are
at hand I am much inclined to this latter hypothesis. The fact that several
hundred feet of the Kansas Permian deposits grade off into typical red beds in a
ver>' short distance in Oklahoma is suggestive of possible conditions east of the
Guadalupes. If such were the case, we would expect the Guadalupian faunas to
cease as abruptly' upon the strata changing to the red beds, as the Kansas faunas
do upon entering the Oklahoma red beds.

"At the same time, owdng to the very nature of the origin of the red beds,
their extreme southwestern part vaay have been deposited slightly later than
the main mass farther to the north and east. However, this is regarded more in
the nature of a possibilit>' than a probability.

"The accompanying map [fig. 3] indicates the probable relationship of the
marine areas during the Council Grove-Chase and Guadalupian time in the
immediate area under consideration. No attempt is made to show the full
extent of deposits laid down at this time. The full lines indicate marine condi-
tions and the lines alternating with stippled ones continental-marine deposition.
The extent to which the two factors contributed to the formation of the red
beds is at present unknown. The area of marine conditions in Central Texas is
to represent the Albany sea."

This position has been contested by Girt>^ as is shown below (page 144),
and the difficulty of such a correlation is quickly apparent when the relation
of the Pecos beds to the Delaware and Capitan limestone is understood.
As has been shown by Richardson and others, the Castile g>'psum and the
Rustler limestone lie beneath the red beds of the Pecos Valley, and there is
every reason to believe that these are equivalent to the uppermost red beds
of Texas or the Triassic red beds; probably both are represented in the better

102

ENVIRONMENT OF VERTEBRATE LIFE, ETC,

sections. The Rustler limestone lies upon an eroded surface of the Delaware
limestone and very probably this eroded surface was at one time covered
by Capitan limestone.

Fig. 3. — Map showing Beede's idea on the paleogeography of red beds in
the southern part pf the Plains Province (after Beede).

In his Review of the Geology of Texas, Udden^ describes the formations
concerned as follows:

"Delaware Mountain Formation.

"This formation is composed of an alternation of gray and bluish limestone
with white and brown sandstone. At the lower part is a blue-black thin-bedded
limestone, shaly in part. The base is not exposed. Toward the north of the
Delaware Mountains the formation becomes more sandy; toward the south the
limestone increases in amount. In the Apache Mountains the formations consist
entirely of massive whitish-gray limestone. As the base of the formation is
unknown, the entire thickness can not be determined, but is at least 2,200 feet.
The Delaware formation forms a broad zone composing the Delaware, the

» Udden, J. A., Bull. University of Texas No. 44, p. 54, 1916.

THE PLAINS PROVINCE 103

Apache Mountains, the lower part of the Guadalupe Mountains, and part of
the Wylie IMountains. It extends into New Mexico.

"Capitan Ldiestone.

' ' This formation is composed of a massive white limestone remarkably homo-
geneous in appearance. The entire thickness can not be determined, but it is
at least 1,700 feet. It is known only in the Guadalupe Mountains and extends
far into New Mexico.

"Castile Gypsum.

"This formation is in great part composed of a massive, white, granular
gypsum, but interbedded with it are thin beds of gray and yellow limestone and
dolomite, as well as thicker beds of the same rock and considerable masses of
gray, red, and green shales and marls. The thickness of this formation is not
exactly known, but two deep wells near Rustler Springs show that it can not be
less than 1,000 feet. The Castile gypsum forms a band about 15 miles broad,
west of the hills comp>osed of the Rustler limestone; toward the north the Castile
gypsum is found also east of the Rustler Hills, so that the breadth of the zone
increases to about 30 miles near the boundary of New Mexico. Some isolated
exposures are found on the west side of the Delawcire Mountains. As far as
known, the Castile gypsum rests everywhere unconformably on the Delaware
formation. Some shale in this formation is sulphur-bearing in Culberson County.

"RUSTLEK FORIIATION.

"Compact, fine-textured, gray dolomitic limestone and dolomite, generally
quite heavy-bedded, compose this formation. At the base there is in most
places a considerable mass of light pink or yellowish brecdated limestone. In
the northern part of the region some yellow sandstone alternating with limestone
is developed below the brecciated limestone. The thickness of the Rustler forma-
tion has not been determined, but it must be at least several hundred feet. The
Rustler formation appears in a series of low hills extending from a point about
12 miles north of Kent to the boundary of New Mexico."

It has been shown by Case^ that the red beds of upper Permian age in
western Texas do not extend across eastern New Mexico in the latitude of
Tucumcari and Las Vegas, and he has pointed out that the beds of Texas
and New Mexico that far north are parts of distinct provinces, a fact borne
out by his discover^' of vertebrates similar to those occurring in Rio Arriba
County, New Mexico, near Socorro. It would, then, appear that the beds of
western Oklahoma (WTiitehorse) are in reality* above the Capitan limestone.
The difference in stratigraphic position, however, need not be alarming, as
in such beds the rate of accumulation might be at times exceedingly rapid.

The suggestion of Beede that the elevation and erosion of the Delaware
and Capitan limestones was an occurrence quite similar to that occurring
in eastern central and eastern North America in the same general time
interval is very pertinent here.

• Case, E. C, The Red Beds Between Wichita Falls, Texas, and Las Vegas, New Mexico,
in Relation to Their Vertebrate Fauna, Jour. Geol., vol. xxii, p. 243, 1914; Carnegie
Inst. Wash. Pub. 207, p. 61, 1915.

104 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Descriptions are given by Udden^ of the beds of certain deposits of
Permo-Carboniferous age in Trans-Pecos Texas:

"The Shafter Region, Presidio County,
"the cibolo beds.

"The rocks which represent the Permian have been called by Udden the
Cibolo beds, and they have been subdivided by the same author from below to
above in: Transition beds. Lower Brecciated Zone, Zone of Sponge Spicules,
Thin-bedded Zone, and Yellow Limestone.

"Transition Beds. — Gray marly shale with lenticular ledges of organic and
siliceous sand. Their thickness is about lOO feet.

"Lower Brecciated Zone. — Grayish-white limestone in heavy ledges often
thoroughly brecciated. The thickness is about 133 feet.

"Zone of Sponge Spicules. — In the lower part, this consists of thinner-bedded
limestone. Above it becomes siliceous and changes into pure sandstone. The
thickness of this bed is 85 feet.

" Thin-bedded Zone. — Dark, evenly bedded and compact limestone, including
some sandy strata. The limestone contains cherty material which weathers
out in rusty edges of plates of irregular shape, or porous spherical shells and
nodules. Much of the rock is bedded in uniformly thin ledges; occasionally the
ledges thicken lenticularly. Thickness, about 470 feet.

" Yellow Limestone. — Hard, yellow, siliceous, and dolomitic limestone, showing
bedding planes only in the lower part, while higher up the stratification becomes
indistinct. Thickness, about 650 feet.

"This series has been observed on Cibolo Creek near the Chinati Mountains
west of Shafter.

"The Marathon Region.

"The Permian is very well developed in the Glass Mountains. It has been
subdivided by Udden in four formations, which are, from below to above, the
Leonard, Word, Vidrio, and Gilliam. The lower formations are found in the
southern and southeastern hills, while the upper ones — the Vidrio and Gilliam
formations — occupy the center and the whole northern slope. Towards the
southwest the continuation of this Permian is found in the Altuda Mountain
and south of it, in the Ord Mountain Range. Toward the north we find an
isolated outlier in the Sierra Madre. The highest parts of the Permian are
covered unconformably by the Comanchean Cretaceous. * * *.

" WORD formation

"In the upper part this is composed of thin- and thick-bedded gray and
yellow to reddish limestone, in part dolomitic, containing chert concretions with
some interbedded strata of sandstone (about 380 feet) ; below this we find some
120 feet of yellow sandstone, 'in part laminated. The lowest part of this formation
consists of 120 feet of heavy-bedded gray limestone, with chert concretions.
The entire thickness of this formation is approximately 600 feet.

"vidrio formation.

"This series is composed of a very uniform, dark to light gray, dolomitic
limestone, or dolomite, with very few layers of pure limestone. The dolomite

^ Udden, J. A., Review of the Geology of Texas, Bull. University of Texas No. 44, p. 50, 1916.

THE PLAINS PROVINCE

105

contains considerable chert in irregular form. In the uppermost part we find
one or two beds of reddish-brown sandstone about 4 feet thick. The entire
thickness of this formation is about 2,000 feet.

GnXIAM FORliATION.

"This series is composed of gray, light-colored, and reddish limestone and
dolomite; both are frequently brecciated. In the upper part the rock is nearly
massive, or, at least, bedding planes are very dim. In the middle the limestone
shows thick lenticular beds, while in the lower part it is decidedly thin-bedded.
At the base we find thicker layers of reddish dolomite alternating with thinly
laminated layers of the same rock and with thin strata of yellowish marly sand-
stone. The thickness of this enormous mass is 2,500 feet, in Gilliam Canyon."

The Permo-Carboniferous appears in other localities near Ord Mountain,
but the general character of the deposits are the same as those already
described. (See page 53 of Udden's report.)

Correlation table of the Texas Permian.
From Udden's Rei»rt, page s6-

Shafter region.

Ord Mountain and
vicinity.

Glass Mountains.

Delaware-Guadalupe
Mountains.

Yellow limestone.

I. Vidrio formation.

Gilliam formation.
Vidrio formation.

2. Sandstones and lime-
stones.

Thin-bedded zone.
Zone of sponge spicules.
Lower brecciated zone.
Transition beds.

3. Shales, sandstones
and limestones.

4. Limestone conglomer-
ate and thin-bedded
or flaggy limestone.

Word formation.

Leonard formation.

PennsyK'anian (Alta and
and Cieneguita beds).

Pennsyh-anian.

' Gaptank formation.

Rustler
limestone.

Capitan

limestone.
Castile ,
gypsum. ; Delaware

/ formation.

Hueco formation.

Two papers in the University of Texas Bulletin^ give additional informa-
tion concerning the late Paleozoic depx^sits of the western part of Texas.
Udden adds three formations to the list quoted above. His series in the
Glass Mountains is as follows:

Permian {?) Permo-Carboniferous:

Tessey 1,400

Gilliam 740

Vidrio 1,700

Word (?) 1,400

Leonard 1,800

Hess 2,100

Wolfcamp 500

Pennsylca nia n:

Gaptank 2,000

' Udden, J. A., Notes on the geology of the Glass Mountains, Univ. Texas Bull., No. 1753,
1917-
Baker, C. L., and W. F. Bowman, Geologic exploration of the southeastern Front Range
of Trans-Pecos Texas. Idem.

106 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

"The Wolfcamp consists mostly of shales which vary in color from almost
black to gray and greenish-gray. Interbedded with this shale are several
layers of limestones which are cemented shell breccias, in places conglomer-
atic. There are also layers of calcareous sandstones." There is possible
an unconformity between the Gaptank and the Wolfcamp, and the fauna
of cephalopods in the upper formation indicates a decided break in the
sequence of life between the two.

The Hess formation consists of limestones, largely oolitic, shales, sand-
stones and a minor amount of conglomerate. "The color of this limestone
is mostly light gray. The individual beds have a uniform development and
can be traced for comparatively long distances. It can also be said that
the general aspect of these limestones resembles that of the Hueco formation
farther west in the State, but sufficient collections of fossils have not been
made from this formation for the purpose of verifying such a correlation.
In its upper part, fossils are quite plentiful in certain layers. It appears
that the dolomitization of the limestones in this formation has proceeded
at quite unequal rates in different places. At the west end of the escarp-
ment, dolomitization is quite general. As we go away from the disturbance
near the igneous intrusions extending northeast from the Iron Mountain,
dolomitic layers appear less frequently than at the west end. The sand-
stones and shales of this formation are present mostly in the lower four
hundred feet. Most of the shale is bluish-light gray in color. The sand-
stones are usually free from limy material, have an open texture, and are
moderately fine grained. In places they show cross-bedding. The basal
conglomerate of the Hess consists mostly of limestone boulders, but it also
contains some boulders of flint and other quartz. All the underlying forma-
tions are represented. It varies from ten to forty feet in thickness." The
Hess is separated from the Wolfcamp and Gaptank by a considerable
unconformity. In his paper Udden says: "It is believed that the Leonard
is to be correlated with the Clear Fork in the west-central part of the State.
Perhaps it includes also the basal part of the Double Mountain, and the
upper part of the Albany limestones. It certainly also contains many of
the forms noted in the Delaware formation of Girty." He further says:
"Apparently there is no doubt that the Word formation belongs to the
Delaware deposits of Girty in the Guadalupe Mountains. It also represents
the main part of the Double Mountain in central Texas."

The Tessey formation is composed mostly of unstratified dolomite re-
sembling the Vidrio. " It is believed that the Vidrio, the Gilliam and the
Tessey formations are in part the equivalents of the Capitan limestone in
the Guadalupe Mountains. Together they have a thickness of 3,800 feet,
which is more than twice the known thickness of the Capitan limestone.
The three formations are conformable and dip to the northwest with an
angle of about eight degrees."

THE PLAINS PROVINCE 107

Baker and Bowen, describing the Front Range west of the Glass Moun-
tains, say that after the Gaptank was deposited the sea "withdrew from the
region, subaerial erosion followed, and a resubmergence brought about the
deposition of some 8,000 feet of Permo-Carboniferous sediments. This
epoch of marine deposition was twice interrupted by uplift which brought
about renewed erosion, as is indicated by two unconformities and basal
conglomerates in the Permo-Carboniferous series."

Both of these papers give detailed accounts and sections with lists of
fossils showing the marine conditions in the Trans-Pecos Texas region during
what the authors call the Permian (?) or Permo-Carboniferous time. In
the opinion of the author of this paper the correlation of the Leonard with
the Clear Fork and the Word with the Double Mountain must await more
proof than is contained in Udden's paper. There is the possibility that they
were formed in the same interval of time, or the Clear Fork and Double
Mountain may have been formed in the long period represented by the
erosion interval described by Baker and Bowen between the Gaptank and
the series of limestones above it.

D. THE LATE PALEOZOIC IN THE NORTHERN PART OF THE PLAINS
PROVINCE AND ON THE EASTERN FRONT OF THE ROCKY

MOUNTAINS.

On page 62 of Publication 207 of the Carnegie Institution the author
has given a resume of the general lie of the Permo-Carboniferous Red Beds.
As was shown in that publication, there is a merging of limestone into red
shales and sandstones in the northwestern portion of the Plains Province
similar to the merging which occurs in Kansas, Oklahoma, and Texas.

The red beds in the North and on the slopes of the eastern face of the
Rocky Mountains have never been placed exactly in the geological column.
This is in part due to the lack of determinant fossils and in part due to the
conditions of deposition, terrestrial deposition prevailing and producing
overlapping and interlocking lenses of relatively small areal extent. More-
over, there is little doubt that the "red beds conditions" extended in time
from late Pennsylvanian into, if not through, Triassic time. It is, so far,
impossible to correlate any of these beds with the more definitely determined
beds in Oklahoma or Texas, but there is no question that at approximately
equivalent intervals of time similar results were produced on the borders,
at least, of the Plains Province by similar conditions.

The age of the red sandstone and shale has been variously reported by
different authors. A portion of the discussion quoted in Publication No.
207 of the Carnegie Institution is repeated here to show the attitude of
various writers:

"In Professional Paper 32, United States Geological Survey, Darton discusses
the character of the Red Beds of the Front Range of the Rocky Mountains. The

108 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

upper Carboniferous limestone is found in the northern part of the Front Range
near the Wyoming line and in the Culebra Range it appears to merge into the
Fountain Red Beds, which he believes to be the exact equivalent of the Lower
Wyoming of Eldridge and the Badito of Hills, and to represent the Amsden
formation and overlying Tensleep sandstone of the Bighorn Mountains and the
Minnelusa formation of the Black Hills. The lower Red Beds of the Rocky
Mountain Front Range have yielded no fossils, but undoubtedly merge into
limestones both on the north and the south and can be correlated with formations
in the Black Hills and the Bighorn Mountains. Darton says further:

'"Throughout the Black Hills, the Bighorns, and much of the region to the
south the upper Carboniferous and Red Bed series presents a general succession
as follows, beginning at the top: A thick mass of gypsiferous, red, sandy shales;
a thin mass of thin-bedded limestone; a thin mass of red sandy shales; a thick,
hard, light-colored, fine-grained sandstone; and, at the base, limestones and sand-
stones giving place to sandstones and conglomerates, the basal series lying uncon-
formably upon the Mississippian limestones, on Cambrian, or on old granites
and schist.

"'Near the Colorado- Wyoming State line the upper Carboniferous limestone
may be seen to merge into red sandstones, apparently by the expansion of included
reddish sandy layers observed northwest of Cheyenne and a corresponding
thinning of the limestones. A mass of red sandstones and conglomerates, which
lies at the base of the limestones for some distance, is seen also to thicken gradually
to the south.

"'The name "Fountain formation" has been used to comprise all of the red
beds in the region northeast of Canyon and southwest of Pueblo, and if, as I
believe, the Chugwater (upper Wyoming) formation thins out a short distance
south of the Garden of the Gods, the Fountain formation corresponds in the main
to the lower Wyoming, and is the product of similar conditions at the same
geological epoch. I do not see the slightest reason for supposing that the two
formations are not equivalent.

"'The character of the beds northwest of Pueblo and in the Garden of the
Gods region is precisely the same as in the district west and north of Denver,
and although I made special search I could find no evidence of overlaps or uncon-
formities of any kind within the great uniform mass of red grit deposits.

"'The upper and lower Wyoming are very distinct from each other from the
Garden of the Gods north to the State line, as recognized by the geologists of
the Hayden survey and clearly set forth in the Denver monograph, where the
terms "lower Wyoming" and "upper Wyoming" were introduced. The upper
Wyoming consists mainly of fine-grained sediments extending from the "creamy
sandstone," which I believe to be the equivalent of the Tensleep, to the base
of the Morrison formation. It consists mainly of bright-red shales, always
with a thin limestone layer or series toward its base, and from Platte Canyon
northward with a massive pinkish sandstone at its top. The included limestone
is believed to represent the Minnekahta horizon of the Black Hills and other
regions, indicating a short but widespread interval of limestone deposition at
this epoch in the West. The few fossils found in this limestone unfortunately
do not settle its age, but there appears to be but little doubt that its representative
in the Black Hills is Permian. The overlying red shales, with gypsum, in northern
Colorado may be Permian or Triassic, for the fossils in the limestones which occur
near the top of the extension of this series into the Bighorn uplift do not indicate
whether the beds are Paleozoic or Mesozoic.

THE PLAINS PROVINCE 109

"'The ChugTxater formation (upper Wyoming Red Beds) is only 140 feet
thick at the Garden of the Gods and appears to thin out and disappear a few
miles south, bringing the Fountain formation into contact with the Morrison,
a relation due either to non-deposition of the Chugwater beds or to their removal
by erosion in pre-Morrison rimes. As it is, the hiatus probably represents part
of the later Carboniferous, the Permian, the Triassic, and all of the Jurassic periods.
South of the Arkansas River some of the Chugwater beds probably appear again,
although at present their identity is not established.

'"The Badito formation of Hills appears to be simply the Fountain formation
of Cross and Gilbert. The Sangre de Cristo formation to which Hills refers in
the Walsenburg folio appears to represent a great development of Fountain (or
lower Wyoming) deposits. It is stated that remains of an upper Carboniferous
fauna and flora occur in this formation, which is added e\-idence as to the age
of the lower Red Beds (Fountain-lower Wyoming) series. These beds overlie or
mei^e into the basal limestone series on the ejistem slope of the Sangre de Cristo
(Culebra) Range, in which Mr. Willis T. Lee has discovered an extensive upper
Carboniferous (Pennsylvanian) fauna.

'"The red beds revealed in the canyons of the southeastern Colorado can
not be classified with certainty from the present e^^dence. On Pulsatory River
and Muddy Creek the principal body of red beds is separated from the Morrison
formation by gypsum or gypsiferous shales, strongly suggestive of the Chugwater
(upper Wyoming) formation. It was immediately under this gypsum in Purga-
tory Canyon that I found the shoulder bone of a supposed belodont. Mr. Willis
T. Lee has traced the Red Beds farther south into northeast New Mexico, where
the gA'psiferous horizon gives place to a massive sandstone, termed the Exeter
sandstone, constituting the summit of the Red Beds, a member which may repre-
sent the distinctive top sandstone of the Chugwater formation in northern
Colorado and in southern Wyoming. The sandstone is prominent in the Two
Buttes uplift, constituting the summit of the Red Beds, and is underlain by red
shales, which contain a thin bed of limestone, noted by Mr. Gilbert, strikingly
like the Minnekahta horizon. I have not made observations on the Red Beds
in Kansas and do not feel that a comparison of the published statements with
my observations in the region north and west should add in the correlation.'

"Girtyi regards the Fountain formation as Pennsylvanian. Henderson* re-
garded the lower part as Mississippian jmd the upper as Pennsylvanian, and
Da\4d White,' from the e\-idence of fossil plants, would place it in Pottsvalle time.

"In Professional Paper 53, L^nited States Geological Survey, Darton speaks
of the Red Beds of Colorado. He says that in southern Colorado the Red Beds
lie on an irregular surface of granite, except in certain embayments, as the ones
at Manitou and Can3'on Cit>', where lower Paleozoic rocks inter\-ene. The Red
Beds have been found to be an extension of the Red Beds underlying the Carbon-
iferous limestone in southeastern Wyoming and of the Permian and overlying
Red Beds of Kansas.* The Red Beds of this region he considers divisible into
three parts: (i) the Fountain or lower Wyoming (the lowest), consisting of coarse
red grits which he found to represent the upper Carboniferous limestone of

* Girty, U. S. Geological Sur\ey, Professional Paper No. 71, pp. 369-370.

* Henderson, Jour. Geol., vol. 16. pp. 491-492, 1908.

•White, Da\id, U. S. Geological Survey, Professional Paper No. 71, p. 370.

* The Red Beds of Kansas have since been shown to be continuous, in part at least, with

the Permian limestones and not to overlie them. — E. C. Case.

no ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Wyoming; (2) the Tensleep sandstone, traced as far south as the Manitou em-
bayment; (3) the gypsiferous red shale and sandstone of the Chugwater, which
represents the red beds of the Black Hills and Wyoming. The lower part of
the Chugwater he considers as Permian, the upper part as Triassic or Permian.

"Henderson' in 1908 gave an account of the Permo-Triassic (?) of the foot-
hills formations of northern Colorado. He distinguishes the upper part of the
Wyoming as partly Permian.

'"Lykins formation. — Conformably overlying the Lyons is a series of varie-
gated, mostly thin-bedded sandstones and shales, rather friable, chiefly deep red
in color, with thin limestone bands, the upper part usually gypsiferous. In the
Boulder district Fenneman names these beds the Lykins formation. It is the
exact equivalent of the upper Wyoming of Emmons in the Denver Basin and
the Chugwater of Darton in northern Colorado. In the Denver Basin monograph
it is given a thickness of 485 to 585 feet; Fenneman makes it 800 feet in Four-mile
Canyon, north of Boulder, and Darton gives it a thickness of 380 feet at Lyons
and 520 feet at Owl Canyon. Though it varies greatly in thickness and in strati-
graphic details, its general characters are constant throughout the region. As a
whole the formation is non-resistant, the greater part being concealed by the
debris in the lateral north-south valleys caused by its destruction.

"'From Owl Canyon to the Little Thompson I have mapped as part of the
Lykins a more resistant sandstone, strongly cross-bedded, which forms a ridge
in the valley and which sometimes extends nearly to the top of the east slope of
the Lyons escarpment. It is difficult to distinguish from the Lyons sandstone,
and should perhaps be assigned to that formation, but is uniformly separated
from the latter everywhere north of the Little Thompson by strata lithologically
resembling the Lykins. In approaching Little Thompson Canyon these inter-
vening beds rapidly play out, bringing the sandstone which is mapped as Lykins
into contact with the Lyons and -making the former the crest of the escarpment,
almost covering the latter. Thence southward it is doubtful if the two sandstones
can be recognized as distinct formations, and nowhere have I found a noticeable
unconformity. As the two sandstones after coalescing form an almost vertical
escarpment, if they are distinct it is practically impossible to represent the Lyons
on the map, yet northward they are quite distinct. The one which is mapped
as Lykins in the northern region passes beneath the "crinkled" sandstone of
Fenneman's report, which is but a few feet above the Lyons north of Boulder.
This problem is worthy of future investigation.

"'In some places certain strata of the Lykins are very massive, though soft,
and portions of the formation are locally calcareous, in addition to distinct
limestone bands.

" ' In the absence of paleontological evidence this formation has been usually
assigned to Triassic-Jurassic age. It seems quite likely, however, that the base
of the Lykins may represent Permian time, as the immediately underlying Lyons
is upper Carboniferous. The upper part of the Lykins is probably Triassic or
Jurassic, as it is overlaid by known Jurassic in northern Colorado, though it is
possible that part of the Jurassic and Triassic is represented by the general un-
conformity between the Lykins and the Morrison.'

"Butters^ in 1913 gave a very detailed account of the 'Permo-Carboniferous'

• Henderson, First Annual Report Geological Survey of Colorado, p. 168, 1908.
' Butters, R. M., Permian or Permo-Carboniferous of the Eastern Foothills of the Rocky
Mountains in Colorado, Colorado Geological Survey, Bull. 5, pt. 2, p. 65, 1913.

THE PLAINS PROVINCE 111

of the eastern foot-hills of the Rocky Mountains in Colorado. After detailing the
conditions in the various sections from the north line of Colorado south he says:

'"Overlying this (the Ingleside) is the Lykins formation, and at one horizon
about 200 feet from the base, at Heygood and Box Elder Canyons, Bellerophon
crassus and Myalina subquadrata were found. The same species are found in the
Fountain and Ingleside below. On this evidence, together with the fact that
there is no angular unconformity, and no marked difference of lithological char-
acter, this basal portion of the Lykins is assigned to the Pennsylvanian period.
On the northern slope of Table Alountain, Larimer County, 40 to 50 feet higher
than the fossiliferous stratum, and separated from it by a g>'psiferous series, another
fossiliferous stratum occurs. This is probably more than 300 feet from the top of
the Lykins formation and seems to be in the same stratigraphic position as the fos-
siliferous beds near Stout, and also those in the crinkled sandstone near Perry Park.
At Stout, and also at Table Mountain, the "crinkly" structure is not present.'

" ' Correlations. — The correlation of the Fountain, Ingleside, Lyons, and Lykins
along the foothills from the line to Colorado Springs is a question of recognizing
the same formation under different names. * * * Thus the Fountain of Fenneman
in the Boulder quadrangle is equivalent to the lower part of the Fountain of
Cross in the Pikes Peak area. The Fountain, Ingleside, and Lyons together
are equivalent to the lower \\'yoniing of the Denver Basin area. The Lykins
is equivalent to the Chugwater of Darton and the upper Wyoming of Emmons.
The upper portion of the Fountain and the Ingleside together are equivalent
to Darton's Casper formation. The Lyons is equivalent to the Creamy sandstone
of the Denver Basin area, but Darton's Tensleep is not the equivalent of the
Lyons and the Creamy sandstone. It is a lower horizon, and can be correlated
only with the lower portion of the Lyons, and also the Ingleside. An explanation
of this requires a description of the conditions in northern Colorado. This has
been made under ' ' Formation names. ' ' Darton's Tensleep in Colorado is probably
in part equivalent to the sandstone-limestone series; that is, the Ingleside series.

'"Owing to the absence of fossil evidence in the Badito formation, and from
the fact that it is separated so widely from any recognized Fountain exposures,
it has not been definitely correlated with the Fountain. Lithologically they are
very similar, and the Badito overlies pre-Cambrian rocks unconformably, bearing
about the same relation to the overlying formations as does the Fountciin. On
these grounds they are at least approximately in the same horizon.

"'The Cutler formation is defined as that portion of the "Red Beds" lying
above the Rico, where that is present, or otherwise as succeeding the Hermosa
and below the Dolores. The Cuder is assigned to the Permian purely on strati-
graphic grounds, and is separated from the Rico by a purely arbitrary line.
There seems to be as good ground for assigning the Lykins, above the crinkled
sandstone, or at least the lower portion of it, to the Permian, and thus correlating
it with the Cutler.

" 'Above the Cutler formation in the San Juan region is a series of sandstones,
sandy shales, and conglomerates which vary in thickness from 800 to 400 feet,
and from that down to 30 feet at the San Miguel River, disappearing entirely
north of this river. These shales and sandstones are a bright vermilion in color,
and are known as the Dolores formation. They are assigned to the Triassic
age because of the scanty, but widespread, vertebrate, invertebrate, and plant
remains. The extreme upper part of the Lykins in Larimer County may be
equivalent to the Dolores and thus be Tricissic. If so, it is impossible to draw a
line between the Permian and Triassic in eastern Colorado.' "

CHAPTER IV.
THE BASIN PROVINCE.

A. THE UPPER PENNSYLVANIAN IN THE BASIN PROVINCE.

Though the distinction between the Permo-Carboniferous red beds of
the Plains and Basin Provinces is clearly marked, there is strong evidence that
the Pennsylvanian limestone which forms their base is continuous at least
across the southern end of the barrier which divided them. (Fig. 4.) There
has been a very general consensus of opinion that the Hueco limestone of
the Guadalupe Mountains is equivalent to the Kaibab of the Grand Canyon
region and connects through that limestone with a series of deposits which
extend through the basin province as far north as Alaska and that the fauna
of the whole horizon places it as equivalent to the Russian Gschelian; but
for a recent expression of a different opinion see the summary of a paper by
Schuchert on page 152. Girty says:^

"The Hueconian fauna is widely distributed over the West, ranging indeed
into Alaska, while it is even recognizable in Asia and eastern Europe. Most of
the occurrences of Carboniferous in the West can be referred to this series,
although some of them present more or less distinctive facies. The more im-
portant of the facies provisionally referred to the Hueconian are these: that of
the Aubrey group of Arizona, rather widely distributed; that of the phosphate
beds of the Preuss formation [Park City formation], local in Utah, Idaho, and
Wyoming; the Spiriferina pulchra fauna with a considerable distribution in
Idaho, Wyoming, Utah, and Arizona; the fauna of the McCloud limestone of
California probably extending into Nevada; and that of the Nosoni formation
of California (in part of the 'McCloud shale'), apparently recognizable to the
ecistAvard and to the North and West, even into Alaska." A little further on in
the same paper Girty says that 'The Gschelian is clearly related to our Hue-

conian.'"

Further remarks by Girty in the same paper^ make it apparent that he
is far from assuming a definite position with regard to the equivalency of
the Russian and American beds:

"I am tentatively assuming, on the grounds noted above, that the Guada-
lupian is equivalent to the Permian or to the Permian and Artinskian, the one
representing a normal marine and the other an abnormal facies. It may prove,
however, that all or part of the Guadalupian is younger than the Permian. * * *
In view of the striking difference between the faunas of the Guadalupian and the

' Girty, Geo. H., Outlines of Geologic History, p. 130, 1910.
* Loc. cit., p. 133.

• 113

114

ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Hueco formation, in which the brachiopods are most in point, of a lack of a
corresponding difference between the Gschelian and Permian, of the marked
resemblance of the brachiopods of the Hueco and Gschelian, and of the lack of
agreement between the Permian and Guadalupian, there is a possibility, if not a
certain probability, that the Artinsk and Permian maybe correlated with the
Hueco formation."

S5 ^- -^ S 1

'#D

/^i\Mi/ / \< / / /A^TXr^^Sli

CuZ/h-

?f^:fy

?^^W\y^&/JMSii

<^y

~f — ?o

V""^

^ /

^>*s .

^iCas

/ ^■<^^L^N{/ r / \ P\-i / "^rh-, L

/ mrS ir^rC /(r-V / W /'WT

n/ ' /^ wV^xOS'C /y^^^^r^-^ /

J>» /

/^^^^~(' 'Y iy^,_7J]^^rrr~

c«

/ ^^^M:^/lwwy]V

'^'1

tf /

^C / ^'^ — ~-p^

4-M

/ ^^*"-''^ / r^ )/ ^ V / x!"" ^— .. J N^^~l

5f
P

^jj<Mj ^^1^-^ £s-IW

/ ( ^mPt^^Mi ^$d:^hwi::^

is /

' ^RTit-^^^JM^mK i ^

"^ .

So. /

^"^

-x

^^"^-if--^ I^ ^---^ ■•--V^N-' Cv

7^ — V-/L I r 7~ ^ ^ V. ^^^

/ '\/v~A~— -if-l^ ^ ^

/ TNVrr ) — V-O-

V .

125"

120" ri5' no' 105* 10

32.
23.
24.
2S.
26.
27.
28.
20.
30.
31.
32.
33-
34.
35.

36.

37.

38.
39-
40.

Guadalupian area.

Southern and central New Mexico.

Bisbee, Arizona.

Clifton- Morenci, Arizona.

Globe, Arizona.

Ojo Caliente. New Mexico.

Zufii Uplift, New Mexico.

Manzano area. New Mexico.

Grand Canyon, Arizona.

San Juan area, Colorado.

Ten Mile quadrangle. Colorado.

Canyon Range. Utah.

Eureka, Nevada.

Battle Mountain, Nevada.

Sierra Nevada, California.

Klamath, Redding, California.

Uintah Mountains.

Wasatch Mountains.

Park City district, Utah.

Wind River Mountains, Wyoming.

Owl Creek Mountains, Wyoming.

Big Horn Mountains, Wyoming.

Yellowstone National Park,

Black Hills, South Dakota.

Laramie Mountains, Wyoming.

Phosphate area, Idaho.

Wood River area, Idaho.

Eastern Oregon.

Three Forks, Montana.

Little Belt Mountains, Montana.

Fort Benton, Montana.

Big Snowy Mountains, Montana.

Phillipsburg, Montana. .

Republic area, Washington.

Duncan map area, Vancouver Is-
land.

Skagit Range, British Columbia.

Hozomeen Range, British Colum-
bia.

Okanagon Range, British Colum-
bia.

Midway Mountains and Anarchist
Plateau, British Columbia.

Rossland, Pend d'Oreille, Selkirks,
and Columbia system west of
Christiana Lake, British Colum-
bia.

Flathead Valley, British Columbia.

Blairmore, Alberta.

Banff, Alberta.

Highland Valley and Thompson
River, near Kamloops, British
Columbia.

Bridge River and Chilkas Lake,
British Columbia.

Stewart and Tacla Lakes, British
Columbia.

Ketchikan and Wrangell district,
Alaska.

Lake Tagish, British Columbia.

Peale and Stewart Rivers, Yukon.

Headwaters of White River, Chisana
and Nabesna Rivers, Yukon.

Headwaters of Copper and Susitna
Rivers, Gulkana River, Alaska.

Upper Yukon and Porcupine Rivers,
Alaska.

Upper Yukon and Fairbanks quad-
rangle, Alaska.

Fig. 4. — Key map giving location of areas of late Pennsylvanian
deposits in the Basin Province discussed in this work.

THE BASIN PROVINCE 115

Despite this uncertainty expressed by Girty, there seems to be a very
definite opinion in the minds of most workers in this region that the fauna
and horizon equivalent to the Hueconian west of the Rocky Mountain barrier
is Gschelian in age.

Beede has expressed the opinion that the Delaware limestone is upper
Pennsylvanian and that the Hueco is Mississippian in large part. This
opinion is frequently expressed by Girty himself, in other papers.

Though the Hueco and its equivalents occupy a position considerably
below the top of the Pennsylvanian, they afford a very convenient base
from which to reckon the changes which occur in upper Pennsylvanian and
Permo-Carboniferous time, and for this reason is traced in considerable detail
in the follo\\nng pages. Except in the Guadalupe Mountains and the
regions far north in the United States, western £md northwestern British
Columbia, and in Alaska, this limestone horizon is followed by shale and
sandstone beds which lead up to red beds or their chronological or conditional
equivalents. In the Guadalupe Mountains such a transition is lacking,
the Delaware limestone following the Hueco limestone directly; in the other
regions cited, the red beds and the transition beds, if ever present, have
been wholly or in part removed by erosion.

(a) Conditions in Texas. — Udden states of the Hueco limestone formation
that it —

"consists almost entirely of a gray, hard, thick to thin-bedded limestone, which
contains ver>' little magnesia, or none at all. At the base of this limestone
generally occur yellow to brown and purple sandstones, and conglomerates, as
well as some gray and yellowish shales. The entire formation is at least 5,000
feet thick." *

It occurs pretty widely through Trans-Pecos, Texas, outcropping in the
various uplifts; the southern limit is unknown. South and southwest of
the Guadalupe Mountains in the Shafter region, Presidio County, the
Cieneguita beds consist of —

"dark to black shales alternating with dark limestones, conglomerates, and heavy
lenticular masses of a clastic rock composed of siliceous fragments cemented by
calcareous clay (mortar rock). The shale predominates. Locally, some layers
of black chert occur. This whole series of rocks is at least i ,000 feet thick." * * *

"The Alta beds, which rest on the Cieneguita beds, show a thickness of about
3,500 feet. They consist of some dark shales below and some yellow sands above.

"The dark shales consist of sharply bedded layers of silt, clay, and some sand,
•with layers of coarser and more purely sandy material. The thickness of this
series is approximately 2,000 feet.

"The yellow sand consists of a soft, occasionally almost crumbling, bluish-
gray sandstone of fine texture. It is a coarse silt of well-assorted quartz grains.
The thickness of this series is about 1,500 feet."

* Beede, J. W., The Correlation of the Guadalupian and Kansas Sections, Amer. Jour. Sd.,

vol. XXX, p. 131, 1910. See his figure reproduced as figure 2 of this pap>er.

* Udden, J. A., Re\-iew of the Geology of Texas, Bull. 44, University of Texas, p. 48, 1916.

116 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

In northern Brewster and southern Pecos Counties is a series of upper
Pennsylvanian rocks called Gaptank by Udden/

"It consists at the base of conglomerates alternating with limestones, sand-
stones, and shales; in the middle part of shales interbedded with limestone; while
the upper portion is mainly composed of several masses of limestone separated by
shaly and sandy mateiial. The thickness of the whole formation surpasses 1,500
feet. It probably corresponds to the highest portion of the Pennsylvanian of
central Texas — the Cisco formation."

The Leonard formation is regarded by Udden as Permian, but seems
more properly to belong with the Pennsylvanian series. Udden describes
it as follows:

"The upper part is composed of thinly laminated yellowish sandstone inter-
bedded with layers of gray limestone, yellow chert, and gray shales.

"The lower part consists of heavy and thinly bedded gray limestone, in part
conglomeratic or containing pebbles of different size. At the base, shales and
soft sandstones are interbedded with a dark gray limestone. The thickness of
the entire series is nearly 1,800 feet. * * *

"Farther to the east, near Word's ranch, the series is much more calcareous
in the upper and middle part; the thickness is more than 2,300 feet. Toward
the west of the first- mentioned section, the shales begin to predominate, while
the limestone is reduced in thickness. At the base of this formation is a con-
glomerate from 20 to 200 feet thick, and this unconformably overlies the Pennsyl-
vanian." [Italics, Case. See also the description of the deposits in Trans-Pecos
Texas on a previous page.]

To the north and west the stratigraphic equivalent, if not the continua-
tion of the Hueco, is the Kaibab limestone of the Aubrey formation in the
Grand Canyon section. A typical section from this region is given by
Darton.'

Feet

Limestone, variegated crimson and lemon-yellow, with nodules of chert and iron, to

top of hills 50

Sandstone, coarse, drab, sometimes pinkish, in places containing many quartz pebbles

and imperfect vegetable impressions 2

Massive cream-colored limestones with geodes containing calcite 1 6

Chert

Cherty limestone 2

Blue limestone with Productus semireticidalus, etc., very abundant 4

Cherty limestone, light blue, containing Productus semireticulatus, P. occidentalis,

Spirigera sublilita, Orthisina umbraculum, Rhynchonella uta, etc 175

Shales, green, red, and white, and snowy gypsum 180

Hard blue limestone containing crinoidal columns, spines of Archaocidaris, Pro-
ductus, Spirigera, etc 100

Soft lemon-yellow limestone with few Productus ivesi, etc 90

Drab cross-bedded sandstones (Coconino).

The connection between the two areas is shown by the occasional appear-
ance in uplifts of rocks of similar appearance and probably equivalent age.

' Udden, J. A., Review of the Geology of Texas, Bull. 44, University of Texas, p. 47.
^ Idem, p. 51.

' Darton, N. H., A Reconnaissance of Parts of Northwestern New Mexico and Northern
Arizona, Bull. 435, U. S. Geological Survey, p. 29, 1910.

THE BASIN PROVINCE 117

According to Darton, such rocks occur in the Zuiii Uplift, at Ojo Caliente,
and in the Nacimiento Mountains, all in northwestern New Mexico.

More directly north of the Guadalupe Mountains, but still west of the
barrier between the two provinces, limestone of this horizon occurs in con-
siderable quantity. Lindgren^ says that a considerable thickness of Penn-
sylvanian limestone existed over the whole State, with a maximum thickness
between Santa Fe and Las Vegas. As far south as Socorro there is an
alternation of shale and sandstone with the limestone, but south of this
point limestone predominates, with a decrease in the total thickness. All
the conditions indicate shore conditions in the northern part of the State,
where some land existed even at this (Pennsylvanian) time. The land
mentioned by Lindgren was, in part at least, the barrier between the Plains
and Basin Provinces.

(b) Conditions in New Mexico. — ^The upper Paleozoic rocks of New Mexico
are divided into the Manzano formation and the Magdalena formation. The
latter, and lower, is entirely Pennsylvanian, the former is in part Permo-
Carboniferous.

Near Socorro, the Magdalena formation is nearly 1,500 feet thick, divided
between the upper Madera blue limestone and shale and the lower Sandia
limestone and shale, with minor quantities of sandstone. Lee' gives a
description of this limestone in its relation to the Manzano formation in
central New Mexico. He says that in late Magdalena time an uplift, or
other change, occurred in the mountain region of New Mexico which caused
a change of sediment from limestone to red beds. In the Mimbres Moun-
tains, near Kingston —

"The basal Magdalena strata consists of about 300 feet of dark-blue and gray
limestone in thick beds with thin shale partings. The upper part of the group
has about the same thickness and consists chiefly of blue and drab shales inter-
stratified with several beds of limestone 15 to 20 feet thick. Unconformably
overlying these beds are red sandstones and shales (Abo sandstone) of the Man-
zano group. * * * There is no evidence on which to separate the group into
Sandia formation and Madera limestone, as in the region farther west." *

(c) Conditions in Arizona. — In southern Arizona the Pennsylvanian (?)
limestone appears at Bisbee and Globe. Describing the first area, Ransome*
says:

"The Naco limestone * * * is made up chiefly of light-colored beds, which
consist essentially of calcium carbonate. The beds range in thickness from a

' Lindgren, W., L. C. Graton, and C. H. Gordon, Ore Deposits of New Mexico, U. S. Geo-
logical Survey, Professional Paper No. 68, pp. 31-32, 1910.

» Lee, VV. T., and G. H. Girty, The Manzano group of the Rio Grande Valley, New Mexico,
U. S. Geological Survey Bull. 389, 1909.

» Darton, \V. H., A Comparison of the Paleozoic Sections in Southern New Mexico, U. S.
Geological Survey, Professional Paper 108-C, p. 53, 1917.

* Ransome, F. L., The Geology and Ore Deposits of the Bisbee Quadrangle, Arizona, U. S.
Geological Surs'ey, Professional Paper No. 21, p. 45, 1904.

118 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

few inches to lo feet, but are usually thinner than those of the Escabrosa
[Mississippian] limestone. [The Naco limestone isj compact and nearly aphan-
itic, ringing under the hammer, and breaking with a splintery fracture, whereas
the Escabrosa limestone is usually more granular and crystalline, and crumbles
more readily when struck. There are, however, exceptions to this rule, dense
aphanitic beds occurring rarely in the Escabrosa formation and granular crinoidal
beds being not uncommon in the Naco limestone. * * *

"While the greater part of the 3,000 feet or more of the Naco formation is
made up of fairly pure gray limestone, certain thin beds of a faint-pink tint occur
at several horizons * * *. These pink beds, which weathering usually shows
to have an inherent lamellar or shaly structure, are very fine grained and compact
in texture. They effervesce freely with cold dilute acid and are evidently com-
posed chiefly of calcium carbonate. Examination of natural surfaces with a lens,
however, shows the presence of minute quartz grains and tiny flakes of mica. * * *

"Chert is not uncommon in the Naco formation; it occurs sometimes as
irregular bunches and nodules in beds of otherwise pure limestone and sometimes
as the result of silicification of thin fossiliferous beds throughout their thickness.
It is also particularly abundant along and near zones of Assuring and faulting.

"Conditions of deposition. — As littoral sediments are entirely lacking and
terrigenous materials represented only by the very minute mica scales and quartz
grains in the pink calcareous shale, it may fairly be concluded that the Naco
limestone, which is not noticeably dolomitic, was deposited in moderately deep
water at some distance from the shore, whence the tiny mica scales were derived.
* * * The region of deposition was in the main beyond the 'mud line,' which the
results of the Challenger expedition showed to lie generally at a depth of 100
fathoms. During certain stages of the accumulations of the limestones, offshore
currents carried some of the finest of the land waste into this area of tranquil
deposition and left records of these occasional incursions in the form of pink
shales."

Girty says,^ after listing the fauna of the Naco:

"Very few of these species have exact representatives in the Mississippi
Valley Pennsylvanian * * *. Prodiictus itesi, Prodiictus occidentalis, and A rcheo-
cidaris ornata are suggestive of the Aubrey limestone of the Grand Canyon
section, just as the abundance of Omphalotrochus suggests the ' Permo-Carbon-
iferous' of California. The whole fauna is closely related to that of the limestones
of the Hueco Mountains in western Texas. * * * Its age seems to be late in the
Carboniferous, perhaps about the same as the series just referred to. The fauna
has at the same time a very different facies from that of the Guadalupe fauna,
as well as from that of the so-called ' Permo-Carboniferous ' of the Mississippi
Valley."

The top of the Naco is cut off by a great unconformity.
The Globe district was described by Ransome^ in 1903. Of the late
Paleozoic deposits he says (page 40) :

"Wherever thick sections of the Globe limestone are exposed it is found that
the alternating buff and gray limestones with subordinate grits are overlain by

1 In U. S. Geological Survey, Professional Paper No. 21, page 54.

^ Ransome, F. L., Geology of the Globe Copper District, U. S. Geological Survey, Profes-
sional Paper No. 12, pp. 40, 109, 1903.

THE BASIN PROVINCE 119

gray, sometimes slightly pinkish, crinoidal limestones, usually in rather thick
beds, but also some cherty beds and an occasional bed of siliceous conglomerate."

The lower part of this limestone is regarded as Mississippian and the
upper part as Pennsylvanian. On page 109 Ransome states:

"From the Devonian to the Upper Carboniferous the region was covered by
a sea of some depth abounding in marine life and depositing abundamt limestone.
Although no characteristic Lower Carboniferous fauna Wcis found, rocks of that
period may be present, and the Globe limestone as a whole contains no \'isible
unconformities. From time to time there were slight incursions of sediment,
and in a few instances bands of siliceous conglomerate were intercalated within
the limestones. The mass of these is unimportant, but they are significant in
showing that this part of the Devonian and Carboniferous sea was probably
neither very deep nor far distant from a land-mass.

"The Upp>er Carboniferous limestone is the latest Paleozoic deposit of which
the region preserves any record. If marine conditions continued into the Permian
the deposits of that pyeriod must have been wholly removed before the strata
were broken up and invaded by diabase. Had Permian or later beds been
invohed in that structural revolution some traces of them would probably have
been preserved in the resulting intricate lithological mosaic. * * * The region
was presumably elevated above sea-level at the close of the Carboniferous and
subject to erosion."

[The Clifton-Morenci district in Arizona shows] "at least 500 feet of heavy-
bedded bluish-gray limestone which unquestionably represents both the Missis-
sippian and the Pennsylvanian. In a general way the lower 200 feet are equiva-
lent to the Modoc formation, although its several members recognized farther
south can not be identified, and the upper 300 feet represent the Pennsylvanian.
It is not possible, however, to draw a dividing line, and the whole therefore has
been included in the single formation named the Tule Spring limestone. * * *" 1

"Above the Morenci shales,' in a deepening sea, were def>osited a series of
limestones, first dolomitic, then remarkably pure and rather coarse, the Modoc
and Tule Spring limestones of the Mississippian and Pennsylvanian epochs.
Throughout the whole of the Carboniferous animal life wais abundant and rich
faunas may be collected in many places."

The Pennsylvanian beds are cut off by the erosion face of a great un-
conformity.

In the Fort Apache, Arizona, region, according to Reagan,' the upper-
most beds belong to the Aubrey group:

"The interstream spaces of all the streams in their upper course as well as
the south front of the Mogollon Range and the Cibicu Di\'ide, where not covered
with later deposits are capped with 280 feet of calcareous sandstone followed by
500 feet of soft red and gray shales, interrupted by sectile limestone. The
Aubery (Kaibab) limestone occurs in one locality — at the head waters of the

* Lindgren, W., Description of the Clifton Quadrangle, Folio 129, U. S. Geological Survey,

P- 5. 1905-

* Idem, p. II.

» Reagan, A. B., Geology of the Fort Apache Region in Arizona, Amer. Geol., vol. zxxn,
p. 280, 1903.

120 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Cibicu and Canyon creeks. The rocks of this group are usually non-fossiliferous;
but fossils enough were obtained (Athyris subtilUta, Productus punctatus, Spirifer
cameratus, Productus, and Bellerophon) to identify it as upper Carboniferous."

The upper Carboniferous of the Zuni Plateau was divided by Dutton^
into the upper Aubrey and the lower Aubrey.

"The lower Aubrey consists of bright-red sandstones throughout, deposited
usually in rather thick, and less frequently in moderately thin, layers. They
are much alike in all outward respects, color, texture, and grouping, and in the
erosional forms sculptured out of them. They are very fine grained, without
traces of conglomerate or coarse shingle or gravels; and having a calcareous
cement they weather easily and break down into very fine red sand. Fossils
are scarce, but may be found here and there in sufficient quantity and distinctness
to identify their age. These fossils, so far as I have seen, are the same as those
which abound in the beds above them.

"The Aubrey is composed largely of sandstones, but they have a very different
aspect from those below. In color they are yellowish-brown, and the cement,
instead of being calcareous, is siliceous, in fact a regular chert. * * * These sand-
stones are often conspicuously cross-bedded, and the silicification of the rock has
in no way obscured jt. * * * There are several bands of these adamantine sand-
stones, and intercalated with them are three or four thick beds of pure limestone,
containing an abundance of fossils of many and characteristic species."

{d) Conditions in Colorado. — In southwestern Colorado the uppermost
Pennsylvanian deposits are the Hermosa and Rico.

The Hermosa is described by Cross and Howe'^ as a series of alternating
limestones, sandstones, and shales having a maximum thickness of 2,000 feet.
In the Animas Valley the lowest third of the formation consists of green
sandstones and shales with some gypsiferous shales; the rest of the forma-
tion has layers of limestone distributed throughout; toward the southwest
the limestones become more important, with some beds reaching a consider-
able thickness.

In the Rico district the massive limestones are abundant only in the
middle of the formation, the upper part being mainly black and gray shales
alternating with green grits and sandstones, and a few limestones.

At Ouray the lower 300 feet are made up of thin alternating beds of
sandstone, shale, and a gnarly, fossiliferous limestones.

The upper and greater part of the Hermosa consists of pink, massive
grits and sandstones, red sandy shales, and gnarly fossiliferous limestones.
The massive sandstones, which are coarse and gritty, vary from 50 to 75
feet in thickness and are separated from one another by the red shales and
thin-bedded sandstones or calcareous layers. Heavy beds of limestone
occur in the southwestern San Juan region, but are lacking near Ouray.

' Dutton, C. E., Mount Taylor and the Zuni Plateau, U. S. Geological Survey, Sixth Annual

Report, p. 132, 1885.
' Cross, Whitman, and Ernest Howe, Ouray Folio, U. S. Geological Survey, No. 153, p. 4,

1907.

THE BASIN PROVINCE 121

The Cutler (Permo-Carboniferous) is apparently conformable on the
Hermosa in the Ouray district.

In the same folio (Ouray, No. 153) Girty gives a list of the fauna occurring
in the Hermosa and says (page 4) :

"The fauna of the Hermosa formation occurs also in the Weber limestone cuid
lower Maroon formation of the Crested Butte district, and in the Weber formation
of the Tenmile and Leadville districts. From this fact and the similarity in
stratigraphic occurrence a correlation of the formations seems to be justified.
The Hermosa fauna represents early Pennsylvanian sedimentation, and it is prob-
ably older than the ' Upper Coal Measures ' faunas of the Kansas and Nebraska
section."

At Rico, Colorado, a distinct series of deposits overlies the Hermosa and
is conformable with the overlying Cutler. This deposit seems to be some-
what local in character.

The Rico formation^ is about 300 feet thick and is made up of sandstones
and conglomerates with intercalated shales and thin fossiliferous limestones
which are usually sandy.

"The general characteristics of the Rico formation in the vicinity of Rico are,
first, its calcareous nature, in which it resembles the strata above and below;
second, the feldspathic constitution and the coarseness of its sandstones, in which
respect it difi^ers from the Hermosa and resembles the Cutler; and third, its
chocolate or dark-maroon color, which contrasts sharply with the gray or green
of the Hermosa and which is more or less distinct from the bright vermilion of
the Cutler and Dolores. * * *

"The bulk of the formation is made of sandstones and sandy shales composed
of such materials as are derived from the disintegration of granite. The s£uid-
stones are mostly coarse or conglomeratic, always showing grains of fresh feldspar
mixed with mica flakes and quartz. * * * The coarser sandstones are usually
cross-bedded and occur in massive beds from 2 or 3 to 25 feet in thickness. Some
of the coarser sandstones are of \ery much lighter color than the mass of the forma-
tion. \Mien fine-grained the sandstones are usually somewhat laminated and
pass into sandy shales. The shales, aside from the sandy varieties, are of two
kinds — the fine-grained, unlaminated, red, marly beds, similar to those of the
Cutler, and the equally fine-grained, laminated clay shales of a green color.

"Intercalated with the sandstones and shales, which are for the most part very
calcareous throughout, there are several beds of impure limestone, some as earthy,
gray, sometimes nodular bands associated with the marly shales, and others as
sandy limestone of a red color, in strata from 6 inches to 2 feet in thickness."

Fossiliferous sandy limestone layers occur as lenses shading into the
sandstone both horizontally and vertically. One or tv\^o are very persistent.
The Rico pass conformably into Cutler with no sharp distinction.

Cross and Howe* give the following statement with regard to the relation
of the Hermosa and Aubrey:

1 Cross, Whitman, and F. L. Ransome, Rico Folio, No. 130, U. S. Geol. Sur., p. 3, 1905.
* Cross, Whitman, and Ernest Howe, Red Beds of Southwestern Colorado and Their Cor-
relation, Bull. Geol. Soc. Amer., vol. 16, p. 466, 1905.

122 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

"The Hermosa and Aubrey [Kaibab] faunas are both regarded as Pennsyl-
vanian, but Mr. Girty informs us that the Hermosa has no species in common
with the typical Aubrey [Kaibab] of the Grand Canyon section, as far as known.
Mr. Girty further states that the lower Aubrey fauna from beds at the junction
of the Grand and Green Rivers, comprising a large part of the Aubrey fauna
described by White in Powell's Unita report, is markedly different from the
Aubrey of the Grand Canyon, as it also is from the Hermosa fauna of Colorado.
These faunal differences must seemingly be explained in one of three ways: (i)
by a rapid gradation of forms within a comparatively narrow zone; (2) by the
assumption of an effective barrier between the Aubrey and Hermosa seas, extend-
ing for hundreds of miles from eastern New Mexico west and northwest across
New Mexico and Utah; or (3) by assuming one of the formations to be younger
than the other, and that the Pennsylvanian section is incomplete both in Colorado
and in the Plateau country * * *."

It is probable that the Hermosa is the equivalent of the Weber interval
to the north and either equivalent to, or a continuation of, the Aubrey.
If the latter, there is evidence of a transition into the conditions which
determined the Weber grits which cover such a large area to the north.

The first appearance of the Weber formation in a locality where it has
been described is in the area of the Tenmile quadrangle in west central
Colorado. In this region, Emmons^ says of it:

" Weber formation. — This formation constitutes the most siliceous member of
the Carboniferous system, and corresponds in a general way to the Weber
quartzite of the Wasatch Mountain section and to the Lower Aubrey [Kaibab] of
the Colorado Canyon section. It includes a lower calcareo-argillaceous member,
designated in the Leadville report the Weber shales, the main siliceous formation
being there called the Weber grits.

"The Weber shales constitute a transition series between the massive lime-
stones below and the coarse sandstones of the Weber grits above. They consist
in the Mosquito Range of argillaceous and calcareous shales, alternating with
quartzitic sandstones. The former are generally very carbonaceous and often
contain seams of impure anthracite, up to several feet in thickness, but of no
commercial value. The calcareous members sometimes develop considerable
beds of impure limestones, generally fossiliferous, which are distinguished from
those of the lower formation by fossils that are exclusively of Coal Measure
aspect. The thickness of this series, which is very variable, is assumed in this
region to be about 300 feet. It occurs in the valley of Eagle River, just west of
the limits of the quadrangle.

"To the Weber grits belong the lowest beds exposed in the Tenmile district
west of the Mosquito fault. Their average thickness here, as in the Mosquito
Range, is about 2,500 feet. They consist mainly of coarse sandstones or grits,
often very micaceous, with a subordinate development of shales and a few thin
and non-persistent beds of dolomitic limestone. The sandstones are generally
light gray in color, but near the base of the series are sometimes quite dark from
the presence of finely divided carbon, probably in the form of anthracite or
graphite. Their prominent constituents are quartz and feldspar, evidently
derived from the Archean; pink orthoclase is sometimes so abundant as to

* Emmons, S. F., Tenmile District Special Folio, No. 48, U. S. Geological Survey, p. i, 1898.

THE BASIN PR0\1NCE 123

impart a reddish tinge to the rocks. The abundant mica is mostly muscovite,
biotite being present in subordinate quantity'. The muscovite is probably of
secondary origin, for it is present in much greater quantitj' than could reasonably
be expected, if it were directly derived from the Archean, and in Iju-ger leaves thcin
is common among the gneisses observed.

" In the Mosquito Range two beds of limestone, each about 50 feet in thickness,
are found about the middle of the formation. In this district the limestones are
more prominent in the upper part, but are very irregularly developed. Six
different beds were observ^ed on the south face of Sheep Mountain, but at other
points not over two could be detected. They are generally rather thin, but the
principal bed in the northwestern part of the district is 60 or 80 feet thick. At
the southern boundary of the area this bed can not be detected, and has appar-
ently thinned out. The great variability of the many thin beds of limestone in
this and the succeeding formation is so remarkable that these have been desig-
nated on the map by a special color, which shows approximately the variable
extent of calcareous sedimentation in the midst of a great thickness of prevailingly
coarse siliceous deposits. The limestone beds are also important in defining
horizons, and to them are confined a large and important class of the ore deposits
of the district.

"The limestones in the Weber grits are all typical dolomites, with a small
but persistent admixture of carbonates of iron and manganese (1.5 to 5 per cent)
and up to 10 per cent of insoluble matter.

" Maroon formativn. — This name is here applied to about 1,500 feet of beds
which in many respects resemble the Weber grits, but which in the Mosquito
Range ha\e a much larger proportion of calcareous and argillaceous beds. This
formation in the Tenmile district consists predominantly of coarse gray and red
sandstones, in some places passing into conglomerates, with many irregularly
developed beds of limestone. As contrasted with corresponding members of the
Weber grits, the following distinctions have been noted. The red color of the
sandstones, which is more common than in the lower formation, though less pro-
nounced than in the beds of the next above, results not from the presence of pink
feldspar, as in the Weber grits, but from abundant iron oxide impregnating the
cement. Hence in depth, as shown in underground workings, the red color
generally gives way to a greenish gray. The strata in which the red color is most
pronounced are fine-grained and often somewhat schistose. These sandstones
also contain a large proportion of feldspar fragments. In one case it was found
that a mixture of carbonates had almost wholly replaced the abundant feldspars,
which, nevertheless, were to all outw'ard appearances quite fresh.

"The argillaceous shales are often black, but seldom contain actucd coal;
they are much more abundant than would appear from a hasty inspection of the
hill slopes, where their outcrops are readily obscured by the debris from the harder
rocks.

"The limestones of the Maroon formation constitute, however, its most
characteristic feature, and, independently of color, seem to aiford the safest means
of distinguishing it from the Weber grits. In outward appearance they are
purer and are evidently freer from arenaceous material. They are light bluish
gray or drab in color, becoming white on weathered surfaces, in strong contrast
with the dirty brown weathered surface of the dolomites. They have also a
conchoidal fracture and lithoidal structure, instead of the rough granular fractured
surfaces which characterize the latter. Chemical analysis confirms these indica-

124 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

tions, an examination of nine specimens proving that, with one exception, they
are non-magnesian limestones with a variable but generally small percentage of
insoluble material, and less than i per cent of (Fe]VIg)C03. In the limestone
from Tucker Mountain, which forms the exception, the dirty brown weathered
surface and granular texture already noted suggest the dolomitic character and
the presence of iron and manganese carbonates. Furthermore, in the absence of
definite faunal distinctions the limestone beds have been used to define the limits
of the formation, the base being taken at the limestone belt known as the Robinson
limestone, and the top of the formation at the Jacque Mountain limestone.
Both these limestones contain an invertebrate fauna of upper Coal Measure
type. The Robinson limestone is somewhat dolomitic at the base, containing
nearly 7 per cent of magnesium carbonate. The Jacque Mountain limestone is
characterized in certain layers by an oolitic structure. The rock is light bluish
gray in color, and the oolitic grains are embedded in a finely granular matrix
of similar color. They are about the size of mustard shot, and have a normal
concentric structure and sometimes a radiate appearance; grains of sand or
crystal particles serve as nuclei. This structure disappears with recrystallization,
and is entirely wanting in certain layers.

"Wyoming formation. — The beds above the Jacque Mountain limestone,
which have a maximum thickness of about 1,500 feet, have been given this name,
not because of any fossil evidence of their age that could be found, but because
by their position and petrological character they most nearly correspond to the
beds of this formation which elsewhere in the Rocky Mountain region have, on
fossil evidence, been determined to be Triassic. If the Permian is represented
in Colorado, the evidence of which appears to the writer as yet very uncertain,
it would be included in these beds, which have evidently been deposited in direct
and unbroken succession over the upper Carboniferous.

"They consist principally of sandstones, of intensely brick-red color where not
metamorphosed, with a moderate development of thin shales between their more
massive beds. Limestones are practically absent, having been found only at a
few isolated points, generally at about the same horizon, showing that the condi-
tions favoring calcareous sedimentation, which had hitherto prevailed so irregu-
larly throughout the region, had at this time almost entirely ceased. The sand-
stones are often coarse, sometimes conglomeratic, and are composed mainly of
distinctly recognizable Archean debris. Feldspar and mica are the most abundant
constituents next to quartz. Near the top of Jacque Mountain there is a con-
glomerate bed containing Archean bowlders as large as 2 feet in diameter; finer
conglomerates are abundant, in beds usually not over i or 2 feet in thickness.
In one place, on the Tenmile slope of Mayflower Hill, a conglomerate bed was
observed where the pebbles were entirely of white quartzite in a matrix of nearly
pure quartz sand. On the slope of the hills bordering Tenmile Valley on the
west, especially of Jacque, Tucker, and Copper mountains, where metamorphic
action has been most pronounced, the red color has disappeared from the Wyoming
sandstones, and the rock has become dark and quartzitic and contains much
bright-green epidote."

Similar conditions occur in the Anthracite-Crested Butte and the Aspen
quadrangles.

It is obvious that the red sandstone here mentioned above the Weber
grits is not to be directly correlated with the Wyoming formation or its

THE BASIN PROVINCE 125

equivalent, for these belong entirely upon the eastern side of the Rocky
Mountain barrier within the Plains Province. Aside from the remote
possibility of an arm or extension of the depositional area from the Plains
Province to the west across the barrier in this locality, the probability is
that these red beds are an extension of the Rio Arriba Beds (New Mexico)
and Cutler Beds (southwestern Colorado) to the north.

Beyond the limits of southwestern and western Colorado, to the west
and northwest, the red beds cease to appear in their characteristic develop-
ment, but the deposit of the equivalent period seems rather to be changed
in character than to be absent from the geological column. To realize the
geographical continuation of the upper Pennsylvanian surface which marks
the beginning of the Permo-Carboniferous series it is most convenient to
trace the position of the Weber quartzite, already fixed in its relations to
the Hueco limestone of the Guadalupe Mountains. The Weber quartzite or
sandstone, which is one of the most widely distributed and easily distin-
guished horizons of the Pennsylvanian, is said by Hague to stretch all the
way from the Front Range in Colorado to the Eureka Mountains and can
be traced north, to southern Wyoming.

(e) Conditiojis in Nevada. — In the Eureka area, according to Hague, the
Weber is overlain by 500 feet of upper Pennsylvanian limestone, but its
thickness has been reduced by erosion. In the northern and central portions
of the State the beds are 2,000 feet thick. They are distinguished by their
light color and the prevalence of fine-grained beds. "These colors are light
bluish-gray and drab, the latter possessing a conchoidal fracture and compact
texture. * * * Throughout the horizon the limestones are interstratified with
belts of grit and siliceous pebbles, held together by a calcareous cement, in
which are intercalated thin beds of purer limestone." These limestones are
very free from MgCOg, the only Paleozoic horizon at Eureka free from dolom-
itic strata.

A similar condition seems to be traceable to the northwest, where at
Battle Mountain, Nevada, HilP reports the following conditions:

"Limestones. — The limestones are well exposed on Antler Peak, and a small
area underlain by these rocks occurs west of the rhyolite cap rock at the east
side of the mountains west-southwest of Battle Mountain. Of the limestones of
Antler Peak the Fortieth Parallel geologists say : ' They extend from the summit to
the very bottom of Willow Canyon * * *, exposing 1,200 feet of heavily bedded,
dark-gray limestones in places somewhat shaly and of light bluish-gray tints.

"Carboniferous fossils were found near the base and 100 feet below the
summit of these beds.

"The writer collected some fossils from the limestones exposed on the east-
west ridge at a point 7 miles west of Battle Mountain, under the rhyolite cap.

' Hill, James M., Some Mining Districts in Northeastern California and Northwestern
Nevada, U. S. Geological Survey Bull. 594, p. 66, 1915.

126 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

The three lots collected were examined by G. H. Girty, of the Survey, who
reports the following species: Amboccelia planiconvexa, crinoid stems, Productus?
sp., Chonetes aff. C. geinitzianus, which he says are 'probably Pennsylvanian,
though the faunas are too limited to be really diagnostic'

''Red sandstones. — The red and brown sandstones and conglomerates form a
broad northward-lying belt along the central part of the eastern slope of the
mountains, being the prevailing rocks in the vicinity of Galena and showing in
great force in Copper Canyon and the ridge east of that canyon. As a rule the
eastward lower beds of this series are arenaceous shales and sandstones of light
red and yellow color. The lower central part of the series is conglomeratic.
The well-rounded pebbles are largely white quartzite and a dense red jasperoidal
material. They range from one-eighth to one-half inch in diameter, though a
few larger cobblestones occur. The matrix is a yellowish, red-weathering sand
of rather coarse texture. Interbedded with the conglomerates there are numerous
irregular, lenslike masses of red sandstone. In the upper part of this series the
material becomes fine grained, passing into dirty brown sandstones that contain
considerable mica.

" White quartzite. — The ridge between Copper Canyon and Willow Creek,
at the southwest side of the mountains, is composed of a white vitreous fine-
grained quartzite, somewhat similar to the hill north of Cottonwood Canyon
in which the Little Giant vein occurs. At the former locality the beds dip
west at very steep angles and apparently have been faulted into their present
position. * * *

"The two formations discussed above (the red sandstone and the white
quartzite) are presumably to be correlated with the Weber quartzite of the
Fortieth Parallel Survey reports."

(/) Conditions in Utah. — From the region east and northeast, the same
horizon can be traced through the Canyon Range and Oquirrh Range
south of the Great Salt Lake as the Bingham quartzite to the Uintah and
Wasatch Mountains. Loughlin^ says:

"The quartzite as a rule is of fine, even grain and varies in color from nearly
white to light and dark brown or reddish brown. Some of its beds are greenish.
Its general appearance is very similar to that of the thick Cambrian quartzite ex-
posed in the Tintic district and at several places along the Wasatch Range. * * *

"The quartzite contains a conspicuous and persistent dark-reddish finely
banded member, 400 or 500 feet thick, which is a convenient horizon marker
and indicator of the geologic structure. * * * One lens of gray limestone * * *
was noted on the north side of Fool Creek canyon. Detailed study may prove
the presence of several such lenses. * * *

"No fossils were found in the quartzite, but its apparent conformable position
above limestone of Madison age suggests that its lower part at least is Missis-
sippian. Its upper part may be Pennsylvanian. A similar quartzite of great
thickness, containing some limestone beds, forms the greater part of the West
Tintic Mountains, the southern end of which is almost connected with the north-
west end of the Canyon Range, and the writer has found upper Mississippian

* Loughlin, G. F., A Reconnaissance in the Canyon Range, West Central Utah, U. S. Geo-
logical Survey, Professional Paper No. 90, p. 54, 1914.

THE BASIN PROVINCE 127

fossils in the limestone beds. Correlation, therefore, with this quartzite fixes
the age of the quartzite of the Canyon Range as upper Mississippian.

"The upper Mississippian studied by the writer in the Tintic Mountains
north of the Canyon Range and east of the West Tintic Range consists of a
thick series of alternating limestone, shale, and sandstone or quartzite beds.
The same series, 5,000 to 6,000 feet thick, is present in the southern part of the
Oquirrh Range, and is overlain by the thick Bingham quartzite, which has been
referred by Girty to the Pennsylvanian series. In the Wasatch Mountains the
same intercalated series of limestones, shale, and sandstone is overlain by the
Weber quartzite of Pennsylvanian age. These data indicate a transition north-
ward and eastward from quartzite into strata composed largely of limestone and
shale, and suggest that in late Mississippian and Pennsylv^anian time the deposi-
tion of siliceous sediment was extended northward and eastward, overlapping the
intercalated beds of limestone, shale, and sandstone."

Granting the equivalence of the quartzite beds in the Canyon Range,
the Bingham quartzite of the Oquirrh Range, and the Weber quartzite of
the Uintah and Wasatch Mountains brings this formation dose to the
western and northwestern edge of the Red Beds in Colorado.

In the Uintah Mountains the Weber formation includes the Weber
quartzite at the base. It is described by Weeks* as follows:

" Weber formation. — The lower part of this formation is a white and gray to
greenish quartzite in thin and thick beds, some of which weather brown. In the
upper part of the formation are alternating blue and white siliceous limestones
and quartzites. The transition to the next series is through blue and reddish
limestones and shales. The greatest thickness occurs on the south side of the
Weber River, on the north slope of the range. * * * This formation, like the
'Uinta,' is quartzitic in the western and central parts of the range and grades
into a rather soft sandstone in the eastern part. No fossils were found in the
Weber formation.

"Permian, Permo-Carboniferous — Nomenclature and correlation. — The Permo-
Carboniferous series of the Uinta Range seems to correspond in position, thick-
ness, and general lithologic characters to the Upper Coal Measure and Permo-
Carboniferous formations of the Fortieth Parallel Survey. On similar grounds
they may be correlated with the Aubrey limestone of Walcott's Grand Canyon
section. The correlation with Powell's section is less definite. The limestones
overlying the Yampa sandstone of the Upper Aubrey group and an undetermined
thickness of the shales and soft sandstones of the Shinarump group app>ear to
correspond to the beds under discussion.

"Description. — The upper beds of the Weber formation are calcareous sand-
stones and siliceous limestones which weather yellow and grade into the thin
red shales and red and blue limestones of the upper part of the Permo-Carbonifer-
ous series. * * *

"One of the best sections occurs on the east side of Duchesne River below
the mouth of West Fork. There the lower 600 feet of the Permo-Carboniferous
are formed of the red and purple shales and blue limestones. Above is 1,000
feet of light gray and white sandstones, with some interbedded limestones in the

* Weeks, F. B., Stratigraphy and Structure of the Uinta Range, Bull. Geol. Soc. Amer.,
vol. 18, p. 438, 1907.

128 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

lower part. In the upper part these sandstones occur in alternating layers of
soft and compact beds full of peculiar black points or specks. These are suc-
ceeded by 800 to 900 feet of red shales, with a prominent band of light-colored
shale at the top."

In northwestern Utah the Weber quartzite continues as a strong horizon,
but gradually plays out to the northeast of Ogden, Utah. The occurrence
of the quartzite is described by Blackwelder:^

"At the type locality in Weber Canyon the quartzite is said to be 5000 to,
6,000 feet thick, but it thins toward the north and entirely disappears within 8
miles, so that farther north the Park City formation is everywhere directly in
contact with the Mississippian strata. At the base of the Weber quartzite and
intergrading with it, there are red beds consisting of brick-red sandstone with
some sandy shale and thin beds of cherty gray limestone. The limestones have
yielded a few fossils that are closely related to those in the lower part of the
Weber quartzite, and are considered by Girty to be of Pennsylvanian age. The
red beds are separated from the underlying limestones by a distinct unconformity,
but the testimony of the fossils seems to indicate that the interruption of deposi-
tion was brief. The Weber quartzite proper consists of creamy-white quartzite
or hard sandstone interbedded, particularly in the lower part, with cherty
dolomites of dark gray to black color. A characteristic of the upper beds of the
quartzite is a coarse pitting of the surface which is probably due to the leaching
out of calcite unevenly distributed through the formation."

In the Park City district of Utah, adjacent to the area discussed by
Blackwelder, the Weber quartzite continues and was described by BoutwelP
in 1912:

"The Weber quartzite, as it outcrops in this region, consists of gray quartzite
with comparatively insignificant occurrences of cherty patches and intercalated
limestone. It is characterized by massiveness both in bedding, the beds being
rarely less than 4 and in many places 8 to 15 feet in thickness, and in the absence
of parting planes. On fresh fracture it is a light brownish gray, and it weathers
to a glistening surface of a lighter shade. The normal quartzite is fine and even
grained and dense. The exceedingly brittle nature of the rock causes it to break
into sharp, irregular fragments, and when ground fine in a fracture zone it appears
as a glistening white sugary filling inclosing larger fragments. * * * It has been
metamorphosed into quartz so completely that the granular form of the original
sand is rarely discernible. * * *

"The middle and basal portions of this formation, which are not present in
this area, outcrop in prominent cliffs just south of the district. Except for a few
thin limestone beds near its top, the middle portion is massive quartzite, but in
the lower part the intercalated limestone members increase in number and thick-
ness. In Big Cottonwood Canyon a few limestones are intercalated. A thin
crinoidal sandstone occurs about 130 feet from the top, a thin pitted, cavernous,
grayish-white quartzite 460 feet below that, and a thinly banded calcareous

• Blackwelder, Eliot, Phosphate Deposits East of Ogden, Utah, U. S. Geological Survey,
Bull. 430, p. 540, 1909.

' Boutwell, J. M., Geology and Ore Deposits of the Park City District, Utah, U. S. Geo-
logical Survey, Professional Paper No. 77, p. 45, 1912.

THE BASIN PROVINCE 129

quartzite 430 feet farther down. In Weber Canyon this great formation is most
characteristically exposed as a massive, dense, homogeneous quartzite. The
insignificant exceptions are a curiously pitted and marked stratum of quartzite
just below the top and a few thin limestones in the basal portions. * * *

"The passage from this great quartzite into the overlying formation has
been a subject of considerable study without definite results. One geologist
reported that a marked uniformity [unconformity?] existed between this quartzite
and the overlying limestone. During the present survey, however, excellent
exposures showed apparently complete conformity. The lithological character
of the sediments also indicated that a full record is here found of a normal gradual
transition. Exposures in W'^oodside Canyon show a succession of calcareous
sandstones, normal sandstones, and arenaceous quartzites immediately above
characteristic massive Weber quartzite — apparently a normal transition. In
Big Cottonwood Canyon, a few miles west of this area, the quartzite gives way
upward to a sequence of sandy beds. In Weber Canyon the precise contact
w^as not sufficiently exposed to demonstrate conformability, but the evidence
obtainable pointed to that condition."

Schultz,' in 191 8, gave the following account of the Weber quartzite
in the Uinta Mountains:

"Weber QuARTzrrE (Pevnsylvanian).

"Overlying the massive Pennsylvanian limestones, which may be considered
as at least in part equivalent to the Morgan formation of the Wasatch Range,
is a thick massive gray to white quartzite or sandstone that has been correlated
with the \\'eber quartzite of the Wasatch Range. With this quartzite are asso-
ciated small quantities of chert and limestone. The absence of impurities and
cementing material is conspicuous. On fresh fracture the quartzite is pure
white to light brownish gray and it usually weathers to a light shade, although in
some localities it has a decided brownish color. The quartzite or sandstone is
fine, even grained, and dense. The exceedingly brittle nature of the rock causes
it to break into sharp, irregular fragments that form conspicuous talus slopes.
In some localities the sandstone has been so completely metamorphosed into
quartzite that the form of the original constituent grains is not readily discernible
by the naked eye. In many other places along both sides of the range, however,
the Weber quartzite is nothing more than a rather soft sandstone in which the
original grains are poorly cemented and readily detected and which on weathering
produces a fine-grained sand remarkably free from impurities. Good exposures
of this formation may be seen along both sides of the Uinta Range west of Green
River wherever the beds haA-e not been obscured by the overlying Tertiary
deposits or by the Bishop conglomerate, of late Tertiary or early Quaternary age.
The distribution of these beds is shown on the accompanying map [PI. V, in the
original publication]. East of the area examined and eeist of Green River the
Weber quartzite beds are exposed by erosion at many places along the south
flank of the mountains and along the crest of the Midland anticline at the south
side of Blue Mountain, in Colorado, where it forms the center of the oval basin
south of Midland Ridge.

* Schultz, A. R., A Geologic Reconnaissance of the Uintah Mountains, Northern Utah, with
Especial Reference to Phosphate, U. S. Geological Survev, Bulletin 690-C, p. 45, 1918.
10

130 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

"The total thickness of this formation is approximately 2,200 feet in the western
part of the Uinta Range and about 1,600 feet in the vicinity of Green River.
Powell gives the thickness in the east end of the range as 1,000 feet or more."

Schultz here discusses the possibility of an unconformity at the base
of the Weber and decides that no such unconformity exists.

(g) Conditions in Idaho. — The Weber quartzite extends into southwestern
Idaho, where it undergoes a considerable change, indicating an approach to
its northwestern limit. Richards and Mansfield^ describe it as follows:

"The Weber quartzite in the area examined in 1909 consists chiefly of massive
white quartzite with subordinate amounts of shale and calcareous sandstone or
limestone. In the region studied during the present year the conditions are
practically reversed, and the true quartzite is subordinate to calcareous sandstone
and limestones. The character and position of the small amount of quartzite
present are extremely variable. * * *"

{h) Conditions in Wyoming. — In southwestern Wyoming the Weber con-
sists of gray and white quartzitic sandstone, often brecciated, reaching a
thickness of 500± feet. Veatch says of this area:^

"From this time [Pennsylvanian] until late Cretaceous there was no profound
disturbance. The strata, so far as can be seen, are entirely conformable and the
series is complete, but the absence of beds found in other portions of the Rocky
Mountains suggests that there were land periods during this interval, produced
by broad orographic movements without pronounced deformation in this area.

:4: ;^ H: "

Blackwelder has given a more detailed discussion of the Weber and
adjacent formations in Wyoming.^

"Above this limestone [the Madison, Mississippian] is a sandy series, which
is somewhat variable in different parts of the region. The larger part of it com-
prises the Weber quartzite in southeastern Idaho and Utah and the Tensleep
sandstone in central Wyoming. Its basal portion generally contains beds of
red shale and purple or gray limestone, and where thus developed in Utah has
been called the Morgan formation, while in eastern and north-central Wyoming
Darton has named the corresponding beds the Amsden formation. The upper
and larger part of the sandy series is generally a massive yellowish or cream-
colored sandstone (Tensleep) in the east, or a quartzite (Weber) in the southwest.
On account of resistance to erosion the outcrops of the Weber, Tensleep, and
equivalent rocks are almost invariably marked by ridges or peaks, and in canyons
by cliffs. * * *

"Between these two conspicuous horizon markers, the yellow sandstone below
and the red shales above, lie the less-resistant strata which include the phosphate
beds. In north-central Wyoming, where these strata are relatively thin, Darton

* Richards, R. W., and G. R. Mansfield, Preliminary Report on a Portion of the Idaho
Phosphate Reserve, U. S. Geological Survey, Bull. 470, p. 385, 1910.

' Veatch, A. C, Geography and Geology of a Portion of Southwestern Wyoming, U. S.
Geological Survey, Professional Paper No. 56, p. 49, 1907.

' Blackwelder, Eliot, A Reconnaissance of the Phosphate Deposits of Western Wyoming,
U. S. Geological Survey, Bull. 470, p. 458, 1910.

THE BASIN PROVINCE 131

has named them the Embar formation. These beds increase in thickness and
change considerably in character in passing westward to the Hoback and Salt
River ranges, and there they may be diN-ided into several formations, corre-
sponding probably to those recognized by Gale in southeastern Idaho and
Utah, namely, the phosphatic Park City formation below, the Woodside shale
in the middle, and the Thaynes limestone at the top. As not all parts of these
strata are fossiliferous, the exact equivalence of the divisions in the eastern and
western sections has not been established, but it seems to be approximately as
stated. The phosphate beds lie near the base of the Embar formation to the
east and the Park City formation to the west and are associated with dark shale
and fossiliferous limestone. In the Gros \^entre Range the limestone above the
phosphate beds is largely replaced by chert and fossils are very scarce.

"The individual beds of phosphate rock are subject to much variation in
character and richness. In the western sections they are generally considerably
thicker than in the east and northeast. The richest variety of phosphate rock
is commonly a black oolithic material — firm but not particularly hard. When
broken it emits a disagreeable odor of petroleum. From this richer variety there
are all gradations, through hard phosphatic limestone and soft phosphatic shale
and sandstone down to beds that contain but little phosphoric acid.

"In the canyon of Snake River in western Wyoming the total thickness of
phosphatic beds, both rich and lean, exceeds 40 feet. * * *

"On the north side of the Wind River Range the phosphate beds have dwindled
to 3 or 4 feet in thickness and consist largely of gray phosphatic sandstone, which
contains only 35 to 45 per cent of tricalcium phosphate. Across the Wind River
\'alley, in the Shoshone and Owl Creek mountains, the deterioration of the phos-
phate deposits is still more marked, for there the beds are but 2 to 4 feet thick
and generally contain less than 20 per cent of tricalcium phosphate. * * *

"From the work of other geologists in Wyoming it is believed that phosphate
deposits exceeding a few inches in thickness do not occur much north of the Owl
Creek Mountains, northeast of the southern part of the Bighorn Range, nor east
of the low ranges between Casper and Lander. There b no information as to
the southward extension of the material, but it has never been recognized south
of the Wind River Range in Wyoming. It is highly probable, however, that lean
phosphate beds of some importance stretch north by west across Yellowstone
Park into southern Montana. * * *"

From the Gros Ventre Mountains northward and eastft-ard it is scarcely
practicable to discriminate the formations known as Park City, Thaynes,
and Woodside in southeastern Idaho and northern Utah. The corresponding
interval is occupied by a gray or buff alternation of shale, limestone, and
chert, with black layers of shale and phosphate rock, which is believed to
represent the Embar formation.

To the west of the areas described by Dale, Richards, and Mansfield
there is evidence of the continuation of the W^eber quartzite horizon.
Lindgren^ in 1899 described the Wood River series. The rocks are all
badly disturbed and in places metamorphosed by contact with the great

• Lindgren, \V., The Gold and Silver Veins of Silver City, De Lamar, and Other Mining
Districts in Idaho, 20th Annual Report, U. S. Geological Sur\-ey, part ni, pp. 89-90
and 194, 1899.

132 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

central granite mass of Idaho. On page 194 Lindgren speaks of these rocks
as follows:

"The rocks of the series consist of a comparatively small amount of heavy-
bedded gray limestones and a large mass of quartzitic sandstones of red, gray, or
brown color. Very frequently these are more or less calcareous. Black shales,
usually calcareous and frequently banded, gray and black, are also abundant,
and contain most of the veins in the district. Occasionally slaty rocks are met
with, showing the effects of great compression in slaty cleavage, frequently in
two directions."

A number of casts of fossils were found in a hard, grayish brown, partly
calcareous quartzite which have been identified by Schuchert as Myalina
(two species), Schizodus, AUorisma, Scaphiocrinus or Graphiocrinus, and
Fusulina (?). Dr. Schuchert^ says concerning this fauna:

"The identification of Fusulina is doubtful, since only two large cross-sections
are shown embedded in the rock. They are pseudomorphs in calcite, and preserve
only the spiral layers of growth. If Fusulina is present the horizon is upper
Carboniferous. The presence of Scaphiocrinus or Graphiocrinus often indicates
a rather lower horizon — lower Carboniferous — though Graphiocrinus is also found
in the upper Carboniferous. The pelecypods indicate no special horizon in the
Carboniferous. For the present I am inclined to view these fossils as probably
upper Carboniferous."

In northwestern Wyoming and southwestern and western Montana,
the horizon of the Weber quartzite becomes somewhat mixed, limestone,
sandstone, and shale layers occurring at frequent levels and sometimes
dominating the quartzite. The deposits in this region are called Quadrant
quartzite, and there has been considerable difference of opinion as to its
exact age and correlation with other beds, but there can be little doubt that
the Quadrant quartzite, or a part of it, is the equivalent of the Weber
quartzite and that its deposition was due to similar conditions.

In the Yellowstone National Park,^ the Quadrant quartzite —

"consists of white, yellowish and occasionally pink beds of quartzite, with
intercalated beds of drab saccharoidal limestones. The quartzite is generally
compact, occurs in beds from 4 to 25 feet in thickness, and weathers in massive
blocks. More rarely it breaks into small fragments that form debris slopes, as
seen in the Teton Range. The total thickness averages 400 feet in the Gallatin
Range. In the southwest corner of the park it is far less prominent than in the
Gallatin, but its resistance to weathering makes it easily recognizable, out-
cropping beneath the soft red clays of the Juratrias. * * *"

{i) Conditions in Montana. — At Three Forks, Montana, the quartzite of
this horizon is less dominant, except at the top. In the description of this
area by Peale^ the following account is given :

* Schuchert, in Lindgren, 20th Ann. Rept. U. S. Geol. Sur., p. 90.

' Yellowstone National Park Folio, No. 30, U. S. Geological Survey, p. 5, 1896.

' Peale, A. C, Three Forks, Montana, Folio No. 24, U. S. Geological Survey, p. 2, 1896.

THE BASIN PROVINCE 133

"After the deposition of the Madison limestones there seems to have been a
marked change in the character of the sediments, due possibly to the prevalence
of much shallower seas or to more active erosion of land areas. The lower beds,
to which the name of ' red ' limestone has been given, ju^ everywhere arenaceous
and argillaceous, and in many localities a conglomerate of limestone pebbles lies
at the very^ base of the formation. Although no true dolomites are found, these
lower limestones are all more or less magnesian. The section varies considerably
at dififerent points, but its thickness is between 300 and 400 feet. The red lime-
stone is from 170 to 200 feet thick, and the brilliant red color of its debris makes
it very conspicuous, while its soft character leads to the development of a ravine
back of the outcrops of the Madison limestone. The fossils found in the red
limestone are of upp>er Carboniferous age. Following the red limestone are from
150 to 180 feet of thin-bedded, cherty limestone, alternating with quartzitic
layers, the latter predominating at the top and being capped by a prominent
bed of quartzite or quartizitic sandstone, which has been taken as the base of
the overlying Mesozoic."

At Old Baldy, near Virginia City, the limestone is capped by the Quad-
rant quartzite. Immediately below the quartzite there is a considerable
thickness of limestone which contains Pennsylvanian fossils. Schuchert
considered these as Pottsvillian (Bull. Geol. Sec. Amen, vol. 20, p. 559,
1910). The Quadrant sandstone is certainly calcareous at its base, and
also fossiliferous.

In the description by Clark^ of the rocks of southwestern Montana the
position of the Quadrant is given as well up in the Pennsylvanian in opposi-
tion to the opinion expressed by Girt>',^ who regards it as Mississippian,
in some places at least.

Near Melrose, Montana, which lies about 27 miles directly south of
Butte, the Quadrant lies in almost the same position and relations as the
Weber occupies in southeastern Idaho and the adjacent portions of Utah
and Wyoming. Gale says:'

"The qucu-tzite of the Quadrant is overlain by light, sandy-weathering blue
limestone, about 130 feet thick in the Melrose section. This limestone contains
much black chert in nodular form and in layers. The phosphate bed immediately
overlies this sandy blue limestone and is itself overlain by ledges containing much
massive chert, so that the stratigraphic section here corresponds remarkably in
lithologic character with the sequence of strata typically associated with the
phosphate beds in southeastern Idaho."

Near Dell and Dillon, Montana, Bowen* reports the following sequence:

"Overl>Tng the limestone (No. 2) (of Mississippian age) is a great thickness of
sandstone containing some highly calcareous beds and possibly some true lime-

* Clark, T. H., Bull. Mus. Comp. Zool. Har\-ard University, vol. LXi, No. 9, p. 361, 1917.

* Comment on manuscript of Professional Paper No. 71, p. 387.

»Gale, Hoyt S., Rock Phosphate Near Melrose, Montana, U. S. Geological Survey, Bull.
470, p. 444, 1910.

* Bowen, C. F., Phosphatic Oil Shales near Dell and Dillon, Beaverhead County, Montana,

U. S. Geological Survey, Bull. 661, p. 316, 1918.

134 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Stone and chert. This formation was not studied in detail, and no fossils were
obtained from it. It seems to correspond in position to the Quadrant quartzitei

"The thick sandstone (No. 3) is overlain by a few hundred feet of gray-
limestone, which is covered by sandy beds. Fossils obtained both from the
limestone and from the sandy beds are assigned by Mr. Girty to 'the Phosphoria
formation, now regarded as of Permian age.'

"The gray limestone (No. 4) is overlain by 1,500 feet or more of thin-bedded
pinkish limestone, in which there may be some beds of shale. Fossils obtained
from the lower part of this limestone are provisionally referred by Mr. Girty to
the Lower Triassic."

In the Garnet Range, a few miles east of Missoula, Montana, the
Quadrant formation is described by Pardee^ as —

"300 feet or more of grayish-brown and yellow quartzite and red shale overlying
the Madison limestone in a narrow area extending from Tenmile Creek south-
eastward, also west of Little Bear Creek. * * * A little phosphatic sandstone
overlies the quartzite in the extreme southeastern part of the area surveyed, but
it dies out to the northwest. This phosphatic sandstone probably represents the
Phosphoria formation."

Still farther to the northwest at Philipsburg, Montana, the Quadrant
preserves the character shown near Melrose. Of this locality Calkins and
Emmons^ say the Quadrant formation is divided into two members:

Upper mainly quartzite.

Impure quartzite and quartzitic sandstone.

Calcareous shale and impure chert limestone.

Light-colored quartzite, fine-grained, thick-bedded.
Lower red magnesian limestone and shale.

Near Boulder Creek the beds are vertical and the quartzite forms promi-
nent reefs:

The upper quartzite of the upper member is less fine-grained and is somewhat
calcareous.

The middle bed is more calcareous and near Philipsburg is phosphatic.

The lower quartzite is very fine-grained and thick-bedded.

The fossils found in the calcareous beds between the quartzites are Cyathophylum ?
sp., Camarotmchia (or Rhynchopora) sp., resembling C. sappho, Camaro-
icechia (or Rhynchopora) sp., resembling C. congregata.

Dr. Girty' says of the fossils:

"These fossils must be Pennsylvanian or Permian. The presence of phosphate
strata at this horizon suggests the Permian (?) phosphate beds of southeastern
Idaho (Phosphoria formation), but the fauna is different. Mr. Gale's suggested
correlation of the red lower members of the Quadrant with the Morgan formation
of northeastern Utah, as given below, is rather confirmed than otherwise."

1 Pardee, J. T., Ore Deposits of the Northwestern Part of the Garnet Range, Montana,

U. S. Geological Survey, Bull. 660, p. 167, 1918.
^ Calkins, F. C., and W. H. Emmons, U. S. Geological Survey, Philipsburg Folio, No. 196,

p. 8, 1915.
' In Philipsburg Folio, p. 8.

THE BASIN PROVINCE 135

Calkins and Emmons continue:

"The kinds of rocks are the same in the Philipsburg section as in the typical
Quadrant of the Threeforks and Yellowstone National Park region, although
the sharp division into a quartzite and a shaly member does not there seem
possible. The assignment of a Carboniferous and probably Pennsylvanian age
to the upper quartzitic stratum is based on lithologic rather than on paleontologic
grounds, for no fossils have been found in it. The upper quartzite was included
in the formation primarily because of its resemblance to the lower. Support is
lent to this part of the correlation, however, by the opinion of Hoyt S. Gale,
who in 1910 examined a section near Melrose, Montana, that is essentially similar
to that of the Philipsburg quadrangle. Mr. Gale considers the lower and piu^r
quartzite equivalent to the Weber quartzite of Utah, and the higher beds here
included in the Quadrant as equivalent to the Park City formation of Utah.
One reason for this correlation is lithologic resemblance, but a stronger one is the
occurrence of a phosphate bed in the Melrose section corresponding to one in
the Utah and southern Idaho sections. This phosphate lies between the two
quartzitic strata and has been found at this horizon on Flagstaff Hill since the
geologic surx'ey of the quadrangle was made. The lower shaly member may have
its equivalent in the Morgan formation on Utah or in similar rocks found locally
in the base of the Weber quartzite."

Under the description of the geologic history' of the region, Calkins and
Emmons remark upon the Upper Pennsylvanian:

"The quartzites of the upper part of the Quadrant are probably beach def>osits,
suj>erp>osed on the fine-grained rocks of the lower member after an interv^al of
erosion, for continuous deposition would have been recorded by a gradual instead
of an absolutely abrupt lithological transition. It therefore may be supposed
that the inland sea of early Quadrant time was filled or upheaved and its bed,
after a brief period of erosion, again invaded by the sea, whose advancing margin
gradually covered the surface with a layer of beach sands. An interlude in these
conditions is represented by the calcareous and phosphatic beds bet«-een the
quartzite strata. The carbonate and the oolitic phosphate of lime are presumably
chemical precipitates, and are most likely to have been formed in a shallow
inclosed sea. The wide expanse and the unbroken continuity of the phosphate
beds in this region indicates that the sea extended continuously ov^er a large
part of Montana, Idaho, Utah, and Wyoming."

In going directly north from the Yellowstone Park region, the type
locality- of the Quadrant formation, into Montana, the change in the deposits
of the horizon are more abrupt than in going toward the northwest. This is
probably due to the northwestward trend of Rock>- Mountains at this pxjint.

In describing the horizon north of the Little Belt Mountains, Weed
writes as follows:^

" Quadrant formation. — This formation, named from its prominence in Quad-
rant Mountain in the Yellowstone Park, varies in character and increases in
thickness from the southern exposure in the canyon of Sixteenmile Creek north-
ward. The southern areas of this quadrangle show a series of beds 230 feet thick.

* Weed, W. H., Little Belt Mountains Folio, No. 56, U. S. Geological Survey, p. 2, 1899.

136 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

The upper layers are compact, hard, pink, and cream-colored quartzites, with
occasional intercalated beds of limestone. The base consists of 80 feet of impure
limestones with interbedded red magnesian shales that are soft and weather
readily, their red muds staining the harder rocks.

"In going northward the formation changes greatly in character and thick-
ness. In the Little Belt Range the quartzites disappear and the most character-
istic feature is the presence of a shale horizon — the Otter shale — whose vivid green
color makes it conspicuous wherever exposed. At the same time the formation
becomes of a variable nature. Limestones, sandstones, and shale beds appear, but
are not persistent. On the Judith River the base of the formation consists of the
red Kibbey sandstone, which frequently contains beds of gypsum. The thickness
of the formation is nearly 1,400 feet at this locality, while it is but 400 feet north
of Castle Mountain. The limestones carry abundant fossil remains, fixing the age
as lower Carboniferous. * * * The region was probably elevated above the sea
at the close of the Quadrant stage."

In the Fort Benton quadrangle, the quartzite has almost entirely given
place to shales and limestones. Weed says of the horizon:^

"Within the limits of this quadrangle the Quadrant formation is a variable
sequence of sandstones, shales, and limestones. The lowest beds are reddish
and yellow clayey sandstones, often holding interbedded layers of gypsum and
constituting the Kibbey sandstone. These are overlain by the Otter shales
holding interbedded limestones. The shales are dark gray or purplish near the
base, becoming a bright coppery-green higher in the sequence. The interbedded
limestones are very seldom more than a foot or two thick, are frequently oolitic,
and carry fossils of lower Carboniferous types.

"[The Quadrant formation] shows a very decided change of conditions of
deposition, indicating a rising of the region, with shore and estuarine deposits
which preceded the emergence of this tract above the sea. The change from pure
limestone to red sandstones with gypsum beds and limey shales is abrupt, but
the Quadrant contains also several beds of very pure limestone.

"* * * though there is a marked change in the character of the forms of life,
there is little change in the character of the beds between this and that of the
overlying Ellis formation, whose fossils are of the Jurassic age."

Beyond Philipsburg and Fort Benton, Montana, there are no reported
exposures of any formations that can be at all closely correlated with the
Weber horizon, but as it has been shown that in this region the Weber is
becoming more of an offshore formation with increasing limestone and is
the uppermost Paleozoic exposed, it is possible to at least trace the upper-
most Paleozoic farther to the north and west. The condition along the
boundary between the United States and Canada is condensed from Daly
on pages 1 71-178, and this area of limestone and quartzite of upper Penn-
sylvanian age is with little doubt to be correlated in a broad way with the
Cache Creek series of British Columbia and as far north as the Yukon
Territory and with the similar deposits of Alaska.

(j) Conditions on the Pacific Coast. — On the west side of the western cor-

* Weed, W. H., Fort Benton Folio, No. 55, U. S. Geological Survey, p. 2, 1899.

THE BASIN PROVINCE

137

dillera of the United States a few areas show the condition during the upper-
most Pennsylvanian time.

In California two formations have been recognized as Pennsylvanian —
Robinson and Calaveras.

The Robinson formation is described by Turner* as "sediments and
trachyte tuffs," with fossils of upper Carboniferous age, and by Diller*
as containing shales, conglomerate, tuff, and sandstone, of which the last
two are the most important. The sandstone is a purplish rock of great
variability'. One-fourth of a mile south 50° west of Robinson's, in Genesee
Valley, it becomes for a short distance an arenaceous limestone.

J. P. Smith^ considers this as equivalent to the Carboniferous portion
of the Pitt formation of Shasta County.

In northern California, north of the fortieth parallel, the upper Penn-
sylvanian is represented by the McCloud limestone and the Pitt shales.
Smith* gives the following table of the formations of Shasta County:

Carboniferous

Middle Triassic.

Pitt formation.

Pitt shales.

Upper Carboniferous.

McCloud shales. Siliceous and calcareous
shales and conglomerates, with upper
carboniferous fauna at base, 1 ,000 feet.

McCioud for-
mation.

McCloud limestones. Massive limestone
and marbles of the McCloud Ri\'er,
rich in corals and brachiopods, 2,000 feet.

Lower Carboniferous.

Baird shales.

The McCloud shales are correlated with the Robinson formation, and
both have been separated off as the Nosoni formation.

The McCloud limestone is described by Diller^ as a dark gray sand,
massive below and lighter colored and somewhat thinner above, with many
nodules of chert and sheets of gray chert, often containing silicified fossils.

The McCloud limestone, according to J. P. Smith,^ is about —

"2,000 feet in thickness, uniform in bedding, and very siliceous in places. Some
few beds are altered to a crj^stalline marble, but in the main the series is made
up of a fine-gTciined hard gray limestone. * * * The McCloud limestone is
probably equivalent to the Caribou formation of Plumas county. But J. S.
Diller thinks they belong to a lower horizon than that assigned them by the
wTiter. The Robinson beds of the Taylorville section are probably higher up
in the section, but nevertheless the McCloud limestone is, in part at least,
equivalent to the Coal Measures."

' Turner, H. \V., Bidwell Bar Folio, No. 43, U. S. Geological Survey, 1898.

' Diller, J. S., Geology of the Taylcrsville Region, California, Bull. Geol. Soc. Amer.,

vol. 3, p. 374. 1892.
' Smith, J. P., The Metamorphic Series of Shasta County, California, Jour. Geol., vol. 2,

p. 602, 1894.
*Id4:m.

* Diller, J. S., Redding Folio, No. 138, U. S. Geological Survey, 1906.

• Smith, J. P., The Metamorphic Series of Shasta County, California, Jour. Geol., vol. 2,

PP- 599 ^^^ ^ii 1^94-

138 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

The McCloud shales (Nosoni formation) are provisionally correlated
with the upper Hueco by Girty, and Schuchert^ remarks: "It also seems to
correlate with the Schwagerina zone of the Russian geologists. This zone
is just below the Perfno-Carboniferous or Artinskian."

They are described as the lower part of the Pitt formation by J. P. Smith,
as follows:^

"The Pitt formation overlies conformably the McCloud limestones, and
consists roughly estimated of about 3,000 feet of siliceous and calcareous shales,
conglomerates and tuffs. * * *

"The oldest fossiliferous strata of the Pitt formation are of Upper Carbonifer-
ous age. * * * The rock is a dark calcareous argilHte. * * * These beds are the
probably equivalents of the Robinson beds of the Taylorsville region, and of the
Little Grizzly Creek beds, Plumas County, which seem to form the top of the
Carboniferous formation. The boundary of these Carboniferous argillites could
not be found, but they probably make up the lower thousand feet of the Pitt
formation."

Diller, in the Redding Folio, says that the McCloud limestone was
formed by quiet sedimentation during Mid-Carboniferous time, that oscilla-
tions began and the McCloud was succeeded in Nosoni (late Carboniferous)
by shales and sandstones with increasing quantities of tuffs and volcanic
flows. The period of deposition was terminated in this locality by an
uplift and extensive volcanic activity.

In the Klamaths proper the McCloud is a fine-grained limestone, very
siliceous in places; it is upper Carboniferous and equivalent to the Caribou
of the Plumas County region.

The Nosoni shale is continued above the McCloud and is overlain by
the Hall City limestone of Permian age.

Girty^ correlated the McCloud limestone with the Hueco. Both Girty
and Schuchert have suggested that the same sea which deposited the
McCloud extended into southwestern British Columbia (Cache Creek), and
into Alaska as far as the Chicagoff Islands.

The deposition of the Nosoni was terminated all along the Pacific coast
by an elevation which was the culmination of the volcanic activity which
furnished the tuffs and flows of the Nosoni. This elevation probably repre-
sents a considerable interval of time before the deposition of the Triassic.
It was in all probability a part of the greater movement which is traceable
from Alaska southward, in elevation and extensive volcanic activity both
subaerial and submarine (southern Alaska) as far as the Redding, Cali-
fornia, quadrangle.

* Schuchert, Chas., Paleogeography of North America, Bull. Geol. Soc. Amer., vol. 20, p.

573. 1910.
' Smith, J. P., loc. cit., p. 601.

• Girty, Geo. H., The Relations of Some Carboniferous Faunas, Wash. Acad. Sci., vol. 7,

p. 16, 1905.

THE BASIN PROVINCE 139

North of the Klamath Mountains no record of the upper Paleozoic is left
in Oregon, and it is not again visible until the Snoqualmie quadrangle is
reached going northward.

Pardee^ reports a "series of sediments and lavas that have been more or
less metamorphosed and include shale, slate, argillite, schist, quartzite,
conglomerate, and greenstone." He states:

" These rocks are most extensively developed in the northeastern part of the
reservation (Colville Indian Reser\^ation), where, underlying most of the Covada
mining district, they occupy a belt about 8 miles wide that extends from the head
of Ninemile Creek east to the Columbia River and thence north to the reserv^a-
tion boundary. Because no well-marked stratigraphic break was seen in these
rocks, because sufficient fossils upon which to base time divisions were not found
in them, and because they are most conveniently mapped and described as a unit,
the name Covada, which has been applied to that portion exposed in the vicinity
of Covada settlement, may be extended to the metamorphic series as a whole,
which is here designated the Covada group."

Pardee suggests the probable equivalence of the Covada group with
Cache Creek and its equivalents of the Boundary Survey and northern
British Columbia and with the McCloud and Nosoni formations. He says:

" Thus a chain of Carboniferous roeks of generally similar lithology, extending
from Alaska to California, seems fairly well established, and the position of the
Colville Reservation strongly sugests that the Covada group is one of the links."

A suggestion of Paleozoic in Oregon near Grant Pass, in the south-central
part, is given by Diller and Kay.^ Here a few poor fossils (crinoid stems)
were found in limestone lentils interbedded with clay slates, siliceous
slates, and tuffs. The bulk of the limestone is Devonian, as shown by the
fossils, but the third belt of the description carrying the crinoid stems is
possibly Carboniferous, possibly Triassic. It is unconformable with the
overlying Jurassic. (See also Oregon Bureau of Mines and Geology and a
Bibliography of Oregon Geology, etc., Oregon University Bulletin, new
series, vol. lo. No. 4, 1912.)

In the Snoqualmie quadrangle. Smith and Calkins' recognized three
layers among the metamorphic rocks, which are composed of sediments and
volcanics : they are the

Peshastin.

Hawkins.

Eciston schist.

These are reported to be strikingly similar to Paleozoic rocks in California,
of the Blue Mountains in Oregon, and in the Okanagon Valley of Washington.

* Pardee, J. T., Geology and mineral resources of the Colville Indian Reservation, Wash-
ington, Bull. U. S. Geological Survey, No. 677, 1918.

' Diller, J. S., and G. F. Kay, Mineral Resources of the Grant Pass Quadrangle and Border-
ing Districts, Oregon, U. S. Geological Survey, Bull. 380, p. 51, 1909.

' Smith, Geo. O., and F. C. Calkins, Snoqualmie Folio, No. 139, U. S. Geological Survey,
1906.

140 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

The inference drawn from this is that during a portion of Paleozoic time the
Pacific Coast region from British Columbia to California constituted a single
geological province. This became land in the Mesozoic and was already
uplifted and folded before the great intrusions which metamorphosed the
rocks.

Farther north in Washington, in the Republic mining district, Paleozoic
rocks are again shown in a considerable exposure. Umpleby,^ discussing
this area, says:

"The oldest rocks exposed in the district are the metamorphic equivalents
of a great series of shales, sandstones, limestones, and lava flows which are of
Paleozoic age, and are provisionally assigned to the Carboniferous. After the
deposition of this series, the area passed through a long period of crustal disturb-
ance which, although not developing sharp folds, metamorphosed the beds and
raised the area far above sea-level. Either during this period of crustal disturb-
ance or shortly thereafter great batholithic masses of grandiorite were intruded
into the Paleozoic series. * * *

"The Paleozoic rocks are very uniformly but not intensely metamorphosed.
True schists are not common, and in many instances the limestone has not been
changed to marble. Nevertheless, the series has been so disturbed that a given
set of characteristics seldom persists for more than a short distance in any direc-
tion. Neither bottom nor top of the series was found.

"Black carbonaceous argillite is the predominant rock type, although bluish-
gray nonfossiliferous limestones have a wide development. Massive gray quartz-
ites were noted in one exposure southwest of Republic. Porphyries of inter-
mediate and basic composition are found both as dikes and sills, apparently
intruded into the series before its metamorphism. The age relations of the
various phases of the series are not obvious from studies in the Republic area, but
to the north, at Phenix, British Columbia, LeRoy reports a section including all
the above types of rocks, which he divides into three parts with an unconformity
between the upper two. His section places the argillites in the upper part,
separated from the limestones and tuffs (no tuffs of this age were noted at
Republic) by a pronounced unconformity, while the lower member is quartzite
with intruded dikes and sills of basic porphyrites. The Paleozoic beds are
folded and metamorphosed, and are in marked contrast with the overlying
Tertiary series, in which folding is less marked and the beds are not metamor-
phosed.

"It is not possible, on the strength of facts now known, to assign this forma-
tion to a definite place in the Paleozoic series. Near Republic the formation
carries certain fossils, not well preserved, but which seem to be crinoid stems.
In an exposure of limestone near the top of Buckhorn Mountain, in the northwest
part of Republic quadrangle, several fossil crinoid stems were found which are
not out of harmony with a provisional assignment to the Carboniferous. These
remains, together with the lithologic characteristics of the series, suggests a
correlation with the Cache Creek series of Dawson, which is of Carboniferous age.

* Umpleby, J. B., Geology and Ore Deposits of Republic Mining District, Washington
Geological Survey, Bull, i, p. 15, 1910.

THE BASIN PRO\aNCE 141

On lithologic grounds, however, it is thought that rocks of more than one age
are present."

KnopP suggests the name Onwenyo limestone for a small mass of lime-
stone appearing on the slopes of the Inyo Range facing Owens Valley, which
he considers as perhaps equivalent to the "Upper Coal Measure limestone"
of Hague's Eureka report. This formation " consists in the main of massive,
grayish, crystalline to compact limestones. The 2-foot basal bed is a blue-
gray compact limestone fossiliferous from the contact and carrying irregular
lenses and stringers of sandstone whose grains are apparently derived from
the Reward below. Here and there through the Onwenyo, particularly in
its upper third, are layers carry^ing rounded chert j)ebbles. The higher
beds are fairly massive and break down in large blocks on weathering. The
limestones as a whole are bluish gray to dark in color, compact to crystalline
in texture, and carry abundant fossil remains." The fossils suggest the
Spiriferina pulchra fauna, which is "more or less characteristic of the
Phosphoria formation of Idaho, the Park City formation of Utah, and
the Embar formation of Wyoming." (Girty in Knopf's paper, page 44.)
Girty's advice is: "You had best refer your collection to the Permian and
correlate it with the Park City, Phosphoria, and Embar, though the Park
Cit>- contains some Pennsylvanian and the Embar contains Pennsylvanian,
Permian, and Triassic."

This connects the upper Paleozoic of the west coast of the United States
with the Cache Creek series of the international boundary and probably
through that series with the northern deposits of Montana, Idaho, and
thence south.

B. PER.MO-CARBOXIFEROUS OF NEW MEXICO.

The southernmost extension of the Permo-Carboniferous beds of the
Basin Province is exposed in the Rio Grande Valley, As far south as
Alamogordo red beds occur which can be traced through the State north
nearly to the north line. These have been included in the Manzano group,
which in general includes the

San Andreas limestone.
Yeso formation.
Abo sandstone.

These deposits were described by Lee and Girty' in 1909. The Abo is the
lowest.

(Page 12.) "It consists of coarse-grained sandstone, dark red to purple in
color and usually conglomeratic at the base, with a subordinate amount of shale,

* Knopf, Adolph, A Geological Reconnaissance of the Inyo Range and the Elastern Slope of
the Southern Sierra Ne\-ada, California, Professional Paper No. no, U. S. Geological
Survey, 1918.

' Lee, W. T., and G. H. Girty, The Manzano Group of the Rio Grande Valley, U. S. Geo-
logical Survey, Bull. 389, 1909.

142 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

which attains prominence in some places. This sandstone, together with the
overlying gypsum, apparently constitutes Herrick's 'Permian.' In the classi-
fication here adopted the upper limit of the Abo formation is drawn below the
gypsum for the obvious reason that in many places the overlying or Yeso forma-
tion contains beds of gypsum and gypsiferous shale at several horizons, through
a thickness in some places of i ,000 feet or more.

" Yeso formation. — The Yeso formation * * * lies with apparent conformity
upon the Abo sandstone, and consists of 1,000 to 2,000 feet of sandstone, shale,
earthy limestone, and gypsum. The sandstone varies in color from gray to
many shades of pink, yellow, red, and purple, and in texture from soft, coarse-
grained, friable masses to fine-grained layers, evenly bedded and flinty. The
shales, frequently gypsiferous, are soft, pink to yellow in color, and beds of massive
white gypsum 100 to 200 feet thick occur in many places.

"San Andreas limestone. — * * * It consists essentially of massive limestone,
which is often cherty and poorly fossiliferous, although several localities were
found where fossils are abundant. * * *"

The discovery by Case of Permian vertebrates in the Abo sandstone
northeast of Socorro, New Mexico,^ similar to or identical with those occur-
ring in the El Cobre Canyon and in the Arroya de Agua, Rio Arriba County,
New Mexico, shows that these deposits belong in the Basin Province.

Whether it will later be proven that the Abo is continuous with the
red sandstone of the Pecos Valley around the southern side of the Guadalupe
range is uncertain, but there can be little doubt that the beds occupy an
equivalent or somewhat lower position. Williston^ and Williston and Case'
have asserted their belief based on the stage of development of the forms,
found, that the beds of Rio Arriba County, are younger than those of Texas,
and later Williston* expressed his belief that the El Cobre beds of New Mexico
are the equivalent of the Wichita beds of Texas. Beyond this there is no
evidence of their relative position.

Lee and Girty" believe that these beds are the equivalent of the Texas
beds. They say that an uplift occurred in the Rio Grande Valley imme-
diately preceding the deposition of the Manzano beds; then the red beds
of the lower part of the Manzano group were deposited, which are equivalent
to the red beds of eastern Colorado and to the red beds of Texas.

To the southwest the lower part of the Manzano group is apparently
lacking. Darton has described the Gym limestone which he regards as
the upper part of the Manzano, in two papers. In 1916 he said:*'

' Case, E. C, Further Evidence Bearing on the Age of the Red Beds in the Rio Grande

Valley, New Mexico, Science, vol. 44, pp. 708-709, 1916.
'Williston, S. W., The Permo-Carboniferous of Northern New Mexico, Jour. Geol., vol.

XX, pp. 1-12, 1912.
' Williston, S. W., and E. C. Case, Permo-Carboniferous Vertebrates from New Mexico,

Carnegie Inst. Wash. Pub. No. 181, 1913.

* Williston, in C. R. Stauffer, Divisions and Correlations of the Dunkard Series of Ohio,

Bull. Geol. Soc. Amer., vol. 27, p. 88, 1915.
« Lee, W. T., and G. H. Girty, The Manzano Group of the Rio Grande Valley, U. S. Geo-
logical Survey Bull. 389, 1909.

* Darton, N. H., Geology and Underground Waters of Luna County, New Mexico, U. S.

Geological Survey Bull. 618, 1916.

THE BASIN PROVINCE 143

(Page 35.) " In the central and southeaistem portions of the Florida Moun-
tains and the central portion of the \'ictorio Mountains and extending part
way around the north end of the Tres Hermanas Mountains there is a thick series
of limestones to which it is proposed to apply the name Gym Hmestone. * * *"

This limestone appears on the top of peaks and fault blocks in scattered
areas through the county. It is the uppermost formation of the Paleozoic.

(Page 36.) "The formation consists chiefly of limestone, in greater part
massively bedded, of light-gray color, and showing a breccia ted structure in many
beds. In Gym Peak and vicinity' the lower member is dark and the one next
above it is much lighter in color, with an abrupt change from one to the other,
and the thickness remaining in this area and west of the peak is at least 700 feet.
In the canyon i mile southeast of Gym Peak limestone apparently in the middle
of the formation dips steeply southeastward under 80 feet of dark-gray fissile
shale which is traceable for about half a mile and again appears along the great
fault on the trail a short distance west of Gj-m Peak. This black shale is overlain
on the east by cherty limestone containing abundant Manzano fossils, and this
limestone is finally cut off by the great fault which crosses the mountain. * * *"

In one locality in the Tres Hermanas Mountains the Gym limestone
passes under "gray quartzite, which is the highest member exposed."

The fossils studied by Girty' indicate that the relation is with the
Manzano, but some of the gastropods indicate the Hueco. Later Darton*
said of the Manzano group:

(Page 53.) "The Manzano group is represented in central and northern
New Mexico by the Gym limestone, which crops out extensively in the Florida
Mountains, tj'pe locality*, and also in the Victorio Mountains. * * * The Gym
limestone also appears extensively in the Tres Hermanas Mountains, where it is
uplifted and cut by porphyry, amd it also crops out in a few small hills rising out
of the desert in the south-central part of the county. The formation has not
been recognized outside of Luna Count>', cilthough doutbless it is represented in
the Manzano and Hueco sections in other areas. * * *"

"The formation consists almost entirely of light-gray limestone, mostly
massive and in part brecciated. An 80-foot member of dark-gray shale is
apparently included on the southeast slope of the Florida Mountains, but this
may be the Percha shcile overlapped, or faulted into its present position. In the
Tres Hermanas Mountains part of the Gym limestone is metamorphosed to
white marble and there is included a member of 50 to 60 feet of gray to reddish
quartzite. * * *"

"In the San Andreas and Sacramento mountains and farther north in New
Mexico the stipposed equivalent of the Gym limestone is separated from the
Magdalena group by a thick series of red beds (Abo sandstone), but these beds
are lacking in the southwest comer of the State and also in the region near and
east of El Paiso.

(Page 55.) "* * * Pennsylvanian and Permian time is represented in the
main by deposits of the Magdalena and Manzano groups and the Hueco and Gym
limestones. The Hueco and Gym are contemporaneous, at least in part, with

• Darton, N. H., A Comparison of the Paleozoic Sections in Southern New Mexico, U. S.
Geological Survey, Professional Paper No. 108-C, 1917.

144 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

the Manzano group, which includes 500 to 1,000 feet of red beds (Abo sandstone)
that thin out to the south."

That the red beds of the Rio Grande Valley and western New Mexico
and Arizona, with their associated rocks, belong in a distinct faunal and
depositional province (the Basin Province), the author has already attempted
to demonstrate (Carnegie Inst. Wash. Pub. No. 207, 1915), but the relation
of the Guadalupian and the overlying beds to the uppermost beds of Kansas
(Kiger of the Cimmaron series), Oklahoma (Quartermaster and Whitehorse),
is a matter of uncertainty. Girty, if I understand him correctly, would
place the Capitan limestone entirely above the Whitehorse beds of Oklahoma,
placing them in the Artinskian or Artinskian and Permian; while Beede
regards the distinction in fauna as due to environmental conditions and
regards both as Permian.

Girty discusses the relations at some length in his monograph upon the
Guadalupian fauna. ^ On page 48 he says:

"In passing northward it appears that the Hueco beds, typically consisting
of dark limestones, change their color and lithology, and are represented by red
beds interspersed with limestones. In the Grand Canyon section they appear
as the Aubrey sandstone and limestone, while in Utah the Weber quartzite seems
to be equivalent to them. These correlations are at present provisional. With
still greater reserve are the red-beds faunas of Wyoming correlated with the
Weber on the one hand and the upper part of the Kansas section on the other.
Their relationship with the eastern fauna is far stronger than with the western.
At present I see no evidence of their being younger than the Weber, but they
may be older. Conservatively they may be placed in the same epoch. If we
accept this correlation of the Hueco formation with the Gschelian on the one
hand and the Kansas Carboniferous on the other, the Guadalupian would conse-
quently correspond to the Artinsk or to the Artinsk and the Permian. * * *"

(Page 50.) "It seems to me more probable that the upper Carboniferous
of the Mississippi Valley represents not the Pre-Hueconian alone of the trans-
Pecos and New Mexico section, but the pre-Guadalupian as a whole * * *."

In another place^ he says that the fauna of Hueco may be equal to eastern
faunas, of the Mississippi Valley, but that the Guadalupian fauna certainly
can not, though the difference may be due to environmental conditions.

"Provisionally I am regarding the Guadalupian as younger than any known
faunas of the eastern region, thus interpreting the faunal differences of the
Hueconian when compared with the Pennsylvanian and Permian of the East, as
due to environment rather than to time."

Beede has long contended that the fauna of the upper red beds of Okla-
homa (Whitehorse) is truly Permian and equivalent in time to the Guada-
lupian. In his review of Girty 's monograph he says:'

* Girty, G. H., The Guadalupian Fauna, U. S. Geological Survey, Professional Paper No.

58, 1908.
' Girty, Geo. H., Outlines of Geologic History, University of Chicago Press, pp. 133, 134,

1910.
' Beede, J. W., Jour. Geol., vol. xvii, p. 677, 1909,

THE BASIN PROVINCE 145

"It would seem that the general physical conditions prevailing throughout
the world at the beginning of and during Permian time must be taken into account
in making broad correlations of Carboniferous and Permian faunas. The sig-
nificance of the evolution of a provincial fauna in a great epicontinental sea,
covering 200,000 or 300,000 square miles, with inadequate and perhaps only
intermittent connection with the open sea of the continental shelves in America,
should be as great as the evolution of a fauna in the Uralian region. This sig-
nificance is increased when it is taken into consideration that both developed
during the time when the water was being drawn from the shelves of both con-
tinents and the areas of the inland seas were being greatly reduced.

" In this light the parallelism in the nature of the deposits of the two regions,
accompanied by a like parallelism of faunal changes, is of fundamental impor-
tance, and deserves a larger consideration than Dr. Girty has given it. For
instance, the introduction of new faunal elements, the sudden and nearly com-
plete disappearance of the Fusulina, and the occurrence of Schwagerina bear the
same relations to the early gypsum deposits and the development of the Red
Beds, in the Kansas section, as they do in the eeistern part of European Russia.
If I read the stratigraphic account of the Guadalupes aright, it seems that the
general considerations of the later Permian apply to them likewise. The un-
conformit>% if such it be, carrying away the Capitan limestone from the flanks
of the mountain of which it forms the top, and over the unconformity the deposi-
tion of the Castile gypsum. Rustler formation, and Red Beds, strongly suggest
that the Guadalupe region was similarly affected with the region to the north-
ward so far as a general Permian emergence is concerned. In this light the
Guadalupian faunas must be largely contemporaneous with the Permian faunas
of America and Eurasia. In the eyes of the reviewer, judging from figures and
descriptions only, there is where their faunal relationships would also place them.

"The point is made that the faunas are so different that, if they axe. contem-
poraneous with those of the Mississippi Valley— of which Dr. Girty seems to be
doubtful — they could not both be covered by a single general term (like Permian?)
for their designation. That they are quite distinct from anything yet brought
to light on the continent will be granted at once by anyone familiar with the
subject. The one is a cosmopolitan, open-sea, coastal-shelf fauna, while the
other is a more isolated epicontinental sea fauna rather thoroughly separated
from its neighbor on the south and perhaps belonging to a different climatic
zone. Should they prove to be equivalent in time, I see no reason why they
might not be covered by a single term of ordinal rank, their local geologic designa-
tions being sufficient to differentiate them.

"That it was impossible for the Guadalupian and Mississippi Valley clear-
water faunas to intermingle to a considerable extent after the time represented
approximately by the Topeka limestone, unless by a circuitous route, no one
acquainted with the geology of the intervening region would hesitate to state."

Beede's explanation of the impossibility of the intermingling of the
waters of western Oklahoma which were already depositing red beds and
the clear waters of the sea which deposited the Capitan limestone are
quoted on page 100. To the author there is considerable difficulty in
accepting either of these explanations.

The red beds of the Pecos Valley are in all probability of equal age with,

and the same in stratigraphic position as, the uppermost red beds of southern
11

146 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Texas, which are, at least, no higher than the Double Mountain, and these,
with the Castile gypsum and Rustler limestone, lie upon an erosion surface
which truncated the Delaware limestone and (in all probability) the Capitan
limestone, but they are below the Greer and Quartermaster (Whitehorse)
of western Oklahoma. Beede says of the Guadalupian and overlying beds:^

"The stratigraphic relationships of these Permian beds are peculiar and
interesting. They are brought to the surface by a westward-facing fault-scarp
as it dies out into a fold to the south. Other mountains occur to the west and
northwest with older faunas, and only in this one locality is the nearly full section
of the Guadalupian rocks shown. The Capitan limestone (white Permian lime-
stone of Shumard) is 1,700 or 1,800 feet in thickness. Below this is the Delaware
Mountain formation, composed of dark limestones and sandstones with a black
limestone 200+ feet thick beneath it, giving, all told, some 2,500 feet to this
formation and a total of about 4,000 feet to the whole Guadalupian section as
shown at the southern extremity 'of the mountains. The stratigraphy was largely
worked out by Richardson. To the east, on the dipslope of the mountain, the
Capitan limestone is wanting. An erosional unconformity is found on the Dela-
ware Mountain formation upon which rests the Castile gypsum. The exposures
of the region show 50 or 60 feet of it and a well at Rustler Spring penetrated it to
a depth of 300 feet. To the east, and upon this, lie the Red Beds."

Richardson says:^

"The Castile gypsum along its western outcrop lies in little knolls and valleys
of the underlying Delaware Mountain formation, indicating an erosional un-
conformity. Another evidence of unconformity at the base of the gypsum con-
sists in the absence of the Capitan limestone. It appears that either the gypsum
was deposited at or near the top of the Delaware Mountain formation as a lens
which did not extend westward to intervene between the Delaware Mountain
formation and the Capitan limestone in the Guadalupe Mountains, or that
erosion removed the former southwestward extension of the limestone (the thick-
ness of which is unknown) before the deposition of the gypsum. The former
supposition necessitates the correlation of the Rustler formation, which overlies
the gypsum, with the upper part of the Delaware Mountain formation or with
the Capitan limestone. But there is little to support this interpretation, and it
is tentatively assumed that the Castile gypsum and the Rustler formation were
formed after the deposition and erosion of a part of the Capitan limestone."

C. PERMO-CARBONIFEROUS OF ARIZONA.

The red Abo sandstone of New Mexico carries vertebrate fossils very
similar to those from El Cobre Canyon and Arroya da Agua, and this sand-
stone is revealed to the west in the Jemez uplift and so carried west to the
region of Fort Defiance, Arizona, whence it can be traced south to Fort
Wingate in New Mexico and the Grand Canyon region. The sandstone in

* Beede, J. W., Review of the Guadalupian Fauna by Geo. H. Girty, Jour. Geol., vol. xvn,

p. 672, 1909.

* Richardson, G. B., A Reconnaissance in Trans-Pecos, Texas, University of Texas Mineral

Survey Bull. No. 9, p. 43, 1904.

THE BASIN PROVINCE 147

northern and central Arizona is the Moenkopie. Darton has g^ven a
general review of the Permian of this region,^ with a map of the outcrops.
He describes the Moenkopie as a mass of shales and sandstones, generally
of a red color and ver>'^ variable in different sections — terminated at the top
by a conglomerate, the Shinarump, which is considered as Triassic.

The following general section of the Moenkopie on the Little Colorado
River is quoted from Ward by Darton:

Feet.

Dark chocolate-brown shales de%-oid of grit and highly charged with salt and gypsum 200

Dark-brown soft argillaceous sandstone loo

Dark-brown shale, highly saliferous and with gypsum layers; becomes calcareous below 200

Shale, mostly white * 100

Brown shale, similar to those above; saliferous lOO

Carboniferous limestone.

"The sandstones occur at various horizons and locally attain a thickness of
100 feet, with more or less intercalated shale. They are mostly soft, and weather
in irregular rounded ledges. The gy-psum occurs largely in thin veins, crossing
the strata at various angles. Toward the base of the formation the shale is
calcareous and nearly every^vhere includes a bed of limestone that merges into
the inclosing strata."

The same formation occurs in the Zuni Uplift, at San Jose, Ojo Caliente,
and Jemez.

These beds extend westward beyond the San Francisco Mountains until
the last remnants appear in small red hills between Ash Fork, Arizona, and
the rim of the Grand Canyon.

A more detailed account of the Permian of northern and northwestern
Arizona appears in Gregory's description of the Navajo Country.* He
says on page 23 :

"In mapping the geology' of the Navajo country it was found that strata of
Permian (?) age are more \N-idely extended than had pre\nously been supposed.
They occur not only in the Littie Colorado Valley, but along the San Juan and
at a number of localities on Defiance Plateau. In the western part of the reserva-
tion they mark the beginning of the red beds and are easily distinguished as a
whole from the underlying Kaibab by abrupt changes in color and in comf>osition."

On page 24 is given a "section of Moenkopi formation in the wall of
Little Colorado Canyon, about 5 miles below Tanner Crossing, Arizona,"
as follows:

Shinarump conglomerate; gray and mottled, cross-bedded; pebbles of quartz, quartzite, calcareous
shale, and petrified wood.

Unconformity; marked by sudden transition of shale to conglomerate and by wavy, irregular con-
tact, including pockets in shale filled by sandstone and conglomerate.

1. Shale, red; bleached white at top, arenaceous and argillaceous, compact, hard, of mtcroscoiMC

fineness; w^eathers into rounded disks 3

2. Shale, red-brown and gray banded, argillaceous, lenticular, with lenses of sandstone at bottom ... 25

' Darton, N. H., A Reconnaissance of Parts of Northwestern New Mexico and Northern

.Arizona, U. S. Geological Surrey Bull. 435, 1910.
* Gregory, H. E., Geology of the Navajo Country, U. S. Geological Survey, Professional

Paper No. 93, 1917.

148 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

3. Shale, red-brown, argillaceous, and thin cross-bedded, like grass or stems of reeds arranged in

masses; contacts not exposed 16

4. Sandstone, chocolate-colored to red, calcareous, extremely fine 3

5. Shale, brown and maroon, arenaceous and calcareous, ripple-marked and mud-cracked 2

6. Sandstone, red-brown, massive, extremely fine-grained, micaceous 2

7. Shale, banded chocolate-colored and red-gray, arenaceous, becoming argillaceous at the top; tiny

veins of gypsum 11

8. Sandstone, red-lsrown, with white band at top, composed of extremely fine quartz grains with

flakes of muscovite; plant impressions 3

9. Shale like No. 27 8

10. Sandstone, chocolate-colored, fine-grained, micaceous, thin-bedded at top; forms bench 6

11. Shales, banded red of various shades and gray-green, arenaceous, argillaceous, calcareous 26

12. Sandstone, thin-bedded, ripple-marked I

13. Shale, red-brown, with streaks of purple and green-gray and blotches of white; includes lenses of

sandstone; bedding planes sun-baked 20

14. Sandstone, red-brown, thin-bedded, cross-bedded, marked by ripples, mud cracks, and worm

casts; muscovite on bedding planes; plant impressions on lumpy uneven surfaces include
striated and radiating groups 2

15. Shale, banded chocolate-color, red, gray, and green, arenaceous and calcareous at bottom, argilla-

ceous at top, ripple-marked, mud-cracked, lenticular 10

16. Sandstone like No. 14 2

17. Shale like No. 15 12

18. Sandstone in 3-inch beds, wavy, irregular, lenticular, cross-bedded, ripple-marked 4

19. Shale, red-brown, arenaceous, calcareous, banded and lenticular; shows worm casts, sun-baked

surfaces, and impressions of plants 7

20. Sandstone, red-brown, thin-bedded lenticular, with bands of argillaceous and calcareous shale;

mud-cracked, and ripple-marked; plant impressions 2

21. Shale, banded dark red, light red, and purple, with elliptical blotches of green-white, yellow-

green, and ash-gray, calcareous, argillaceous, and arenaceous 18

22. Sandstone like No. 20 2

23. Shale like No. 21 ; bands of color one-half inch to 3 inches thick 5

24. Sandstone like No. 20 2

25. Shale like No. 21 2

26. Sandstone, chocolate-colored, highly calcareous, with scattered limestone pebbles 3

27. Shales, chocolate-colored to red, with gray and lavender lenses; arenaceous, imbricated, ripple-

marked; 15 feet from the bottom is a 6-inch bed of sandstone, and thin sandstone lenses
occur throughout; the top 10 feet is dark-red argillaceous shale in regular beds traversed by
veins of gypsum, probably of secondary origin 40

28. Sandstone, chocolate-red, with streaks of maroon and purple; fine-grained quartz with calcareous

cement; cross-bedding both angular and tangential; fine to medium grained; size of grain
varies with each lamina. Near middle of bed are lenses of conglomerate, 2 inches to 12 feet
wide, 6 inches to 100 feet long, highly irregular in shape, and composed of chunks and slabs of
argillaceous shale, sandy shale, and sandstone; muscovite abundant on bedding plane;
forms vertical cliff 52

29. Shales, chocolate-colored with white bands; arenaceous and micaceous strata, thin as cardboard

or 2 to 3 inches thick; show ripple-marked, mud-cracked, sun-baked surfaces, curled disks,
tiny folds, and faults 100

389

(Page 25.) "The bedding is very irregular throughout. Strata of shale and
sandstone appear and disappear along the strike, and individual laminae within
the beds are markedly discontinuous. Arenaceous beds prevail, typical clay
shales are very rare, and pure limestone is absent. Mr. Heald noted that the
strata became increasingly calcareous upward until bed No. 3 is reached.
Gypsum in tiny horizontal and vertical seams is common. A large part of it,
perhaps all, is secondary. Several small unconformities were noted, but no
hiatus that necessarily involved a long period of corrasion or of weathering.
Part of the color banding* appears to be genetically related to conditions of deposi-
tion ; much of it is better explained as due to leaching by ground water. Frequent
exposure to the atmosphere as the Moenkopi beds were forming is indicated by
the almost universal presence of sun-baked surfaces and ripple-marks. Plant
impressions are common and appear to represent several different species."

THE BASIN PROVINCE 149

In discussing other sections related to this, Gregory makes frequent
reference to "mud-lumps" and fragments of shale in the different layers.
He also speaks of the discontinuity of the layers (p. 27): "All the beds in
this section [2 miles east of Holbrook, Arizona] "decrease and increase within
short distances along their strike, and most of them retain their individuality
only for a few tens or a few hundreds of feet. It is difficult to locate equiva-
lent strata in two sections measured a mile apart."

The beds are prevailingly quartz sands and the limestone present is in
lumps, grains, lenses, and concretionary layers. The g^-psum is ver>' largely
secondary in origin.

(Page 30.) "The Moenkopi formation is assigned to the Permian (?) on
both stratigraphic and paleontologic evidence. It possesses essential unity in
structure, texture, color, composition, and conditions of sedimentation. An
erosional unconformity with the Kaibab limestone marks its lower limit in the
Little Colorado Valley of Arizona; smd though clear evidence of such relation
has not been obtained in the San Juan region, the fossiliferous Goodridge beds
are separated from the Moenkopi by a sharp lithologic break. The Shinarump
conglomerate (Triassic) unconformably overlies the Moenkopi or the DeChelly
sandstone, which is also assigned to the Permian (?). The paleontologic e\adence
obtained both within the Navajo Reservation and along its borders is conflicting.
Fossils collected on the rim of the Little Colorado Canyon by Mr. Pogue include
many fragmentary bivalves and some gastropods. Professor Schuchert reports:

" T see Bakewellia, Pinna, Schizodus, and Bellerophon. The horizon is clearly
above the Pennsylvanic and is the Permic molluscan fauna devoid of brachio-
pods. The horizon may be high in the Permic, that is, above Lower Permic, as
the term is understood in America, say about Middle Permic'

" Fragmentary plant remains, including species of Walchia, were collected at
a number of localities and in 1913 E. C. Case and W. B. Emery, of my party,
obtained determinable plant fossils from the middle Moenkopi beds 3 miles west
of Fort Defiance. Regarding this collection David WTiite writes:

"'The large fragment with closely placed lateral twigs belongs to another
Walchm resembling Walchia hypnoides. It is perhaps identical with that de-
scribed by Dawson as Walchia gracilis. One or two small fragments in one of the
loose rock pieces agrees still more closely with Walchia gracilis. These forms of
Walchia are characteristic of the Permian and are present in Okleihoma and in
the Wichita formation of Texas.' "

The beds of the Kanab Valley described as Permian by Walcott seem
now to be very definitely assignable to the Triassic (Meekoceras beds) and
equivalent to the Permo-Carboniferous of the Uintah Mountains.

(Page 31.) "On the northern flanks of the Zuni Mountains, near Fort
Wingate, Dutton found ' several specimens of Bakewellia and an attenuated form
of Myalina corresponding to the forms of the latter genus which are common
in the Permian.' The description of the strata from which these fossils were
obtained indicates their equivalency with the Moenkopi at Fort Defiance and
elsewhere. S. W. Williston states that 'there are genuine Permian red beds' in
the Zuni Mountains and that 'a Paleozoic brachiopod was obtained by Mr.
Miller in the (Moenkopi) cliffs at Holbrook.'

150 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

" In its stratigraphic position the Cutler formation of the San Juan Mountains
corresponds to the Moenkopi formation, and the two are lithologically somewhat
similar."

The Be Chelly sandstone. — A massive, cross-bedded sandstone peculiar
to the Navajo Country is named by Gregory the De Chelly sandstone.
This had previously been correlated with the Vermillion Cliffs or the Wingate
sandstone.

(Page 32.) "Section of De Chelly sandstone at west entrance of Bonito Canyon, near Fort Defiance.

"[Measured by K. C.Heald. Dip, 16° E.)

"Shinarump conglomerate.

"Unconformity. Feet.

1. Sandstone, light red, fine-grained; clear- white and red rounded quartz grains; calcareous and

ferritic cement; contains rare pebbles one sixty-fourth to one-sixteenth inch in diameter;
massive, cross-bedded in places; weathers into rounded knobs; in two beds, 13 and 15 feet
thick 28

2. Sandstone, tan to brown, fine-grained; clear, well-rounded quartz; calcareous cement; many

specks of limonite; even-bedded to slightly cross-bedded; hard; forms nearly vertical clifT;

in three beds, 7, 3, and 6 feet thick 16

3. Sandstone, chocolate-colored to gray-brown, fine to medium grained; clear, well-rounded quartz;

massive; parts of the bed show no structure; other parts cross-bedded with curved laminae
tangential to a horizontal surface; weathers in rounded bosses 77

4. Sandstone, chocolate-colored, shaly, largely concealed by talus 27

5. Sandstone, light red, fine-grained; clear to red rounded quartz grains; bottom 5 feet thin-bedded;

in the center gray, cross-bedded, resistant sheet, I J feet thick; remainder massive, incon-
spicuously cross-bedded 115

Moenkopi shales.

263

This section is characteristic of the De Chelly sandstone in a general
way wherever it occurs. It is mentioned as occurring at Defiance Hogback,
Buell Park, lower Black Creek, Canyon De Chelly, San Juan Valley in
Utah, Monument Valley, etc.

(Page 33.) "Structure, texture, and composition [of De Chelly sandstone]:

"With the exception of the Navajo and Wingate sandstones, which it resem-
bles in many physical features, the De Chelly sandstone presents the most
massive strata of all the red beds. In the wall of Canyon Bonito beds 5 to lo
feet thick are found near the top; but in the same locality 60 to 70 feet of strata
with most obscure bedding stand vertically in the wall. At Oljeto a single bed
is 85 feet thick, and in Canyon del Muerto and Canyon De Chelly there are
massive beds 200 to 300 feet thick, with no definite planes of separation. Here
and in Monument Valley giant slabs of rocks splitting off from the massive beds
leave clean, smooth faces hundreds of square feet in area, marked only by the
delicate tracery of cross- bedding laminae.

"The De Chelly sandstone is fine-grained throughout and remarkably uniform
in texture. It consists essentially of grains of two sizes — spherical grains of
quartz, averaging about 0.19 mm. in diameter and making up the bulk of the
rock, and less well-rounded grains 0.5 to 0.6 mm. in diameter. In places grains
of the two sizes are intermingled, but commonly the larger grains are sprinkled
over the surfaces of cross-bedding laminae. Here and there slightly larger grains
are found as lenses or strings marking cross-bedding division planes, and rarely
scattered pebbles one-sixteenth to one-eighth inch in diameter are seen. White
rounded quartz grains constitute about 95 per cent of the rock; red and amber

THE BASIN PROVINCE 151

quartz grains are also found, but the prevailing light-red to red-yellow hue of the
strata is maintained chiefly by the ferritic pigment which, with calcite, constitutes
the cement. Light-colored specks of kaolin are present in the hand specimen
and in places gi\'e the rock an appearance of a mixture of salt and cayenne pepper.
Mica and black quartz are also sparingly distributed.

"Cross-bedding is a characteristic feature of the De Chelly sandstone. Here
and there the entire wall of a canyon consists of interlocking cur\ed beds; else-
where massive cross-bedded strata are replaced along the strike by horizontally
foliated sandstones. The cross-laminae may be a foot or more in thickness, but
usuallj' they measure less than an inch and in many places the division planes are
so closely spaced that the structure is concealed, the rock surface being completely
overspread by a lace-work of intricate curves. Typically the canyon walls in
the De Chelly sandstone are marked by sweeping curved bands 20 to 200 feet
long, tangential to a horizontal surface and flatly convex upward.

"The De Chelly sandstone is traversed by wide-spaced joints which, together
with the curved cross-bedding foliation, allow the agents of erosion to carve
alcoves, recesses, and tunnels in great variety and on a scale that ranges from
ornamental pockets to great arched-roof alcoves in which, high on the canyon
walls, are tucked away single houses or whole villages of cliff dwellings.

"Physiography of Pekuan Tme.

"Gilbert conceived the whole plateau country as 'covered by an inland sea
entirely separate from the ocean * * * from the close of the Carboniferous to
the beginning of the Cretaceous.' As a result of studies in the Grand Canyon
r^on Walcott reached the conclusion that —

" ' It is probable that the era of the deposition of the Permian was one of slow
movement of the sea bed. Elevation and depression are indicated strongly by a
marked unconformity, by erosion, in the lower portion of the upper Permian.
* * * The sediments are mostly detrital in chju^cter, and ripple-marks jmd other
indications of a littoral deposit are also seen at several horizons.'

" Robinson considered the Moenkopi of the San Francisco Mountain volcanic
area as ' flu\4atile or lacustrine' in origin. Huntington and Goldthwciit concluded
that ' the Moenkopi series was probably laid down in a shallow sea where estuarine
conditions may possibly have prevailed.' In a later paper Huntington ascribed
these beds to alternate lacustrine and subaerisJ deposition incident to the expan-
sion and contraction of waters of a lake contained within an inclosed desert basin.

"The Plateau Province during Permian (?) time was probably a r^on of low
relief bordering the sea and having an arid climate. Over the lohg slopes and
into the flat-floored depressions, sediments were carried from surrounding lands
and deposited on flood plains, piedmont slopes, and the floors of fresh and alkaline
lakes. The remarkable banding of subequal dimensions displayed in certain
localities and so vividly described by Dutton indicates cycles of change of roughly
equal length. In some places the sediments suggest change from subaerial to
lacustrine deposition; in others marine strata are interbedded with materials of
flood slopes. Deposits of gypsum alternating with ripple-marked beds of lenticu-
lar sand point to fluctuation in ^•olume of the water contained by ephemeral lakes.
Ancient plaj-as, deltas, and flood plains are suggested by rain prints, mud cracks,
and the almost universal presence of shining films of clay and mica and halite
pseudomorphs that coat the planes of foliation. The general absence of fossils,
other than fragments of vertebrates and xerophilous plants, is suggestive of

152 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

continental conditions. Aridity is suggested by the presence of feldspars and
by the prevailing reds and browns of the rock, which are inconsistent with the
presence of ground water near the surface. The cross-bedding also points to
aridity, for while angular cross-bedding of types common on alluvial plains and
even on the seashore occurs in many strata, tangential cross-bedding of the eolian
type is prevalent.

"Toward the end of the Permian epoch aridity reached a stage where sand
dunes became a prominent feature. These are best preserved along the east
and northeast sides of the area, where the De Chelly sandstone displays the record
of wind work during late Permian (?) time. The walls of Canyon de Chelly
consist in part of overlapping heaps of wind-blown sand now weakly cemented
into rock. In the picturesque Navajo language they are 'frozen dunes.'

"The exact sequence of events during Permian (?) time has not yet been made
out, but the final explanation must allow for extensive subaerial sedimentation
under arid conditions and for two or more invasions of the sea. It is not neces-
sary, however, to assume that all parts of the great area in which Permian
deposits occur had the same physiographic history."

In a recent paper upon the Carboniferous of the Grand Canyon,
Schuchert^ has suggested that the Kaibab limestones, the Coconino sand-
stone and even the upper part of the Supai formation are Permian. He says :

"That the Kaibab limestone is of early Permian age is now admitted by most
American stratigraphers. This view, however, has been attained rather from
its field relations than through a study of its marine fossils, for these in several
forms are very much like those of the Pennsylvanian. The fauna as collected
by Noble in the Shinumo quadrangle is listed by Girty and he here correlates
the Kaibab limestone with the Manzano group of New Mexico. He also suggests
that the Kaibab may be equivalent to a part of the Guadalupian of southwestern
Texas, a formation of undoubtedly Permian age."

The remarks made by Girty scarcely seem to convey to the author so
definite an idea of the Permian age of the Kaibab as is received from them
by Schuchert. Girty says in a letter to Noble^ concerning the fauna collected
from the Kaibab:

"The list is typical of the fauna of the upper Aubrey, the general character
of which has long been known through similar lists made up by Meek and others.
I have been tentatively correlating the Aubrey with the Manzano of New Mexico
and with the upper part of the Hueco formation of western Texas. Consequently,
it would be older than the Guadalupe group, which overlies the Hueco formation.
The fauna listed above, however, contains a number of species which are very
similar to or identical with species which occur in the Guadalupian fauna, and
in spite of the fact that most of the Guadalupian species have not been found in
the Aubrey group, it seems less improbable than it did several years ago, when the
Guadalupian fauna was under investigation, that the Kaibab limestone is of the
same geologic age."

1 Schuchert, Chas., On the Carboniferous of the Grand Canyon of Arizona, Amer. Jour.

Science, vol. XLV, p. 347, 1918.
* Girty, in L. F. Noble, The Shinumo Quadrangle, U. S. Geological Survey Bull. 549,

p. 71, 1914.

THE BASIN PROVINCE 153

Schuchert regards the Coconino sandstone as, in part at least, formed
from wind-blown sand derived from the north and northwest:

[This sand] " should be expected in near-shore deposits of Permian time because
of the then prevalent arid climates. The eolian sand, it appears, has been blown
into rivers that have brought it from a long distance to the northward and out of
it in the course of transportation has been washed or blown almost all other dis-
integrated rock material than the quartz. * * *" [The conclusion as to the origin
of this sandstone reached by the writer while in the field is that it represents the
material of a large delta of continental deposit laid down under constant but
probably local sheets of water that were evidently entirely fresh. The Coconino
may be the deposits of dunesands swept from the north into a series of basins or
fresh-water lakes like the present fresh- and brackish-water lakes on the outer
borders of the Nile delta.]

"That the Coconino sandstone Invaded to the southward a land composed
of the Supai formation is shown not only in the very different nature of these
underlying strata and the sharp contact between them, but especially in the
fact that the surface of the Supai has memy vertical solution joints now filled
with the Coconino sands."

On page 352 Schuchert makes the suggestion that —

"It may well be that the marine Kaibab limestone and the Coconino sand-
stone toward the east change finally into desert dune deposits and that the De
Chelly is the time equivalent of more or less of the Moenkopi, Kaibab, and Coco-
nino formations.

"In the Upper Supai have been found plant remains which David White has
determined as Callipteris cf. sp. C. conferia; Walchia cf. W. gracilis; Gigantopteris ?
cf. Sphenophyllum."

David White, in a letter quoted on page 354 of Schuchert's article, says :

"The condition of preservation of the frcigments is so bad that caution is
necessary in basing conclusions of any kind on the material submitted. How-
ever, the presence of Gigantopteris, Walchia, and probably of Callipteris, if my
tentative generic identification of the latter is correct, points to the Lower
Permian age of the flora. * * * In any event, it appears probable that the flora,
when it is better known, will be found to indicate a level not below the highest
stage of the Pennsylvanian."

Schuchert notes the occurrence of these plants further east in Arizona
and concludes:

"It should be noted that these fossils are found immediately above a marked
erosional unconformity. If, therefore, we give full significance to this uncon-
formity, and with it bolster up White's provisional conclusions as to the age of
the plants, the upper 290 feet of the Supai are to be referred to the Permian
system."

154 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

D. PERMO-CARBONIFEROUS OF SOUTHWESTERN COLORADO.

The upper Paleozoic series of southwestern Colorado is generally con-
sidered as comprising the

Cutler — Permo-Carboniferous.
Rico ]

Hermosa [- Pennsylvanian.
Molas J

The Moenkopie of northwestern Arizona has been shown to be similar
or identical in stratigraphic position with the Cutler of southwestern Colo-
rado. The most complete description of the red-bed series in the latter
region was given by Cross and Howe.^

"The Cutler formation embraces somewhat more than the lower half of the
Red Beds section of southwestern Colorado. Its strata are invariably red in
color and include sandstone, arkose grit, conglomerate, shale, and limestone.
The maximum observed thickness is about i,6oo feet.

"The formation seems conformable with the underlying Pennsylvanian beds,
but above it occurs a stratigraphic break with at least local unconformity. The
base of the formation is indicated by the Pennsylvanian fossils of the Hermosa or
Rico formations and in a broad way by the color line. No fossils have been found
in the Cutler beds.

''Details of lithologic character. — Great variability in lithologic constitution,
both vertical and lateral, is one of the most striking features of the Cutler forma-
tion. The sandstones are sometimes fine-grained and massive, but bedding is
ordinarily distinct and few homogeneous beds exceed lo or 15 feet in thickness.
All strata are calcareous, and the finer grained sandstones grade into calcareous
shales and impure marls or into sandy limestones. These rocks are naturally
more or less friable and crumbling.

"The finer-grained strata are of the strongest red color, which is due to a
ferritic pigment, and they are also commonly characterized by abundant bronze or
rusty mica, which renders them fissile. Clay beds are rare, as is massive limestone.
Commonly the more calcareous strata are nodular or gnarly and grade into calca-
reous sandstones. Greenish and grayish tints are locally found in the nodular lime-
stones and a mottling with red is common. Some of the nodular limestones
appear to be intraformational conglomerates.

"The sandstones frequently grade into arkose grits and these into conglomerate.
With increasing coarseness of grain the red changes to pink, and locally beds of
coarse grit are gray or almost white. In other cases the finer matrix of grits and
conglomerates is dark red. The cement of the strata is calcite, and most of the
conglomerate and arkose beds are comparatively resistant to weathering and form
prominent ledge outcrops on all steep slopes.

"The grit beds often reach 35 feet in thickness. They are variably massive,
being in some places almost homogeneous from top to bottom, while more fre-
quently divided by several thin shale or sandstone layers. Cross-bedding is
almost universal. Sporadic pebbles are present in all grits, and with their in-
crease the stratum becomes a conglomerate.

^ Cross, Whitman, and Ernest Howe, Red Beds of Southwestern Colorado and Their Cor-
relation, Bull. Geol. Soc. Amer., vol. 16, p. 461, 1905.

THE BASIN PRO\TNCE 155

"The sandstones are mainly quartzose, the grits contain much feldspar, mica,
and small f>ebbles like the larger ones of the conglomerates. The latter contain
pebbles of granite, gneiss, and various schists, of quartzite and limestone, of
greenstone and porphyry, and many of red, pink, smoky, or white quartz, part
of which may come from veins.

"The pebbles are in general larger near the San Juan mountains. Boulders
a foot in diameter are occasionally present, but most pebbles are only a few inches
in diameter. The relative abundance of difiFerent rocks among the pebbles varies
according to locality'. * * *

" Taking the formation as a whole, the grits and conglomerates comprise about
one-third or less of its total thickness in the quadrangle surv^eyed, and they are dis-
tributed throughout the section. It maj' be assumed that as distance from the
source of the pebbles increases, the formation becomes more and more a series of
fine-grained scmdstones and shales, with subordinate grits and conglomerates.

" Typical sectioit composed of heo sections made in the Dolores vaOey a few mUes below Rico.

Top. Feet.

60. Coarse sandstone or grit, cross-bedded, locally conglomeratic, rather purpfeh in tone 100

59. Calcareous sandstones, sandy shales, often micaceous and fis^e, either red or mottled red and

green in color 120

58. Grit -conglomerate, of variable texture, fwining ledge outcrops 10

57. Fine-grained calcareous sandstones, sandy shales, with occasional thin layers erf harder sandstones

in red or variegated red and green 80

56. Sandy shales, ■with thin sandstones at intert-als; variegated or mottled dark red and light green;

have peculiar nodules 55

55. Grit -conglomerate 20

54. Calcareous sandstone or sandy shale 100

53. Grit -conglomerate, very similar to number 60 30

52. Friable sandstone 35

51. Grit-conglomerate, hard and forming a prominent ledge outcrop 15

50. Sandy shales and crumbling sandstone, strong red color, partly cakareous no

49. Calcareous shales 20

48. Coarse arkose sandstone 30

47. Calcareous shales 30

46. Sandstone variegated 20

45. Calcareous shales 15

44. Coarse sandstone, friable, variegated 15

43. Hard standstone, with few small pebbles of quartnte 15

42. Dark-red sandy shales; in places calcareous and then massive; in other parts micaceous and then

fissile 90

41. Compact arkose sandstone, n-ith few pebbles; cross-bedded; friable sandstones near top 15

40. Calcareous clay, or marl, reddish, with spots of various shades 33

39. Friable sandstone, largely quartzitic, x-ariegated in color 15

38. Conglomerate and arkose grit; most conglomeratic in center; a grit with some pebbles in upper

and lower portions; pebbles of greenish slates and schists, quartzite, granite, and greenish

porphyry 30

37. Calcareous shales, fine-grained, variegated 30

36. Fine-grained, reddish shale 15

35. Mas^\-e sandstone, coarse-grained, cross-bedded, streaked with light-colored layers; some shaly

partings 15

34. Friable sandstone, purplish and graj-ish layers alternating im^ularly 20

33. Sandstones, finely laminated, compact, cross-bedded, \-ariegated 25

32. Calcareous clay shales, graduating into more massive sandy shales and a rather tough sandstone. 10

31. Sandstone, micaceous, red or green, carrying a thin layer of coarse congloniefate near the top. ... 10

30. .\rkose sandstone, cross-bedded, white 20

29. Sandstone, micaceous, red and green 6

28. -Arkose sandstone, coarse grained, white 7

27. Sandstone, compact, micaceous, salmon coltH* to red 10

26. Calcareous shales, green and brown 5

25. Sandstone, micaceous and compact, containing layers of gnarly linoestone from 2 to 12 inches

thick 18

24. Calcareous sandstone, generally red, but mottled; contains thin layers of gnarly limestone 15

156 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

23. Arkose sandstone, cross-bedded and conglomeratic in part; this layer is purple at the base, and
this color alternates with greenish-yellow and red bands; in the upper part a white and cream

color prevails 30

22. Sandstone, rather flaggy, containing a layer of gnarly limestone at the base 18

21. Calcareous sandstone, shaly in the lower part and containing gnarly limestone near the top; color,

bright red lO

20. Sandstone, micaceous, massive, becoming flaggy at the top, where it contains limestone nodules;

color, red , 5

19. Calcareous sandstone, containing small limestone nodules near the top; color, red 5

18. Calcareous sandstone and red arkose, containing limestone pebbles; purple and white in the

upper part; red below 12

17. Covered 12

16. Arkose sandstone, variegated, red, white, and purple 6

15. Sandstone, micaceous, flaggy 5

14. Arkose sandstone, flaggy, becoming finer grained and micaceous in the upper part; color, pink, with

narrow white bands 15

13. Arkose sandstone, somewhat conglomeratic in the upper part, and with thin shaly bands near the

center and at the top; color, white and pink, irregular bands 35

12. Sandstone, micaceous, flaggy, very thin bedded at top; color, red 10

II. Calcareous sandstone; a few thin layers of nodular limestone; color, red 7

10. Sandstone, micaceous, flaggy; color, red 6

9. Sandstone, micaceous, flaggy, becoming more compact in upper part 15

8. Arkose sandstone, white, banded with red 7

7. Calcareous sandstone, rather poorly exposed in the upper part, but flaggy and somewhat shaly,

becoming more shaly in the lower part, and containing nodules of limestone; color, red 35

6. Arkose sandstone, red above, white below 6

5. Shale, probably calcareous; contains nodules of limestone; color, red 5

4. Sandstone, micaceous; color, dark purplish red 5

3. Calcareous sandstone, irregularly nodular and flaggy; contains gray nodules of limestone, but no

well-defined limestone ; color, red 25

2. Arkose sandstone, micaceous, thin conglomeratic, cross-bedded; near the base there is a thin

black shale which is quite variable; color, white, with bluish and green zones 15

I. Calcareous sandstone, with a gnarly gray limestone at the top; color, red 10

Total 1,508

The most complete section of the Cutler occurs in the Dolores Valley,
below Rico, Colorado.^ There is exposed 98 feet of the formation, the
upper 32 to 35 feet are arkosic and the rest shale and sandstone with a
subordinate amount of limestone; a section is given on page 48 of the
article cited. The space immediately below the Cutler is covered and it is
not known whether the Rico is present or not.

The Hermosa is made up of 490 feet of shale, sandstone, limestone, and
grit. (Section on page 48 of Cross and Larsen.)

The Molas is at least 100 feet thick, mostly calcareous shale. There is
much of the formation that is "almost a red clay, but 'CVhere the calcareous
element is prominent the mass becomes irregularly nodular or lumpy."
Layers or lenses of limestone are rare. This formation is persistent through-
out the San Juan region.

In 1910, Cross and Spencer^ first described the Rico formation:

"It is here proposed to apply the name Rico to a formation assumed to be
about 300 feet in thickness, occurring between the Hermosa or characteristic
Pennsylvanian Carboniferous and strata assigned at present to the Trias of the

' Cross, Whitman, and E. S. Larsen, Contributions to the Stratigraphy of Southwestern
Colorado, U. S. Geological Survey, Professional Paper No. 90, p. 39, 1914.

' Cross, Whitman, and A. C. Spencer, Geology of the Rico Mountains, Colorado, 21st
Annual Report U. S. Geological Survey, part 11, p. 59, 1900.

THE BASIN PROVINCE 157

San Juan region — the Dolores formation. It is made up of sandstones and con-
glomerates with intercalated shales and sandy fossiliferous limestones. In its
lithological features it resembles the strata immediately above it, but its fossils
are distinctly of Paleozoic age, and while many of its forms are common to the
Hermosa formation, others are of Permian t>'pe, so that it seems proper to desig-
nate its age Permo-Carboniferous, to indicate that it is transitional between
these divisions of the Carboniferous system. In the Rico region the formation
is conformable upon the Hermosa and is followed by the Dolores vnth seemingly
perfect parallelism of stratification. The fauna as a whole has an aspect quite
different from that of the Hermosa, since it is largely composed of lamellibranchs
as opposed to the brachiopod assemblage of the lower formation. The boundcuy
betn-een the Rico and Dolores formations is at present entirely artificial, being
based upon the highest known occurrence of the Rico fossils. The former is
made to include only strata characterized by the Rico fauna, while the latter
comprises the apparently unfossiliferous medial portion of the Red Beds, together
with the upper part, of knowTi Triassic afiinities. The actual age of the un-
fossiliferous Red Beds is thus left in doubt; they may eventually prove to be
either Permo-Carboniferous, true Permian, or Trias. They corresp)ond to [a
part of] what has been called Trias throughout the Rocky Mountain province."

In their 1905 paper, Cross and Howe^ say of the Rico formation:

"The views expressed in the Rico report on the age and relations of the Rico
formation were based mainly on the opinion of G. H. Girty as to the invertebrate
fauna. That opinion was more completely stated by Girty in his full discussion
of the Carboniferous faunas of Colorado. The more recent work in the San Juan
region, and especially in the Animas Valley, has shown that the Rico formation
is not a persistent feature of the Red Bed section, nor its fauna so markedly dis-
tinguishable from that of the Hermosa beds, as was seemingly the case from
obser\-ations in the Rico mountains. At even a few miles distance to the east
or southeast from that district, the transition from the unfossiliferous Red Beds
to the Hermosa is no longer through a meu-ked reddish zone 300 feet in thickness.
The Rico fossils are found in certain peculiar limestones, plmnly to be correlated
with those so marked in the Rico formation, but these fossil-bearing strata are
not necessarily intercalated in a red section and are limited to a very narrow
band. Moreover, there is mingling of forms supposed originally to be char-
acteristic of the Rico with those of the Hermosa. These observations make the
Rico formation a local development, having less importance in the analysis of
either the Red Beds or the Carboniferous section than was at first assigned to it."

Cross and Howe, in the pap>er just cited, make several observations w^orthy
of note on the correlation and interpretation of the red beds of southwestern
Colorado.

(Page 466.) "If the Aubrey and Hermosa are practically equiv^alent, as the
stratigraphic relations suggest, the Cutler beds occupy a position corresponding
to that of the Permian of the Kanab \^alley in Utah and the formation of the
Zuni Plateau referred to the Permian by Dutton * * *."

' Cross, Whitman, and Ernest Howe, Red Beds of Southwestern Colorado and Their Cor-
relation, Bull. Geol. Soc. Amer., vol. 16, p. 452, 1905.

158 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

(Page 472.) "The general conditions under which correlation of the San
Juan formations with those of the Plateau section must be made are as follows.
Adjacent to the mountains there is a broad zone of gentle westward slope in
which Cretaceous beds occur. The main streams flowing west and south cut
valleys into and in some places through the Cretaceous into underlying formations.
Nearer the canyon of the Colorado the valleys widen and broad platforms and
terraces of Jurassic and Triassic beds appear, the Cretaceous being restricted to
the divides and isolated mesas. The Paleozoic formations appear at first only
in isolated exposures in the deeper canyons, but far to the southwest rise to form
the broad plain called the Colorado Plateau, on the south side of the Grand
Canyon. Thus the older the formation the greater are the gaps between districts
of good exposures, and the greater the likelihood that in the covered tracts un-
suspected complications have entered into the problem."

(Page 473.) [In 1899, A. C. Spencer made a trip into the Paradox and Sinbad
Valleys, where he found below the Dolores] "coarser Red Beds, often conglomer-
atic, with pebbles 3 inches or more in diameter, and several hundred feet of such
strata were noted. No opportunity was found to measure a section showing the
full thickness of these coarser Red Beds, but, as observed by Peale, they are under-
lain by fossiliferous Petinsylvanian Carboniferous in Sinbad Valley * * *."

Cross and Howe suggest (page 475) that the lower 514 feet, red sand-
stones, of Newberry's generalized section of the valley of the Colorado
(Report of Expedition from Santa Fe, New Mexico, to the Junction of the
Grand and Green Rivers of the Great Colorado of the West in 1859) is
equivalent to the Cutler.

(Page 477.) [Button referred the lower 450 feet of Newberry's saliferous
series to the Permian. This consists of] "'sandy shales, containing gypsum and
selenite in abundance, with here and there thin bands of limestone.' At some
unspecified horizon in this formation Dutton found ' several specimens of Bake-
wellia and an attenuated form of Myalina.' On this ground he correlates these
beds with the Permian of the Kanab Canyon district, where Walcott had dis-
covered a more extensive fauna. 'The Permian beds are distinguished for their
dense and highly variegated colors — chocolate, maroon, dark brownish reds
alternating with pale, ashy gray, or lavender colors.'

"The Permian strata thus described are overlain by 'a very coarse, almost
conglomeratic sandstone,' some 50 feet in thickness, which Dutton correlates
unhesitatingly with the 'Shinarump conglomerate' (a particular conglomerate
within the Shinarump group), referring to the fact that it is persistent and uniform
in aspect wherever it appears through the plateau country of Utah and Arizona."

Cross and Howe suggest (page 478) the equivalency of Dutton's Permian
and the Cutler, but note that there is no community of species in the under-
lying Pennsylvanian Aubrey and Hermosa, thus indicating that though the
stratigraphic position is the same, there is no certainty the formations are
equivalent.

The deposits of the upper Paleozoic of west-central Colorado are dis-
cussed in the chapter upon the Pennsylvanian beds of the basin region.

THE BASIN PROVINCE 159

E. PERMO-CARBOXIFEROUS OF THE NORTHERN PART OF THE

BASIN PROVINCE.

A discussion of the Permo-Carboniferous beds of the Basin Province
was given by the author in Publication 207 of the Carnegie Institution.
To this are added recorded observations completing the description and
showing the relation of the beds of the Basin Province to those of the Plains
Province.

As has been shown on page 154, the Cutler formation of southwestern
Colorado and adjacent parts of Utah show the last traces of the red beds
of the Permo-Carboniferous. North and west of this region the equivalent
horizon is either absent or is occupied by diflFerent deposits.

The Weber sandstone horizon, traced on pages 120-136, becomes of im-
portance about where the Red Beds of Permo-Carboniferous age disappear.
It is ver\' possible that it is represented by the Lower Aubrey sandstone of
the Grand Canyon section, but it becomes more prominent and definite in
Central and Northern Utah and breaks up in southern Montana. Using
this well-determined horizon as a guide marking the upper Pennsylvanian,
the layers above it are mostly regarded as Permian, or Permo-Carboniferous.

The condition of the upper beds in the Wasatch and Uintah Mountains
is as follows:

King's description of the upper part of the Weber Canyon section is
quoted in Professional Paper No. 71 of the United States Geological Survey,
page 377:

"Conformably overlj^ng the [Weber] quartzite is a very heavy bed of much
altered gray limestone from 600 to 700 feet thick. The bedding planes are often
entirely obliterated and the material extremely crystalline, showing traces of
great interior disturbance. The lower beds show a true conformity with the
underlying quartzite. * * * The average colors of these limestones are cream-
graj's, inclining often to white in the more crystalline portions. * * * 0\'erlying
this main body of 700 feet of limestone is a series of yellow shaly limestones 175
feet thick. * * * Overlying these calcareous shales, as heretofore quite conform-
able, is a series of sand and mud rocks, all more or less calcareous, varying in color
from chocolate to olive, with red argillaceous sandstones, the whole about 225
feet thick. It has the appearance of a comparatively shcillow water deposit,
made of argillaceous material, limestone, and sand, the thickness of the individual
beds being unusually limited. There are very many beds not over an inch thick.
On the upper surface of the strata, at several horizons, ripplemarks are preserved
with unusual distinctness and on a scale of fineness not often seen, the distance
between the wa%-e and the trough being frequently not over an inch or an inch
and a half. Alternating dark chocolate and olive-colored shales form the lower
200 feet of this group, while the upper 25 or 30 feet are pretty solid sandstone.
Over these, still conformable, are 100 feet of yellow and olive calcareous shales,
which are so earthy as usually to decompose, yielding a bad outcrop. Above this
is a bed of bluish-gray limestone, rather compact, about 150 feet in thickness.
Next comes 20 feet of reddish-browTi clayey sauid, hardly compacted into rock,

160 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

containing thin stony seams intercalated at intervals in the soft, easily eroded
matter. This is immediately followed by 75 feet of a yellowish-gray, brittle,
easily decomposed limestone. Next above are 100 feet of light-colored, very
thinly bedded limestones, that give way to 100 feet more of dark, siliceous, tough
limestone, which breaks under the hammer with great difficulty, yielding an
exceedingly rough, ragged fracture."

Blackwelder^ regards the alternating series of dark limestone and shale
with local sandstone beds, which rest upon the Weber quartzite as corre-
sponding to Boutwell's Park City formation. "There is considerable al-
though not conclusive evidence of an important unconformity between
the Weber and the Park City formations." ^

In the Uintah Mountains, as described by Weeks,' there are 600 feet of
Permo-Carboniferous red and purple shales and blue limestone on the east
side of the Duchesne River below the mouth of the West Fork, followed by
"1,000 feet of light gray and white sandstones, with some interbedded
limestones in the lower part. In the upper part these sandstones occur in
alternating layers of soft and compact beds full of peculiar black points and
specks. These are succeeded by 800 to 900 feet of red shales, with a promi-
nent band of light-colored shale at the top."

In northeastern Utah, southeastern Idaho, and southwestern Wyoming
the various limestones and shales of the Wasatch and Uintah Mountains
give place to the phosphate-bearing Park City formation of Boutwell,
originally described and named in 1907.* It was more fully discussed in
19 1 2.' The later description is quoted in part below:

"This formation is made up largely of calcareous members, but it also embraces
several sandstones and quartzites. * * * In general the formation comprises a
thick limestone in its lower part, several minor limestones in its upper part, and
a number of thin calcareous beds near the base, with intercalated quartzites and
sandstones."

The type section of the formation is exposed in the Big Cottonwood
Canyon :

Grayish-white limestone, with fine gray and white cherts increasing toward bottom 19

Shale and fine buff sandstone 19

Dark-gray limestone; thin chert, red shale, and porous loose members at base 7

Sandy shale 11

Yellowish-gray quartzitic sandstone changing into cherty white lime below / 21

Gray and white banded chert with few white sandstone intercalations 52

Fine calcareous sandstone, with lentils of chert and brecciated fragments of sandstone 8

* Blackwelder, Eliot, New Light on the Geology of the Wasatch Mountains, Utah, Bull.

Geol. Soc. Amer., vol. 21, p. 517, 1910.
' Willis, Bailey, Index to the Stratigraphy of North America, U. S. Geological Survey,

Professional Paper No. 71, p. 379, 1912 (citing Blackwelder).
' Weeks, F. B., Stratigraphy and Structure of the Uintah Range, Bull. Geol. Soc. Amer.,

vol. 18, p. 439, 1907.

* Boutwell, J. M., Stratigraphy and Structure of the Park City Mining District, Utah,

Jour. Geol., vol. 15, p. 439, 1907.

* Boutwell, J. M., Geology and Ore Deposits of the Park City District, Utah, U. S. Geo-

logical Survey, Professional Paper No. 77, p. 49, 1912.

THE BASIN PROVINCE 161

Float of buff sandstone and shale, becoming more shaly and calcareous at base 104

Siliceous arkose comprising mainly rounded quartz grains and feldspars cemented with ferruginous

material 18

Compact grajish quartzite 20

White compact sugary sandstone fossiliferous at base 8

Fine gray and pink massive quartzite with brown sandstone and gray-white chert bands near base ... 3/0
Light-gray limestone -weathering whitish gray with an imbricated pattern; fine gray lime near base

carries good faunas at two horizons in particular, 20 and 55 feet above the base 27

Gray calcareous sandstone 24

Fine gray limestone .' 9

Float showing bits of gra>'ish and brown calcareous sandstone 36

Sandy limestone more calcareous at base -with ca%'emous weathered surface 23

Float; upper sandy beds at top of Weber quartzite 31

"Deposition. — The conditions which prevailed during the deposition of this
formation, a few hundred feet in thickness, marked the transition from those
under which the great thickness of sandstone had been laid down to those which
followed, when the sediments formed red shale. The composition of the lime-
stones, sandstones, and shale points to their deposition in comparatively shallow
water and the limestones contain shallow-water remains. The repeated alterna-
tion of these lithologic t\'pes shows unsettled conditions either as regards elevation
or dep>osition along shore, and it is probable that both occurred.

"Age and strati graphic relations. — The fauna of the Park City formation,
which is better known from areas in Idaho, Wyoming, and other parts of Utah
than from the Park City district itself, may be properly limited to two facies —
one which is best known in the dark phosphatic and calcareous shales around
Montp>elier, Idaho, and one which occurs in the limestones that at some points
overlie these shales and at others seem largely to replace them. The fauna of the
phosphate shales and limestones has been described in United States Geolc^iceil
Surxey Bulletin 436. The fauna of the limestones is characterized by the remark-
able species Spiriferina pulchra, with which are associated tj-pes of Productus,
Bryozoa, etc. Both faunas are suggested by the collections from the Park City
district, the one by more or less abundemt Lingulidiscina tUahensis, the other by
spiriferinas, probably referable to 5. pulchra. The age of these faunas is now
provisionally determined as Permian.

"No unconformity was observed with the underlying Weber quartzite or the
overlying shale or wthin the formation. Accordingly it would seem that sedi-
mentation proceeded unbroken from Mississippian time through that part of the
Pennsylvanicm which is represented by the Park City formation."

"The conditions which prevailed during the closing part of Permian time, as
shown by the alternating limestone, sandstone, and shale of the Park City
formation, were at different stages those of sea bottom, shallow shore bottom, and
exposed shore with mud flats." ^

The Permo-Carboniferous age of the Woodside member of the Park City
formation was demonstrated by Girty,* who writes:

"The three members of the Park City formation vary from point to point
in lithology and in thickness, as well as in fauna. In the Montp>elier district the
upper limestone marks an important horizon for determining the position of the
phosphate deposits. It is massive, contains here and there much black chert,

• Boutwell, J. M., U. S. Geological Surrey, Professional Paper No. 77. p. 104.

' Girty, G. H., The Fauna of the Phosphate Beds of the Park City Formation in Idaho,

Wyoming, and Utah, U. S. Geological Survey Bull. 436, p. 6, 1910.
IS

162 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

and is at many places full of fossils, especially of several species of Productus,
from which fact it is often called the 'Productus limestone.' Not far to the north,
in the Swan Lake district, the upper limestone is replaced by siliceous or cherty
shale of dark-purplish color. This cherty shale is not as a rule a prominent
feature of the stratigraphy nor does it at many places contain fossils.

"The lower limestone in the Montpelier region is of a whitish or buff color
and at some places appears to be a fine-grained calcareous sandstone rather than
a limestone. As a rule its fossils are very few and so ill preserved as to be indeter-
minable. In the Swan Lake district, on the other hand, this bed is a massive
whitish, more or less siliceous limestone containing in abundance poorly preserved
silicified fossils, among them species of Spirifer, Squamularia, and probably
Composita. In its upper portion a large semireticulate Productus is found, and
fine, black, earthy limestones that locally appear at its very top contain numerous
specimens of Spirifer, Productus, and Composita. In this region the lower lime-
stone serves much better than the upper as a guide for finding the phosphate
beds, and for several reasons, more or less obvious, it seems to have been generally
inferred that the guide rock was the same in both areas and that the series was
overturned in the Swan Lake region. There is, however, hardly room for reason-
able doubt that the stratigraphic sequence is normal in both regions and that the
beds themselves differ in character in the two areas.

"The phosphate beds consist mosdy of soft rock, shales, phosphates, and
impure limestones, the latter seldom more than a few inches thick. The shales
are more or less phosphatic and the phosphate bands are more or less argillaceous.
Their prevailing color is black, weathering to brown, but in the Beckwith Hills
the color of the phosphate and associated rock is buff or even reddish. The
thickness of the phosphate-bearing shales ranges from 60 to 100 feet. The main
deposits of phosphate occur, as a rule, at the base of the series, so that in the Mont-
pelier district they lie about that distance below the 'upper Productus limestone,'
but in the Swan Lake district they occur immediately above the 'lower Productus
limestone.'

"The remaining formations concern the present report less closely. The beds
below the Park City formation in southern Idaho have been identified with the
Weber quartzite, which holds a similar position in the Wasatch Range. The
equivalence of the strata in the two sections, especially in detail, is not entirely
clear. In Idaho the 'lower Productus limestone' abruptly grades below into
white sandstones and quartzites, and the Mississippian limestones are succeeded
above by light-colored limestones that include more or less interbedded quartzitic
sandstones, these being probably of Pennsylvanian age. Between these two
quartzite- bearing groups comes in a series of soft beds approximately 1,000 feet
thick. They are poorly exposed, but seem to comprise soft sandstones and soft
earthy limestones of reddish or yellowish tints. If the upper quartzitic beds are
called the Weber, the division between the Weber and the Park City is not easy
to determine. It may prove desirable to draw the line at the base of the phos-
phate shales and to include the siliceous limestones and calcareous sandstones of
the 'lower Productus limestone' in the Weber. If so, the thin stratum of black
limestone which at some places occurs at the base of the phosphate beds and has
here been spoken of as part of the 'lower Productus limestone' may perhaps
better be united with the Park City formation, because the fossils obtained in it
indicate a certain change from the fauna of the white limestone below and an
affinity with the fauna of the phosphate series above. In any event, the Weber

THE BASIN PROVINCE 163

quartzite, whose fauna is almost unknown, seems to show considerable modifica-
tion in the Idaho sections.

"In notable contrast to the Weber formation, the beds above the Park City
formation show striking persistence in their main lithologic and paleontologic
characters. These are the 'Permo-Carboniferous' beds of the King Survey and
were di\-ided by Boutwell in the Park City district into the Woodside, Thaynes,
and Ankareh formations. 1 1 seems all but certain that the ' Permian ' of Walcott's
section in Kanab Canyon, in southern Utah; the 'Permo-Carboniferous' of the
Wasatch Mountains, in northern Utah; and, in part, the 'lower Triassic' of
southeastern Idaho are one and the same series. The Woodside, Thaynes, and
Ankareh do not, perhaps, maintain precise boundaries throughout all this terri-
tory, and in Idaho the first occurrence of Triassic ammonites (Meekoceras beds)
is conventionally taken as the base of the Thaynes. * * *"

"The Triassic age of at least the major portion of the ' Permo-Carboniferous'
(Thaynes and Ankareh) seems to be shown by fairly satisfactory evidence — the
presence of an extensive ammonite fauna of Triassic type and the practical
absence of any distinctive Carboniferous forms. In advance of a detailed study
of these faunas, however, it may be pointed out that above the Meekoceras beds
there are zones which contain great numbers of KhynchoneUa closely related to the
Carboniferous Pugnax Utah and many specimens of apparently true Myalina,
not unHke Carboniferous species.

"It is much less certain that the Woodside formation is not Paleozoic (Per-
mian?). A preliminary study of the fauna of the Woodside shows that, except
that it has yielded no ammonitic forms, it does not differ materially from the
fauna of the Thaynes and presents a strong contrast to the Carboniferous fauna
of the Park City. Lithologically also there is a well-marked division between
the Woodside and the Park City formation, and no lithologic boundary can be
traced between the Woodside and the Thaynes. That the Woodside, Thaynes,
and Ankareh form a natural group is indicated by the classification of these rocks
adopted by most geologists. If the Thaynes is Mesozoic, the obvious line between
the Mesozoic and the Paleozoic would seem to be the line between the Park City
and the Woodside. If, then, as may be tentatively concluded, the Woodside
does not represent the Permian, the natural question to follow is. Does not the
Park City formation belong in the Permian? A decisive judgment on this point
should wait upon a careful study of the faunas obtained from other members of
the Park City beds, as well as upon a study of other related faunas less certainly
appearing at the same horizon. Because of the close relationship or identity of
many species of these faunas with the Gschelian fauna of Russia, I am pro-
visionally holding that the Park City formation is older than the Permian.

"Anyone at all familiar with the Carboniferous faunas of the Mississippi
Valley will at once recognize the fact that the forms found in the phosphate beds,
individually as well as collectively, are quite different from any others found in
that area. In fact, but few of the phosphate species have closely related forms
in the Pennsylvanian, and a correlation by paleontology with any definite portion
of the Pennsylvanian section is at present impossible. Even among western
faunas this has an extremely individual and novel fades, one which is known to
me as occurring only in a well-defined area. * * *

"Though the phosphate fauna possesses a remarkable individuality of fades,
it is not altogether out of relationship with other formational faunas, for with
the aid of the fossils from the associated rocks it can be recognized as belonging

164 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

to a fauna widely dispersed over the West and traceable, it is believed, even into
Alaska, Asia, and eastern Europe. The other western faunas of about the same
geologic age are largely composed of representatives of the Brachiopoda and
Byrozoa, especially of the Productidae and Spiriferidae. Thus there is little
common ground for comparison, but though the abundant and characteristic
features of the phosphate fauna are peculiar to it, where a common ground for
comparison does exist an agreement can be found to a considerable extent,
and some of the less abundant forms have a wider distribution. These western
faunas have not been studied in detail and in many localities the rocks in which
they occur have not been discriminated into formations and named, so that it is
possible to speak of them only in a general way. If, however, we eliminate the
beds of recognized lower Carboniferous age, such as the 'Wasatch limestone' of
Utah (lower part), the Redwall limestone of the Grand Canyon section, and the
Baird shale of California, and also certain younger divisions such as the ' Permo-
Carboniferous ' of the Wasatch Mountains and surrounding region, the Permian of
the Grand Canyon section, the Guadalupian series of Texas, and, perhaps, a few
others, we have left a group of rocks, as already remarked, which is widely dis-
tributed throughout the West (including Alaska) and which constitutes the major
portion of the Carboniferous representation in that extensive area. The fauna
of these formations thus tentatively grouped together, though presenting many
local modifications, can, in a general way, be correlated with the Gschelian fauna
of the Ural Mountains, some of the American assemblages presenting truly
remarkable resemblances to those of Russia."

The Park City formation, which Schultz^ regards as equal to the Phos-
phoria formation and the upper part of the Wells formation in eastern
Idaho, is exposed on both the northern and southern sides of the eastern
and western ends of the Uinta Mountains. A geologic map showing the
exact location of the outcrops is given in his publication. His description
of the Park City formation is as follows ■}

"Park City Formation (Pknnsylvanian and Permian).

"The Weber quartzite is overlain by a series of limestones of Pennsylvanian
and Permian age, with which are associated some calcareous sandstones, shales,
and chert. This formation contains the phosphate deposits and has yielded the
bonanzas that during the last decade have made the Park City mining district,
in the Wasatch Range, Utah, a famous lead and silver camp. The formation
consists for the most part of thin-bedded arenaceous limestones and shales, with
some massive limestone ledges near the base. Although consisting chiefly of
limestones, it may be subdivided into four more or less distinct members which
are usually recognizable in different parts of the field: (l) the upper thin-bedded
shaly gray limestone series, which weathers readily, forming depressions, and at
a distance has the appearance of a clay or shale bank; (2) the upper or cherty
limestone beds, which contain a large percentage of concretionary chert nodules
and lenses; (3) a phosphatic shales series, in which bands of chert and limestone
occur; and (4) the lower limestone member, consisting primarily of massive light

1 Schultz, A. R., A Geologic Reconnaissance of the Uinta Mountains, Northern Utah, U. S.

Geological Survey Bull. 690-C, p. 76, 1918.
' Loc. cit., p. 46.

THE BASIN PROVINCE 165

and gray limestones, some beds of which are from 5 to 25 feet thick. The upper
member of shaly limestones has thus far not furnished any identifiable fossils,
but nevertheless has lithological characteristics so distinctive that it is easily
recognized wherever it is exposed. On Brush and Little Brush Creeks and in the
vicinity of Green River on both sides of the range this member is considerably
thicker than at the west end of the field and consists of grayish-drab limestone
and shaly limestone with a few streaks of pink or red clay near the top, all
weathering to a dull slate gray, so that the entire series, seen from a distance,
looks like a gray clay bank. In some respects these beds resemble the Dinwoody
formation along the east side of the Wind River Mountains, in western Wyoming.
Future detailed work may result in differentiating them into a formation, but
for the present they are retained as a part of the Park City formation.

"The upper cherty limestone member is variable in detail, but is prominent
and easily recognized throughout the range. It consists of massive gray to
cream-colored limestone 20 to 25 feet thick (underleiin by gray and greenish
dark chert in a matrix of shale). It is a controlling factor in the topography and
produces long faceted dipslopes along the south side and a part of the north side
of the mountain front. Certain parts of this member are highly fossiliferous and
contain abundant specimens of Leioclema, Derbya, Spirifertna pulchra, and
Lingulidiscina utahensis, other parts lesemble somewhat the Rex chert member
of the Phosphoria formation of eastern Idaho.

"The phosphatic shale is probably the most distinctive member of the Park
City formation. It is made up largely of black and green, decidedly fissile shale
40 to 50 feet thick and beds of limestone, sandstone, chert, and phosphate ranging
in thickness from a few inches to several feet. Some of the beds of limestone and
phosphate contain an abundance of fossils, a few of which have been collected for
identification. In addition to the Brj'ozoa there are numerous comminuted
fossil fragments, glauconite, and scattered foraminiferal shells. This member is
probably equivalent to the lower part of the Phosphoria formation in eastern
Idaho, as described by Richards and Mansfield.'

"The lowest member consists chiefly of massive limestone with some beds of
shale and sandstone. It is more variable than the overlying members and in
some localities appeals to be entirely missing, as the phosphate shale or phosphate
bed rests directly upon the Weber quartzite. It is probably owing to variations
in this member and the upper member of the formation that the Park City
shows so great differences in thickness throughout the field. The total thickness
ranges from 250 feet or less to 850 feet or more in parts of the field where the beds
have been measured. In certain portions of eastern Idaho where the phosphate
deposits have been studied in great detail the lowest member of the Park City
formation as here described was considered a separate member of the Park City
formation and later was included with the Weber quartzite and underlying beds
as a part of the Wells formation. It appears, however, from the facts thus far
gathered in the Uinta range, that it must be considered as a part of the Park City
formation as defined by Boutwell in his Park Cit\' reports.

"The contact between the Weber quartzite and the overlying Park City
formation has been a subject of considerable study without definite results.
The beds in many localities appear to be conformable, but the relations from
place to place and the position of the phosphate series with regard to the Weber

* Richards, R. VV'., and G. R. Mansfield, Geology of the Phosphate Deposits Northeast of
Georgetown, Idaho, U. S. Geological Survey Bull. 577, 1914.

166 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

quartzite make it appear that the Park City formation unconformably overlies
the Weber quartzite. However, until the beds are studied in greater detail and
careful areal mapping is completed it will not be possible to state definitely the
extent or magnitude of the unconformity. It seems improbable, however, that
the relation between the Weber quartzite and the phosphate beds of the Park City
formation can be satisfactorily accounted for in any other manner. Boutwell,
who studied the formation in the Park City region of Utah, comes to the con-
clusion that the strata lie conformably upon the underlying Weber quartzite
and that sedimentation proceeded unbroken from Mississippian time to the end
of Park City time. He calls attention, however, to the fact that one geologist
reported a marked unconformity at this horizon and that his own studies were
not as conclusive or definite as might be wished. Blackwelder,' in his studies of
the Wasatch Range, recognized the unconformable relation between the Weber
quartzite and the overlying Park City beds and concluded that in the Wasatch
Range there is an unconformity between the Weber quartzite and the overlying
Park City phosphatic beds of Permian age. In another report Blackwelder'
states that the base of the Park City formation is generally marked by beds of
white and pink soft sandstone separated from the Weber quartzite by an obscure
unconformity, which appears nevertheless to be one of considerable magnitude.
Similarly in the Wind River and Gros Ventre Mountains the Park City is known
to rest upon the Tensleep sandstone (horizon of Weber quartzite), in apparent
conformity at many localities, but close examination reveals evidence of an
uneven erosion surface at the base of the Park City beds. A similar unconformity
has also been reported by Richards and Mansfield in eastern Idaho.' The
conditions that prevailed during the period of deposition of the Park City forma-
tion marks the transition from those under which the great thickness of Weber
sandstones was laid down to those under which the sediments formed red shale.
The strata indicate deposition in comparatively shallow water and show an
unstable or fluctuating shore-line, as it appears that both elevation and depression
of the shore occurred.

"The irregularities and differences in thickness of the Park City formation
may best be illustrated by giving some of the measured sections in different parts
of the field. Boutwell's type section of the Park City formation is Big Cotton-
wood Canyon, Utah, given below (see page i6o of this work), does not state the
thickness of the phosphate bed or the phosphatic shale, nor does he show the
horizon at which it occurs. However, the work done by Gale in the Park City
district in 1909 proves that the phosphate occurs near the middle of the formation
and is included with the 104 feet of beds overlying the 18 feet of siliceous arkose.
In discussing the phosphate deposits Boutwell states that the phosphate bed noted
at several points is about 3 feet thick and contains 32.6 per cent P2O6, or 71 per
cent bone phosphate."

Schultz gives several sections other than the type section of Boutwell
cited above. In commenting on the phosphate rock, he describes its oolitic
character and notes the irregularity of the layers:

* Blackwelder, Eliot, New Light on the Geology of the Wasatch Mountains, Utah, Bull.

Geol. Soc. Amen, vol. 21, pp. 531-533. 542, 1910.
' Blackwelder, Eliot, Phosphate Deposits East of Ogden, Utah. U. S. Geological Survey,
Bull. 430, pp. 536-551, 1910.

* Richards, R. W., and G. R. Mansfield, loc. cii., pp. 16, 22.

THE BASIN PROVINCE 167

"When compared over large areas these layers are found to be variable as
to both character of the bed and quantity of phosphoric acid present, not only
vertically but horizontally, and yet in many respects they are rather uniform and
constant in character and have throughout the field certain common characters."

The Park City formation extends nearly as far north as the Weber and
Quadrant quartzites.

Veatch^ described briefly the occurrence in southwestern Wyoming.
Here the formation occurs in the Tunp Range, about 15 miles west of
Kemmerer. It is " a series of very arenaceous thin-bedded limestones which,
from its stratigraphic position, approximately represents the Upper Coal
Measures limestone of the Fortieth Parallel Survey." It is very probable
that the same formation occurs in the Crawford Mountains, which extend
to the southwest into the type area of the Park City formation or very
near to it.

In eastern W'yoming the Park City horizon is represented by the Embar
limestone ; the equivalence of these beds seems to be more and more definitely
established as work in the region progresses.

Woodruff,^ in describing the Lander region, says that the Embar forma-
tion occurs in the Wind River Mountains adjacent to the Lander oil field.
"About half of the Embar formation consists of limestone, most of it mas-
sive and crystalline, but certain members of this half are shaly and cherty;
the other half of the formation is shale."

The upper member, as given in the section below, is evidently above
the Park Citv^ formation and is Triassic. The lower part may be called
Permo-Carboniferous.

Feet.
Limestone, shaly and tan-colored; sandy at top; lies immediately below the red shales at tte base

of the Chugwater formation. (Triassic) 18

Limestone, massive, crj-stalline, slightly cherty, and with distinct majw joints 23

Limestone, concretionary, cherty, slightly shaly 45

Shale, drab, sandy 40

Sandstone, bituminous (Permo-Carboniferous fossils) 5

Limestone, shaly, crystalline 5

Shale (?) 24

Shale 21

Limestone, massive, impure, with well-developed major joints 73

Limestone; contains occasional thin, shaly limestone layers 90

Darton' in his description of the Big Horn Mountains speaks as follows
of the Tensleep and Embar:

(Page 34.) "Tensleep Sandstone. — Its thickness varies from 50 to 100 feet
in the northeastern portion of the region to over 200 feet in the east-central
portion and from 250 to 300 feet to the south and west. * * *

1 Veatch, A. C, Geography and Geology of a Portion of Southwestern Wyoming, U. S.

Geological Survey, Professional Pai)er No. 56, p. 50, 1907.
* WoodruflF, E. G., and C. H. Wegemann, Lander and Salt Creek Oil Fields, Wyoming,

U. S. Geological Survey Bull. 452, p. 12, 191 1.
' Darton, N. H., Geology of the Big Horn Mountains, U. S. Geological Survey, Professional

Paper No. 51, 1906.

168 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

"The predominant rock is white to buff sandstone in thick massive beds,
cross-bedded and often weathering into irregular pinnacled forms. * * * On the
west side of the range, in the vicinity of Paintrock and Tensleep creeks, the forma-
tion varies in thickness from 75 to 150 feet, and in its thicker portions its lower
part includes some softer, fine-grained sandstones and the upper member is
strongly cross-bedded. * * * The upper portion of the Tensleep sandstone often
is calcareous and at many points there are one or two beds of a mixture of sand
and lime sediments."

The Tensleep sandstone is said in this paper to represent the upper
portion of the Minnelusa of the Black Hills.

(Page 35.) " The Embar Formation. — In the southern part of the Bighorn
uplift between the red beds and Tensleep sandstone is a limestone with some
associated shaly and cherty beds, which, with gradual increase in thickness, is
continued westward in the Bridger Range and Owl Creek Mountains. Appar-
ently it is neither a development of the basal portion of the Red Beds nor of the
calcareous sandstone which sometimes occurs at the top of the Tensleep sand-
stone in the region north. * * * On the east side of the Bighorn Mountains the
formation first appears near the West Fork of Powder River, and, on the west
side, in the slopes south of Redbank. It finally attains a thickness of about 200
feet on the ridge south of Thermopolis, where it constitutes an extensive dip
slope several miles wide, extending along the north slope of the Bridger Range.
Prominent exposures appear in the upper canyon of the Bighorn River, which
cuts deeply into this slope. Here the formation consists of 50 feet of massive
limestone, underlain by calcareous shale filled with nodules and lenses of chert,
a member which merges down into sandy shales and impure limestones. At the
base there is a thin mass of sandstone breccia lying on the massive Tensleep
sandstone. Owing to extensive faulting, the formation appears only at one point
on the south side of the Bridger Range, but it outcrops prominently at the south-
western termination of the Bighorn uplift, 3 miles east of Deranch. Here the
limestone constitutes a line of low hogback ridges, at the base of which appear
sandy and cherty shales of buff color lying on the Tensleep sandstone. A thick
mass of the limestone appears in the 7,000-foot knob 6 miles northeast of Deranch,
which at first sight might be mistaken for Madison limestone. Two miles south
of No Wood, at the northern end of a short, deep gorge of No Wood Creek, a
partial section is exposed in which the Embar limestone is seen to be 10 feet
thick, somewhat cherty, and lying on 10 feet of cherty shales, which extend to
the top of the Tensleep sandstone. The limestone is overlain by a yellowish
sandy bed, which may constitute the base of the Red Beds. At the head of
West Kirby Creek the limestone is underlain by gray and reddish sands, in all
about 30 feet thick. A short distance north of No Wood post-ofhce the following
section is presented:

Section on No Wood Creek i mile northeast of No Wood, Wyoming.

Feet.
Red beds, with lo-foot bed of limestone 100 feet from base (Chugwater).
Limestone, light yellow, weathers in thin beds, has a layer of flint near the center, and occasional

flinty concretions lO

Shale, light yellow 25

Limestone, massive, of light-gray color, with chert concretions and layers of black chert between

bedding planes 20

Buff shale o to 2

Soft white sandstone (Tensleep).

THE BASIN PROVINCE 169

"A mile north of this locality there is a similar section, but the basal limestone
is thinner, not over lo feet thick, is yellowish in color in its lower portion, and lies
directly on the Tensleep sandstone. At the western entrance of the deep canyon,

4 miles north of No Wood, the top limestone underlying the Red Beds is 12 feet
thick, massive, cherty, impure, and of yellow color. Next below are 40 feet of
shales and soft sandstones, partly pale red, but yellowish near the top. They
contain a few layers of limestone and lie on the Tensleep sandstone. On the
east slope of Bighorn Mountains the formation appears at intervals in the valley
of BuflFalo Creek, near the Hole-in-the-wall. The limestone is 20 feet thick,
with a 2-foot massive layer at the top, and with thinner-bedded slabby limestones
of greenish-gray color and green shale below, lying on the Tensleep sandstone.
The formation is traceable continuously northward to the Red Fork of Powder
River, but gradually thins in that direction. On the anticline in Red Fork Valley,

5 miles northeast of Bamum, the Tensleep sandstone is overlain by a thin mass
of limestone breccia, merging up into 6 feet of buff sands and greenish shale,
which probably represent the northeastemmost extension of the Embar forma-
tion."

Blackwelder,^ as already indicated, has suggested the equivalence of
certain beds in eastern and western Wyoming. The Tensleep of the Big-
horns is the same as the Weber of Utah and the southw^est; the Embar is
the equivalent of the Park City; the Morgan formation of Utah is the
equivalent of the Amsden of the Bighorns. In the Gros Ventre Range
the phosphate beds become cherty and unfossiliferous. Toward the W'ind
River Range, Shoshone, and Owl Creek Mountains the phosphate thins and
deteriorates and beds exceeding a few inches in thickness do not occur north
of the Owl Creeks, northeast of the north part of the Bighorns northeast
of the low ranges between Casper and Lander.

Condit,in a recent paper,' gives a description of the Phosphoria formation
in Wyoming and Montana, with a map showing the distribution of the Phos-
phoria and Embar formations and a series of stratigraphic sections. He
says (page 113):

" The Quadrant proper is overlain by 100 to 250 feet of dark-gray cherty
quartzite, layers and ropy masses of nodular chert and shale. The sequence is,
regarded as equivalent to part of the Park City formation of Utah and Wyoming,
but more nearly equivalent to the Phosphoria formation of Idaho, as was recog-
nized in 1913 by Richards and later by Stone and Bonine. * * *"

"The uppermost beds of the Phosphoria contain poorly preserved brachiopod
shells and fish bones, probably of Permian age."

' Blackwelder, Eliot, A Reconnaissance of the Phosphate Deposits of Western Wyoming,
U. S. Geological Survey Bull. No. 470, p. 458, 1910.

'Condit, D.D., Relations of the Late Paleozoic and Early Mesozoic Formations of South-
western Montana and .\djacent Parts of Wyoming, Professional Paper No. 120-F, U. S.
Geological Survey, 1918.

CHAPTER V.

THE LATE PALEOZOIC OF BRITISH COLUMBIA.

It is impossible to identify deposits of Permo-Carboniferous age in either
the Basin or the Plains Province farther north than has been indicated in
the preceding summary descriptions. It is altogether probable that the
uplift which occurred in western British Columbia and in Alaska prevented
the northern extension of both the basins of deposition ; if any deposits were
laid down beyond the limits to which they have been traced they have
either been removed by erosion or are buried beneath later deposits, in such
large measure that no trace of them has yet been discovered. It is certain,
however, that the uplift in the latter part of the Paleozoic terminated marine
deposition in northwest North America and formed a surface which must
be considered as no small factor in the environment of the life of the time.
The exact age of the development of the land conditions is uncertain and the
subsequent profound orogenic movements, resulting in metamorphic and vol-
canic action, were so vigorous as to mask much of the record and make the
interpretation exceedingly difficult.

The following descriptions of the beds in British Columbia and in
Alaska must serve in part as an illustration of what may be accomplished
when more exact knowledge of the age of the rocks and their original condi-
tion and extent has been gained.

On the southern part of Vancouver Island, Duncan map area,^ there
are two series of volcanic and metamorphosed sediments which are provi-
sionally referred to the Carboniferous:

Malahat volcanics: Massive and schistose meta-dacites and meta-andesites, tuffs,

and fine-grained cherty rocks.
Leech River formation : Slates, slaty and quartzose schists, micaceous quartzites,

amphibolites, and chloritic schists. If this series is to be correlated with

the northern deposits at all it is with the lower half of the Ketchikan, etc.,

below the prevailing massive hmestone.

In 1 912 appeared Daly's report on the Geology of the Rocky Mountain
Cordillera at the Forty-ninth Parallel, Memoir 38 of the Canadian Geological
Survey. The detailed account of the geology permits an attempt to cor-
relate the Upper Carboniferous and Permian (?) rocks at this latitude.

* Clapp, C. H., and H. C. Cooke, Geology of a Portion of the Duncan Map-area, Vancouver
Island, British Columbia, Summary Report, Canadian Geological Survey, for 1913, p.
89, 1914.

171

172 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

In the Skagit Mountain range the following series has been made out
(page 546) :

Unconformity.

upper Carboniferous : I S-.r^" ^'^^''T''^. (f ^red gabbroid rock).

trr J ^ Lhuhwack volcanic lormation.

Upper Carboniferous (and older) { Hozomeen Jeries."

The oldest, the Hozomeen series, covers 3 or more square miles north of
Glacier Peak and just east of the main divide of the Skagit Range. The
prevailing rock is a cherty quartzite, occurring in thin, flaggy beds, from
I inch or less to 3 inches in thickness. Occasional bands of probably con-
temporaneous and extrusive greenstone occur, but there is no limestone in
the main area.

The Chilliwack is exposed on the Chilliwack River. A typical section
about a mile west of monument 48 is as follows:

Top (?) unconformably overlain by Cretaceous (?). Feet.

Quartzitic sandstone 50 +

Dark gray argillite 20

Light gray limestone, fossiliferous 50

Gray calcareous quartzite and dark gray calcareous argillite 60 +

Andesitic tuffs and flows and agglomerates 2,000 +

Gray and brown shale and sandstone; thin conglomerate bands; crumbling, thin-bedded, highly

fossiliferous 200

Light gray, generally crystalline limestone, fossiliferous 600 ±

Shale, sandstone, and grit 90

Massive, light gray limestone, fossiliferous no ±

Dark gray and brown shales, fossiliferous 300 ±

Massive hard sandstone lOO ±

Hard sandstone and black and red shales with bands of grit and thin beds of conglomerate;

roughly 1,400 ±

Hard, massive sandstone with gritty layers 800 ±

Dark gray to black, often phyllitic argillite with quartzitic bands 1,000 +

The 2,000 feet of andesitic tuffs and flows and agglomerates near the
top constitute the Chilliwack volcanic series.

Girty, after listing the fossils obtained from above and below the Volcanic
series (pages 510-513), says:

"On the whole, from the little that I understand of the stratigraphic relations
and from the relationship manifested by the most marked of your faunas with
that of the Nosoni formation, I am disposed to correlate all your beds in a general
way with the latter. They may contain measures younger or older than the
Nosoni, but from the absence of the well-marked Baird and McCloud facies it
seems probable that none of the horizons from which your collections came is
as old as the McCloud."

In the Hozomeen Range east of the Skagit Range the Hozomeen series
is typically developed ; because of the absence of fossils its position is uncer-
tain, but it is provisionally correlated by Daly with the Cache Creek series
for reasons given on pages 502-504 of his Memoir No. 38.

The series is composed of quartzite, chert, limestone, and dominant
greenstone. The limestone occurs as rare intercalations of white or pale
gray rock squeezed into "pods" by later movements and deprived of all

THE LATE PALEOZOIC IN BRITISH COLUMBIA 173

trace of the original bedding planes. WTiile Daly regards it as not safe to
assign any definite age to the three dominant series, he says:

"It seems best to believe, as a working hypothesis, that the Hozomeen green-
stone and limestone are younger than the principal quartzite (phyllite) group and
overlie the latter conformably."

On page 504 Daly says:

"It seems probable, therefore, that the Hozomeen series is to be con elated
with the Anarchist series, and both of them, with Dawson's Cache Creek series
as well as with the likewise fossiliferous Chilliwack River series."

Farther to the east through the Okanagan Mountains, Kruger Plateau,
Midway Mountains, and the Anarchist Plateau (see plate 3, Memoir 38,
and maps 10 to 13), the Anarchist series, so far as made out, seems to
represent the same horizon of the Carboniferous. It is largely phyllite,
quartzite, limestone, and greenstone. The quartzite and phyllite are most
abundant, then the greenstone and the limestone, in "pods."

"This oldest group is almost certainly the same as that which crops out at
interv-als betv^een the Columbia River and Midway, and, in the Rossland district,
bears obscure fossils referred to Carboniferous species. Though the lithological
similarity of the Anarchist series to these Rossland rocks and, as we shall see, to
the ver>' thick, fossiliferous, undoubtedly Carboniferous rocks found in the
Skagit Range, may be an accidental and illusory resemblance, it seems best to
correlate the Anarchist series, or much of it at least, with the Carboniferous rocks
of western British Columbia." (Page 422.)

In the center of the Columbia Mountain system beti^^een Christina
Lake and Midway is exposed a series of argillites, quartzites, and limestone
called by Daly the Attwood series.^ This resembles the Cache Creek of
the Kamloops district. The limestone is generally white and cr>'Stalline,
but occasionally drab or black, the argillites are or were carbonaceous. The
sediments "probably form part of a once extensive series of sediments
which covered southern British Columbia." Daly agrees with Brock in
correlating this series •with the fossiliferous (probably Carboniferous) series
of argillites, quartzites, and limestone in the Rossland Mountains.

In the Rossland Mountains, and especially in the Little Sheep Creek
Valley, there are fossiliferous limestones, cherts, and quartzites, and in the
Pend d 'Oreille region phyllites, quartzites, limestones, etc. In the Little
Sheep Creek Valley there is a blue-gray to white limestone, brecciated, and
with lenses of chert and quartzite. In the breccias have been found Carbon-
iferous fossils {Lonsdalia).

In the Selkirk Range the Pend d'Oreille group includes greenstone, phyllite,
amphibolite, etc. The rocks are similar, lithologically, to those found in the
Rossland district, and in central Idaho Lindgren' found closely similar rocks in
isolated areas, the Wood River series, which carry Carboniferous fossils.

' Memoir 38, p. 382, and R. W. Brock, Annual Report Canadian Geol(^cal Survey, p. 96,

1902.
* Lindgren, \V., The Gold and Silver Veins of Silver City, etc., 20th Annual Report U. S.

Geological Sur\ey, pt. in, pp. 85-90, 1900.

174 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

About 100 miles to the north and northwest of the boundary section at
the Pend d'Oreille are found considerable areas of stratified rock which
Brock referred to the Cache Creek series, or the Slocan series, which he
regards as equal to the Cache Creek.^

In the Flathead Valley there is, according to Dowling,^ a "down-tilted
block * * * of Carboniferous limestone with reddish upper beds that may
be Permian or Triassic in the higher members." This lies due south of
Corbin, at the mouth of Squaw Creek.

At Blairmore, in the South Fork River coal area, MacKenzie' describes a
small area of pure and cherty limestone. Similar beds occur in the anticline
of the Turtle Mountains. The age of this bed is uncertain, but is pro-
visionally placed as Devono-Carboniferous.

On the eastern edge of the Rocky Mountain Front, in the Sheep Creek
anticline, on Sheep Creek south of Clagary, Alberta, Dowling* describes
Mesozoic rocks and continues (page 149)*:

"It is quite certain that the floor on which the above-described Mesozoic
deposits were laid down consists of a series of limestones of which the western
portion is formed of Carboniferous rocks with possibly some Triassic and Permian
sediments lying here and there upon them."

"The Cache Creek formation, as shown on the present map and as now
understood, must therefore be regarded as including a very thick series of Paleo-
zoic rocks, of which the greater part is definitely referable to the Carboniferous
period by means of its fossils, but of which it is scarcely probable that the upper
and lower limits agree precisely with those of the typical Carboniferous. It may
very possibly be found at the base, particularly, to transgress these limits and
to include beds older than those of the system.

"In attempting a brief general description of this formation, it must in the
first place be observed that the extremely broken and disturbed character of the
rocks almost everywhere renders it next to impossible to learn much about their
attitude or sequence in any one locality. It is very generally impossible to deter-
mine whether the dip of the beds is normal or has been overturned. It is thus
only by following the general association of the rocks from place to place and by
piecing together facts observed at many different places that it becomes prac-
ticable to outline the salient features of the whole.

"The western part of the Kamloops sheet, between the Thompson and Bona-
parte Rivers on one side and the Eraser on the other, is the typical area for the

* Brock, R. W., Explanatory Note to the West Kootenay Map Sheet, Canadian Geological

Survey, 1902.
' Dowling, D. B., Coal Areas in the Flathead Valley, British Columbia, Summary Report

of the Geological Survey of Canada for 1913, p. 139, 1914.
' MacKenzie, J. D., South Fork Coal Area, Oldman River, Alberta, Summary Report

Canadian Geological Survey for 1912, p. 235, 1913.

* Dowling, D. B., Geological Notes on the Sheep River Gas and Oil Field, Alberta, Summary

Report Canadian Geological Survey for 1913, p. 145, 1914.
'This account is taken from Professional Paper No. 71, United States Geological Survey,
page 390. It is a condensation of reports by G. M. Dawson, Report on the Kamloops
Sheet, British Columbia, Annual Report of the Canadian Geological Survey, new series,
vol. 7, page 39, 1896, and Duplication of Geologic Formation Names, Science, n. s.,
vol. 9, page 592, 1899.

THE LATE PALEOZOIC IN BRITISH COLUMBIA 175

Cache Creek formation, and the most definite feature which can be traced through-
out is the belt of massive * * * £md whitish limestones, sometimes marbles. • * *

"Practically the entire mass of the Marble Mountain Range is composed of
these limestones, as well as the whole eastern part of the Pavilion Mountains.
They include comparatively insignificant intercalations of argillite, cherty quartz-
ite, and materials of volcanic origin. Farther south, in the region to the east of
Hat Creek, such materials become more abundant and form thick beds among the
limestones, particularly the cherty quartzites and the greenstones. In this region
it is probable that the lower part of the great limestone series is most prominentiy
displayed and that the higher beds are more characteristic in the north, particu-
larly in the Marble and Pavilion Mountains. The earlier stages of the great
period of limestone deposition appear to have been marked by frequent interrup-
tions, during which argillaceous and volcanic products were laid down; while
in its later stages the deposition of the limestone must have been almost unbroken.
The interlocking of the diflFerent classes of materials is such, however, as to show
the close connection which obtains between the Marble Canyon limestones and
the lower parts of the Cache Creek formation. * * *

"The extremely unsatisfcictory condition of the rocks of the Cache Creek
series for all purp>oses of measurement [is such that] in endeavoring to give some
idea of the total volume of the formation, no even approximately correct data
can be quoted. The subjoined summarized section is therefore merely an attempt
to indicate the general order of succession, and to some extent the importance
of the formation, in the western part of the area of the [Kamloops] map. The
order is descending.

Feet

1 . Massi\'e liinestones (Marble Canyon limestone), with some minor intercalations of vtdcanic rocks,

argillites, and cherty quartzites. At least i,ooo feet seen in some single exposures. Total
thickness probably at least 3,ooo

2. Volcanic materials and limestones, with some argillites, cherty quartzites, etc Minimum thick-

ness about 2,000

3. Cherty quartzites, argillites, volcanic materials, and serpentines, with some limestcme. The

thickness of these beds, or of a part of them, was roughly estimated in two places as between
4,000 and 5,000 feet. Minimum total thickness, say 4,500

9.500

"Thus, the entire volume of the rocks of the Cache Creek formation, as this
is now defined, may be assumed to be about 10,000 feet as a minimum, while I
am inclined to believe that it really exceeds 15,000 feet.

"A few characteristic fossils have been obtained in a number of places beyond
the limits of the present map. At Stuart Lake (latitude 54° 30'), Dease River
(latitude 59" 15'), Frances River (latitude 60° 30'), and on Tagish Lake (latitude
60°), fusuline limestones have been obser\-ed.

"To the westward of the Coast Ranges (in which it is probable that numerous
infolds of Paleozoic rocks will yet be found) a formation known from its fossils
to be of Carboniferous age is again well represented. This has, so far, not been
very minutely examined or reported in detail, but it is known to comprise thick
beds of limestones, argillites, and volcanic materials, the latter being even more
characteristic and in greater development than in the region here specially dealt
with.

"In the Rocky Mountains proper, or eastern member of the Cordilleran sys-
tem, the section which must now be regarded as the typical one for these latitudes
is that worked out by Mr. R. G. McConnell. In this section the Carboniferous

176 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

period is represented by the Banff lime series, which, including two shaly zones,
has a thickness of 5,100 feet. This has yielded a number of fossils and these
show that the series as a whole represents the lower part of the Carboniferous,
passing below into the Devono-Carboniferous. The later part of the Carbonifer-
ous period seems either to be unrepresented, or, if represented at all, to find but
a partial equivalent in the upper shales. It is thus very probable that before
the close of the Carboniferous the present position of the Rocky Mountains
formed part of a land area.

****** iti

"It has already been noted that the lower portions of the Cache Creek forma-
tion may be older than the Carboniferous period. The very general blending
of the Carboniferous and Devonian systems in the West shows that no well-
marked line need be anticipated at the base of the Carboniferous. The separation
of any beds of Devonian age can only be made in the event of the future discovery
of characteristic fossils. The same may be said respecting the possible existence
of Silurian or Cambro-Silurian beds."

In the Summary Report of the Canadian Geological Survey for 1915,
Camsell,^ writing upon the Stewart, Tacla, and Trembleur Lakes area and
the Omeneca Placer district describes the rocks of the various regions. On
Stewart Lake there is a massive blue limestone which carries Fusulina.
This led Dawson^ to refer it to the Carboniferous. Below the limestone
are cherty quartzites, argillites, and greenstone schist, the latter probably
derived by alteration from volcanics. The beds are much disturbed and
dip at high angles. On Tacla Lake to the north of Stewart Lake there are
blue and gray sandstones and slates with small bands of dark-blue limestone.
In near-by localities (Germania Creek and Mansofi River) are volcanic
flows, tuffs, and breccias or chloritic schists and slates with narrow bands
of ferruginous dolomite. These beds are correlated with those of the
Stewart Lake area by lithological and structural characters only.

(The similarity of limestone here described and its position above the
series of metamorphic sediments and volcanics, to the limestone on the
western side of the great bathylith is at least striking and suggestive.)

In the Bridge River Map Area near Chilkas Lake, Drysdale^ has de-
scribed Paleozoic rocks which he calls Devono-Carboniferous and places in
the Cache Creek Series. He gives the following section :

White Cap schist series 8,ooo± feet. Quartzite, mica schist, andalusite schist,
squeezed conglomerate and sandstone, phyllite, talcosic, sericitic, and chloritic schist.

Bridge River series 9,500 ± feet. Mainly contorted chert and cherty quartzite. Red
arenaceous schist, metargillite, and crystalline limestone in lenses. Flows of basalt mainly
near the top.

' Camsell, Charles, Exploration in the Northern Interior of British Columbia, Summary

Report Canadian Geological Survey for 1915, p. 70, 1916.
' Canadian Geological Survey, Report of Progress, 1871-77, p. 55.
' Drysdale, C. W., Bridge River Map-area, Lillooet Mining Division, British Columbia,

Summary Report Canadian Geological Survey for 1915, p. 75, 1916.

THE LATE PALEOZOIC IN BRITISH COLUMBIA 177

In the Lillooet map area^ and in the area between Lillooet and Chilkas
Lakes practically the same series occurs, according to Bateman.

If any part of this section is Carboniferous it is the White Cap series,
which has been very seriously affected by the great bathylithic intrusions
to the west.

The Highland Valley Copper Camp, Ashcroft Mining Division, near
Kamloops Lake, shows, according to Drysdale,' the Cache Creek series,
including cherty quartzite, argillite, greenstone, and limestone (Marble
Canyon limestone). This is lithologically near to the Bridge River series
and possibly represents the lower part onl}^ being Devonian in age.'

DrN'sdale, in the last-cited paper, after giving the section on the Thompson
River near Kamloops Lake quoted above, makes some remarks, page 132, ufxjn
the —

" Conditions of deposition. — The rocks belonging to the Cache Creek formation
were probably laid down in a Carboniferous sea ('Vancouver Continental Sea'),
slowly transgressing from the northwest up>on a low-lying area ('Cascadia'
positive element) which probably supported abundant vegetation. In this con-
tinental sea, argillaceous, arenaceous, and calcareous sediments were deposited.
The limestones represent the off-shore deposits, whereas the carbonaceous argil-
lites and sandstones represent the inshore phases.

"Marine sedimentation was interrupted at inter\'als by volcanic activity,
which resulted in the accumulation of much volcanic dust and the outpouring
of lavas."

He also remarks upon the "age and correlation," from which the follow-
ing is summarized. Fossils have been found in the Cache Creek series
near Thompson River (on the Caribou Wagon Road, 2.5 miles above the
89-mile stable) which belong between the base of the Devonian and the
top of the Permian. Beyond this region to the north Fusulina has been
found at Stewart Lake, latitude 54° 30', Dease River, 59° 15', Frances
River, 60° 30', and Tagish Lake, 60°.

The Cache Creek series may be correlated with the Attw^ood series of
Daly (Memoir 38, Canadian Geological Survey), zmd the upper part of the
Brookl\'n limestone of Phoenix* and the Gloucester limestone of Franklin.*
With the existence of an angular unconformity above this series and below

' Bateman, A. M., The Lillooet Map-area, British Columbia, Summary Report Canadian
Geological Survej' for 1912, p. 188.
Exploration between Lillooet and Chilfcas Lake, British Columbia, idem, p. 177.

* Drysdale, C. W., Highland Valley Copper Camp, Ashcraft Division, British Columbia,

Summary Rejxjrt Canadian Geological Sur\'ey for 1915, p. 89, 1916.
' For further description of this region see:
Dawson, G. M., Annual Report Canadian Geological Sur\'ey, vol. 7, p. 1-427B, 1896.
Drj'sdale, C. W., Geology of the Thompson River Valley below Kamloops Lake, British

Columbia, Summar>- Report Canadian Geological Surrey for 1912, p. 115, 1913.

* LeRoy, O. E., The Geology and Ore Deposits of Phoenix, Boundary District, British

Columbia, Memoir 21, Canadian Geological Survey, 1912.
' Drysdale, C. W., Geologj' of Franklin Mining Camp, British Columbia, Memoir 56,

Canadian Geological Survey, pp. 14, 144, 1915.
13

178 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

the Triassic it is at least possible that the massive limestones described
are a continuation of the one to the north assigned to Gschelian age by Girty.

At Banfif in Alberta the upper Carboniferous and Permian (?) are repre-
sented by:*

The Upper Banff Shale. — This is a series of brown, calcareous and often
arenaceous shales, lying conformably upon the quartzite below. It is often
sun-cracked and is interbedded with thin layers of sandstone. A very
common fossil is Schizodus.

The Rocky Mountain Quartzite. — A clean quartzite which represents a
sudden shallowing of the water which did not become muddy. This forma-
tion is about 800 feet thick at Banff, but thickens rapidly to the east, so
that at Lake Minnewanka, 12 miles east, it is 1,600 feet thick.

The Upper Banff Limestone. — This is shaly at the bottom, but more
massive near the top. The lower shaly portion has cherty lenses and cherty
shales interbedded with the limestone.

The Upper Banff shales have been shown by Lambe and Kindle to be
probably of lower Triassic age :

"The Upper Banff shale has been referred in some of the recent reports of
this Survey to the Permian. Since the original reference of the Upper Banff
shale to the Permian was of a provisional character it has seemed desirable to re-
examine the question of the age of these beds in the light of additional evidence
of the last season's collections. I have accordingly brought together all of the
available collections from these beds, including Professor Shimer's collection
from the Lake Minnewanka section, and referred them to Dr. Geo. H. Girty of
the United States Geological Survey, who has a wide acquaintance with the
faunas of this and related horizons in the Rocky Mountains of the United States.
Dr. Girty concludes that these faunas represent the horizon of the Lower Triassic
(Meekoceras beds) of Idaho, Utah, and Wyoming. If this opinion is correct, as
I believe it to be, the reputed Jurassic beds from which the fossil fish transmitted
to you were obtained and the 'Permian' of the Banff map as well should be re-
ferred to the Triassic. Inasmuch as most of the species in this fauna are new,
this determination will have to rest for the present on evidence of a somewhat
general character." "

The evidence for the Permian age of these beds is set forth by Shimer,'
Burling,* and Lambe.*

' Allan, J. A., Rocky Mountain Section Between Banff, Alberta, and Golden, British Colum-
bia, along the Canadian Pacific Railway, Summary Report Canadian Geological Survey
for 1912, p. 173, 1913. Rocky Mountains, Guide Book No. 8, pt. n, p. 183, 1913,
issued by the Geological Survey, Ottawa, 1913.

' Kindle, E. M., in Lambe, Ganoid Fishes from near Banff, Alberta, Trans. Royal Society
of Canada, series 3, vol. 10, p. 37, 1916.

' Shimer, H. W., Lake Minnewanka Section, Alberta, Summary Report Canadian Geo-
logical Survey for 1910, pp. 145-149, 191 1.

* Burling, L. D., Notes on the Stratigraphy of the Rocky Mountains, Alberta and British

Columbia, Summary Report Canadian Geological Survey for 1915, pp. 97-110, 1916.

* Lambe, L. M., Description of a New Species of Platysomus from the neighborhood of

Banff, Alberta, Trans. Royal Society of Canada, vol. 8, p. 17, 1914.

CHAPTER VI.

THE LATE PALEOZOIC IN ALASKA.

While our knowledge of the geology of Alaska is still very imperfect,
reconnaissance work has gone so far that we may say with some confidence
that no rocks of true Permian age exist in that region and none, with the
possible exception of certain series in the southeastern portion, which may
represent the interval of time recorded in the "Permian Red Beds" of the
southwestern United States. The series reported in the earlier reports as
Permian have one after the other been shown by their invertebrate fauna
to be lower in the series, as their fossils have very decided affinities with the
Gschelian, upper Pennsylvanian, stage of Russia rather than with the true
Permian.

It is, as yet, impossible to delimit the areas of deposition either strati-
graphically or geographically \\-ith any certainty or to give any thing more
than a very approximate account of the conditions of deposition, but a very
general idea of the prevailing conditions at the end of the Paleozoic may be
oflFered with some confidence in the accuracy of the broad outiines.

The following statement from Brooks^ is an excellent introduction to
such a description:

"The records of the succeeding ef>och are obscure, but indicate that lime-
stone deposition continued in some places during early Carboniferous times,
while in others a considerable land area was exposed to erosion. Deposition was
probably almost entirely checked by a crustal mo^•ement which took place about
the beginning of the Permian, and this was followed by subsidence. On the
Yukon there is evidence of an extensive period of erosion which immediately
antedated the deposition of the Permian sediments, but this has not been recog-
nized elsewhere.

"The Permian sea seems to have covered much the lai^er part of the province,
for its sediments have been found along the Yukon, in the Panhandle, and in
the Copper Basin, where they aggregate 7,000 to 8,000 feet. In the two latter
districts the deposition apparently continued unbroken into the Triassic and was
characterized by a gradual change from limestones to shales. It was ended
by a crustal movement which deformed the beds, exposed a considerable land-
mass, and thus b^an another period of erosion. This uplift seems to have begun
in Permian times in northern Alaska and progressed gradually southward during
Triassic times, for the latter period does not seem to have left any sedimentary
record north of the Pacific Mountains. In southeastern Alaska the indications
are that the Permian-Triassic cycle of deposition was closed by an extensive

* Brooks, Alfred H., The Geography and Geology of Alaska, U. S. Geological Survey,
Professional Paper No. 45, p. 265, 1906.

179

180

ENVIRONMENT OF VERTEBRATE LIFE, ETC.

dynamic revolution which metamorphosed and deformed the Paleozoic sediments
and probably left them very much as they are now."

On page 224 of the same paper Brooks makes the following statement:

"A belt of Carboniferous rocks (probably Permian) has been fairly well traced
throughout southeastern Alaska; the continuation of the strike of these would
carry them into the White, Tanana, and Copper River basins, where there is a
very extensive development of Permian beds and also some lower Carboniferous
rocks; Permian beds have also been found on the Yukon, near the Arctic Circle,
resting unconformably on strata of lower Carboniferous age. In northern Alaska
the presence of two Carboniferous horizons is established by fossil evidence,
though there is little definite knowledge of their extent or stratigraphic and
structural relations. In all except this northern field igneous intrusives are found
cutting the Carboniferous rocks, and in southeastern Alaska it is probable that
some of the associated greenstone schists are ancient lava flows. In the Copper
River basin volcanic rocks are characteristic accompaniments of the sediments
referred to the Permian [Upper Pennsylvanian]."

It will be noted that Brooks speaks freely and with some degree of cer-
tainty of the age of the rocks which he regards as Permian, but, as is shown
by the quotations given below, it is altogether probable that these rocks
are of a lower horizon equivalent to the Gschelian of Russia, and such a
correction has been inserted parenthetically in the quotation from his paper.

Following are given characteristic sections and descriptions of some of
the better-known areas which will serve to illustrate the conditions which
prevail in the Alaskan Province.

Rocks originally considered as of Permian age are known from the
Panhandle and from all of southeastern Alaska at intervals from the lower
end of the Alexander Archipelago to the headwaters of the Copper River.
Wright and Wright^ give the following succession :

Permian (Upper Carboniferous)

Probably Permian (Upper Carboniferous)

Permian to Pennsylvanian (Upper Car-
boniferous)

(?) _ Feet.

Cherty limestone, breccia, and conglomerate with

fossils 1,000 ±

(?)
White, cherty, semi-crystalline limestone with fossils. 800 ±
Conformity.

Conglomerate, no fossils; same as above 100 ±

(?)

Slate, greywacke and conglomerate; no fossils 3,000 +

Unconformity.
Slate, greenstone, lava, agglomerates, tuffs, and
breccias, intermixed with argillaceous graphitic
slates and schists; metamorphosed and inter-
folded in bed-rock complex. No fossils 4,000 +

(?)
Slate, amphibole, chlorite, and mica schists, inter-
stratified with f ossiferous limestone 4,000 ±

(?)

Light colored limestones, fossils 600 +

Conformity.
Sandstones, conglomerates, and argillites, containing

igneous material, with fossils

(?)

' Wright, F. E., and C. W. Wright, The Ketchikan and Wrangell Mining Districts, Alaska,
U. S. Geological Survey Bull. No. 347, p. 34, 1908,

THE LATE PALEOZOIC IN ALASKA 181

The four members of the last series are typically exposed; the first, on
the west coast of Baranoff Island, Chicagoff Island, Douglas Island, Gravina
Island, and the Cleveland Peninsula; the second, near Taku Harbor and
George Inlet; the third, at Saginaw Bay on Kuiu or Kiku Island; the
fourth, at Soda Springs Bay. The upper series (Permian) (Upper Penn-
sylvanian) is well shown at Hamilton Bay, on the Screen Islands, and at
Pybus Bay, where the rocks are steeply tilted, folded, and metamorphosed;
the middle series occurs at Sitka, Cape Edward, Douglas Island, and the
Glass Peninsula, and the rocks of this series are also tilted, folded, meta-
morphosed to a schistose condition, and indurated. Fossils from the upper
series of rocks were examined by Girty, who records his opinion in the
Bulletin cited, page 53, that though they have been previously determined
as of Permian age in the Alaskan Range, he regards them as rather lower in
pKJsition. Of the two upper series he says:

"These two series, but esf>ecially the upper, are what have previously been
determined as Permian in the Alaska Range, but I really find that the resemblance
with the Gschelian stage of the Russian section is greater than with the Russian
Permian. Provisionally, therefore, I will correlate this horizon with the Gschel-
stufe, in which occur a great number of equivalent or identical species. This
fauna is entirely unlike anything in the Mississippian Province of the United
States, but some of our western faunas resemble it."

This fauna is the same as is found in the McCloud formation of northern
California.

As this fauna is quite similar to others collected in the same general
region, the remarks quoted above might be applied to the whole of the
heavy limestone in the Admirality Islands. A faunal list is given on pages
52 to 55 of Bulletin 347, United States Geological Survey.

Wright and Wright, speaking of the Upper Carboniferous as including
the Permian, say that argillites were extensively laid down at the end of
the Carboniferous, which indicates an elevation of the land at the close of
the Permian (?) in some adjacent region (probably to the north and east —
Case), and that a period of volcanic activity of long duration ensued.
"The beds of lava and ash ejected from the volcanic vents were contem-
poraneous with the slate beds, and because of their intimate association
with the sediments of volcanoes are regarded as submarine intrusives."
The series of limestones, and with the associated argillites, greenstones,
schist, etc., the Ketchikan series, extends in a generally northwest direction
toward St. Elias. Brooks states that Wright found doubtful Permian
fossils in the Porcupine placer gold district in southeastern Alaska, but in
Wright's report^ I find only mention of Mississippian fossils. On the slopes
of the Alaska Range the same series occurs.

* Wright, C. VV., The Porcupine Placer District, Alaska, U. S. Geological Survey Bull.
236, 1904.

182 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

From the Nabesna and White Rivers, Moffit and Knopfs reported the
following section:

Massive limestone.

Shale, some tuffs, and lava flows.

Basic lava and pyroclastic beds with some shale and limestone beds.

The upper member, the massive limestone, is the one traced by Brooks
along the north base of the St. Elias Range to the Nabesna River through
the White River Basin (Professional Paper 45, p. 222.)

In the Chisana- White River District Capps^ reports,

Lava and pyroclastic beds, with some shale.

Massive limestone with shale, thin bedded limestone and a little sandstone and

conglomerate.
Lava and pyroclastics with a small amount of sediment.
Massive limestone of Skolai Creek with interbedded lava and minor amounts of

shale and conglomerate.
Basic bedded lavas with little sedimentary material.

The lower part of this section may be Devonian or Lower Carboniferous
as the Devonian shades into the Carboniferous; the upper two members of
the section are probably Pennsylvanian (Permian). The upper part of
the section is closed by an unconformity.

The fossils are from one general zone "which seems the same zone as
that near Circle City described by Spurr (Takhandit series) or a closely
related one. I have made no specific determinations, since the fauna is
not to be correlated with the Upper Carboniferous of the Mississippi Valley,
but with the Fusulina zone of China, India, and the eastern slopes of the
Urals." »

Abundant fossils were collected in the Nabesna-White River region,
which were mostly referred by Girty to the Russian Gschelian.* It is very
probable that the massive limestone is the same as that in the Panhandle,
which has been traced into this region, and that the two lower series are the
equivalent of the lower beds in the same region.^

In general. Carboniferous rocks make up the north flank of the St.
Elias Range between the White and the Nabesna Rivers. Most of the
pyroclastic rocks were laid down in water, for they are in places interbedded
with sediments and locally contain fossils. It is probable also that some
of the lavas were discharged into the sea and cooled under water. The
tuffs and breccias are made up of material which was ejected violently
from volcanoes and which fell into bodies of water, there to be deposited in

' Moffit, F. H., and Adolph Knopf, Mineral Resources of Nabesna-White River District,

Alaska, U. S. Geological Survey Bull. 417, p. 16, 1910.
' Capps, S. R., The Chisana-White River District, Alaska, U. S. Geological Survey Bull.

630, p. 39. 1916.
' Girty, Geo. H., in Bulletin 630, just cited, p. 45.
* Moffit, F. H., and Adolph Knopf, Mineral Resources of Nabesna-White River District,

Alaska, U. S. Geological Survey Bull. 417, p. 20, 1910.
'^ Idem, pp. 24-27.

THE LATE PALEOZOIC IN ALASKA 183

company with shales, sandstones, and flows of lava that entered the sea
and cooled. The pyroclastic beds are composed of angular fragments of
rocks, little decomposed, and contrasted with the materials derived from
the decomposition and erosion of land-masses, carried by streams or ocean-
currents and deposited to form ordinary sandstone and shale. The shales
are black to bluish and gray. All gradations occur from typical fine-grained
black shale through limy shale to impure argillaceous limestone and fine
sandy shale to sandstone.

In the central Copper River district, MendenhalU reports Permian (?)
sandstone, shale, and limestone 6,000 to 7,000 feet thick, with included
intrusive sheets and some pyroclastics. His section is as follows:

6,700 Black shale.

6,000 Limestone, very fossiliferous.

Black shale.
5,000 Heavy-bedded, pure-gray limestone, diabase intrusives.

Limestone and sandstone beds, fossils.

Dark limestone.

Sandstone, fossils.
4,000 Black limestone.

Shale.

Gray feldspathic sandstone.

Buff limestone.
3,000 Thin limestone and sandstone.
2,000 Fossils.

Thin sandstone and tuffs.

Coarse tuff.
1 ,000 Greenish sandstone and shale.

Fossils,
o Thin limestone.

This section was made in an isolated mass cut oflF from the rest of the
country by a fault, but it is probably of the same Gschelian, and lower, age
as the Ketchikan series. Mendenhall considers this series equal to the
Chitistone limestone of Schrader and Spencer and the Nabesna limestone
of Schrader. It is very probable that the upper prevailingly calcareous
part is to be reckoned as Gschelian in age and the lower arenaceous and
tufaceous part as lower Carboniferous.

Similar deposits are found in the Kenai Peninsula. On the headwaters
of the Gulkana and Susitna Rivers, Martin, Johnson, and Grant^ and
Moffit' record Carboniferous slate, tuff quartzite, and limestone. In the
Hanagita-Bremmer region Moffit* records Carboniferous (?) schist, slate,
and limestone. Many beds of limestone are from 100 to 200 feet thick.

' Mendenhall, W. C, Geology of the Central Copper River Region, Alaska, U. S. Geological

Survey, Professional Paper No. 41, p. 40, 1905.
' Martin, G. C, B. F. Johnson, and U. S. Grant, Geology and Mineral Resources of the

Kenai Peninsula, Alaska, U. S. Geological Survey Bull. 587, 1915.
' Moffit, F. H., Headwater Regions of Gulkana and Susitna Rivers, Alaska, U. S. Geological

Survey Bull. 498, 191 2.
* Moffit, F. H., Geology of the Hanagita-Bremmer Region, Alaska, U. S. Geological Survey

Bull. 576, 1914.

184 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

The earliest history of this whole region (Nabesna, White, Copper, and
Chisna Rivers) —

"shows that marine conditions prevailed throughout the region, but that the
normal course of sedimentation was repeatedly interrupted by the extrusion of
andesitic and basaltic lavas and the ejection of tuflfs and breccias. Where the
accumulation of ordinary clastic sediments went on uninterruptedly, sandstones,
shales, and limestones were laid down, and now bear evidence in their wealth of
fossil remains, that the seas teemed with life, among which huge zaphrentoid
corals flourished in great abundance. The marine occupation appears to have
continued until early Mesozoic time." ^

In central Alaska, Prindle^ reports Permian or Pennsylvanian rocks in
the Fairbanks and Rampart Quadrangles of the Yukon-Tanana region.
These consist of gray, greenish and black shale with siliceous beds. Prindle,'
writing upon the Fairbanks Quadrangle, gives the following list: Gray and
black shale, with some chert, and probably some limestone resting uncon-
formably on the beds below. This series is regarded as equal to the Nation
River series of the Yukon, but may be equal in time to the unconformity
between the Nation River and the Calico Bluff (Mississippian).

Brooks and Kindle* report for the Upper Yukon a heavy limestone
(Gschelian in age) unconformably above a mass of sandstone and shale,
3,700± feet thick, with fragments of plants and some small seams of bitumi-
nous coal. There are two layers of conglomerate, one at the base, very mas-
sive, and a second, not so heavy, about i,ooo feet above the base.

Invertebrates are reported from a heavy limestone just below an un-
conformity in the Upper Yukon region, which Girty considers as Gschelian
in age, the same fauna as at Pybus Bay, Kuiu Island. The rocks are not
metamorphosed in this region and there is no evidence of volcanic activity,
except at Glenn Creek.

In the upper part of the Tanana Basin, Schrader reports Permian fossils
in the Suslota limestone, which lies above the Nabesna limestone.

Keele" and Camself found a limestone between the Peele and Stewart
Rivers which is probably the equivalent of the massive limestone of the
Yukon area. This limestone is a "massive, granular limestone containing

1 Moffit, F. H., and Adolph Knopf, Mineral Resources of the Nabesna-White River District,

Alaska, U. S. Geological Survey Bull. 417, p. 47, 1910.
' Prindle, L. M., The Fairbanks and Rampart Quadrangles, Yukon-Tanana Region, Alaska,

U. S. Geological Survey Bull. 337, 1908.
' Prindle, L. M., A Geologic Reconnaissance of the Fairbanks Quadrangle, Alaska, etc.,

U. S. Geological Survey Bull. 525, 1913.

* Brooks, A. H., and E. M. Kindle, Paleozoic and Associated Rocks of the Upper Yukon,

Alaska, Bull. Geol. Soc. Amen, vol. 19, p. 255, 1908.
^ Schrader, F. C., Recent Work of the United States Geological Survey in Alaska, Bull.
Amer. Geog. Soc, vol. 34, pp. 1-145, 1902.

* Keele, Jos., Report on the Upper Stewart River Region, Yukon, Annual Report Canadian

Geological Survey, vol. 16, pt. c, p. 14, 1906.
' Camsell, Chas., Report on the Peele River and Tributaries, Yukon and Mackenzie, Annual
Report Canadian Geological Survey, vol. 16, pt. cc, p. 16, 1906.

THE LATE PALEOZOIC IN ALASKA 185

fossils." Keele refers to it as a "mass of white, bedded, crystalline lime-
stone forming the greater part of a mountain group."

Dawson found a siliceous crystalline limestone on Tagish Lake which
he considered as Carboniferous and correlated with the Cache Creek series.
(Canadian Geological Survey, 1887, part B, p. 170B-171B.) This limestone
has been shown by later work of Dawson, Gwillen, and Kindle to be very
doubtful in position and unsafe for use as any basis of conclusion.

The following summary statements are given by Mendenhall, and by
Brooks and Kindle. Mendenhall* says:

"We have seen that the earlier and middle Paleozoic history seems to have
been largely continental — that of a land-mass, or its shore-lines, with coarse
sediments and volcanic materials; but with the end of the Paleozoic the sea
invaded the pro\ince through a general subsidence following the outflows of the
Nikolai greenstone, and although no doubt varying in depth and shifting its
outlines, practically maintained control until the middle of the Mesozoic.

" In the northern part of the province the sea seems to have only gradually
invaded the land areas. The earliest Permian sediments there are shore deposits
in part, and include sands, tufaceous beds, and flows, which denote the dying out
of the earlier eruptive epoch, perhaps the last feeble activity of the Tetelna stage.
But toward the middle of the period represented by the Permian rocks truly
marine conditions prevailed and abundant marine life existed, while heavy
limestones and fossiliferous black shales were being laid down in what is now
the upper Copper River Valley. It is probable that this Permian sea wais wide-
spread in eastern Alaska. Its records are preserved on the middle Yukon, in the
Copper \'alley, in the valley of the upper ^^^lite, and east of there toward Pyramid
Harbor. It extended also to southeastern Alaska, where, however, the recognized
beds belonging to it are marine sandstones instead of limestone. The even,
exclusively marine phase of the upper part of the deposits makes it wholly im-
probable that there existed at this time any marked relief where the present great
ranges stand. They may have been mountains before the Permian and after,
but probably not during this era.

"In the Chitina Valley a shallow sea held possession through the Permian
epoch and well into the Triassic, the deposition of thin, dark limestones and black
shales continuing uninterruptedly."

Brooks and Kindle say:'

"The Carboniferous of Alaska is represented by three stages, the first of
heavy limestone partly Devonian and partly lower Carboniferous. This steige is
succeeded by a series of rocks in the Yukon area which 'are made up of littoral
and in part at least of fresh-water deposits, embracing some very coarse material
and aggregating nearly 3,000 feet in thickness. This same epoch of deposition is
probably represented in the White-Copper River region and in the Alaskan Range,
where, however, it appears to form the base of the Carboniferous, indicating that,
if the older limestone had ever been present in this area, it was removed by

' Mendenhall, VV. C, Geology of the Central Copper River Region, Alaska, U. S. Geological

Survey, Professional Paper No. 41, p. 76, 1905.
' Brooks, A. H., and E. M. Kindle, Paleozoic and Associated Rocks of the Upper Yukon,

Alaska, Bull. Geol. Soc. .\mer., vol. 19, p. 304, 1908.

186 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

erosion before the coarse sediments were laid down. The same horizon appears
to be represented in southeastern Alaska by argillites. The third epoch of deposi-
tion is one of calcareous material, and here again there was an abrupt change, this
time to deep-sea conditions and marked by the appearance of a new fauna. This
Upper Carboniferous sea was probably widespread. Its thickest recognized de-
posits are found in southeastern Alaska, but it seems quite probable that the
upper limestone member of the Cache Creek series was deposited in it. Sedi-
mentation may have continued unbroken into Triassic times in the Yukon
district, but in southeastern Alaska a period of erosion intervened between the
highest Paleozoic and the lowest Mesozoic.'"

CHAPTER VII.

INTERPRETATION OF THE ENVIRONMENTAL CONDITIONS
FROM THE EVIDENCE OF THE DEPOSITS.

A. PERMO-CARBONIFEROUS CONDITIONS VERSUS PERMO-

CARBONIFEROUS TIME.

From the preceding account of the stratigraphy of the upper Paleozoic
it is evident that a distinct line may be drawn between the red beds and the
beds beneath them in all of the provinces; it is equally clear that this line
will not lie at equivalent stratigraphic levels in all places. If, as is assumed
as the thesis of this paper, the red beds and their equivalents are indicative
of new environmental conditions, then a clear distinction may be drawn
between what has long been called Pennsylvanian and Permo-Carboniferous
(Permian) time, and Pennsylvanian and Permo-Carboniferous conditions,
and the two are not coincident in all places.

The convenience and relative ease of synchronizing definite time inter-
vals with definite deposits for purposes of classification is obvious, but it is
equally obvious that time is not the sole nor even the dominant factor in
evolution. Environmental conditions are far more effective in determining
the rate and direction of evolutional and distributional changes, and these
conditions may migrate both vertically and horizontally, independent of
natural or arbitrary divisions of time. To assume that similar conditions
necessarily prevailed over large areas through even a minor interval of
geologic time is to introduce a possible error of the first dimension. This
warning is especially applicable in interpreting a series of terrestrial or semi-
terrestrial deposits. The evidence that changes in environment, climatic,
physiographic, topographic, and organic, migrate across large areas is too
full and conclusive to admit of any question that such processes are at work
at the present time. That such changes might have originated and migrated
in a strictly similar manner in past time is equally beyond question. When
past time is considered it is evident that migration may have proceeded in
a new direction, upward, for if the migration is a slow one the new conditions
will reach any new localities at progressively later dates and will be recorded
in stratigraphically higher deposits.

The advance of conditions fitted for a certain group of organisms might
then be recorded in the stratigraphic series by observable changes which

187

188 ENVIRONMENT OF VERTEBRATE LIFE, ETC,

would be oblique to the beds; time and environment will not coincide, and
to correlate widely separated groups of beds as synchronous in deposition
because of a similarity, even approaching identity, in the fauna or flora
would be a serious error.

The condition, or better, the changing conditions, of upper Paleozoic
time, treated in a broad way, furnish an excellent concrete example of such
a migration of environment both vertically and horizontally.

It is well known that under the influence of a major earth-movement
the continent of North America was uplifted in a progressive way from east
to west during upper Paleozoic time.^ Swamp and terrestrial conditions
initiated in the east advanced toward the west, displacing the marine condi-
tions even to the westernmost limits of the Plains Province and over the
southern part of the Basin Province.

In this connection certain remarks made by Lee* are of especial interest:

"During much of the Carboniferous period sea water covered large portions
of the area now occupied by the mountains. Marine limestone of Pennsylvanian
age is abundant in central and northern New Mexico and in central and western
Colorado. It has been the belief of many geologists that open sea conditions
prevailed in western America during the time that the coal measures were form-
ing in the eastern and central parts of the continent. Statements are frequently
made that ' in the western part of the United States there are no coal accumula-
tions of this age (Pennsylvanian).^ There is unquestionably a large amount of
limestone of marine origin in the rocks of Pennsylvanian age in the Southern
Rocky Mountain Province, but there are also thin beds of coal and plant-bearing
sedimentary rocks which indicate lowlands and coastal swamps in Pennsylvanian
time. These have been found in New Mexico, near Socorro; in the mountains
east of Albuquerque; near Santa Fe on the western slope of the main range; in
the Pecos Valley between the mountain ranges; east of the main ranges near Las
Vegas; and farther to the north in the Moreno Valley. Thin beds of coal of
Pennsylvanian age have been reported from many places in central and western
Colorado and in eastern Utah, both north and south of the Uinta Range. These
coal beds are not thick enough to be of commercial value, but they prove that the
physiographic conditions of the Rocky Mountain region during the early part of
the Pennsylvanian epoch were not so different from those in eastern and central
North America as many geologists have supposed. However, later in the epoch
these coals were covered by the sea in which was formed the massive limestone
of Pennsylvanian age in New Mexico and southern Colorado, which seems to
indicate clear water and open sea conditions."

* Girty, G. H., Outlines of Geologic History, chap, vi, pp. 126-127, 1910.
Schuchert, Chas., Paleogeography of North America, Bull. Geol. Soc. Amer., vol. 20,

chart of submergences and emergences and map of Lower Permian.
Ulrich, E. O., Revision of the Paleozoic System, Bull. Geol. Soc. Amer., vol. 22, chart on

P- 343-
' Lee, W. T., Early Mesozoic physiography of the Southern Rocky Mountains, Smithsonian

Miscellaneous Collections, Vol. 69, No. 4, p. 4, 1918.
' Schuchert, Chas., Textbook, p. 745.

INTERPRETATION OF ENVIRONMENTAL CONDITIONS

189

In the northwestern part of the continent this uplift was approximately
paralleled in time by a disturbance which started in Alaska and advanced
to the south as far as northern California. Pirsson and Schuchert* say of
this latter disturbance:

"While it appears that the greater part of the continent was not undergoing
more than crustai warping in Pennsj'lvanian time, there seems to have been more
decided unrest along the entire Pacific area from northern California into arctic
Alaska. Here the limestones and calcareous shales of the Pennsylvanian and
the early Permian are interbedded with much extrusive igneous material. The
thickness in California is not less than 4,600 feet, with a maximum of 10,000 feet,
while in the Copper River region of Alaska it is nearly 7,000 feet. The calcareous
deposits often abound in fossils unrelated to those found elsewhere in North
America; they are of the Pacific realm, while the life record of the eastern seas
accords better with that of northern Europe, though in the main it is best regarded
as constituting the North American province. Long after Pennsylvanian time
had begun there was also decided crustai movement in western Colorado, Utah,
and Arizona, and this is probably to be associated with the deformation of the
Pacific border."

As shown in the summary description of Alaska, the limestones and cal-
careous shales of Alaska, British Columbia, and the Pacific coast involved

Table 3. — Abbreviated correlation table showing the rise of the red deposits from easttovoest in the stratigraphic
series. Red beds or equivalenl intervals lie above the korisontal lines in the table.

N. M.

Okla.

Tn«»

Kansas

Ho.

Iowa

iDd.

Ky.

Ohio

W. Va.
and Peon.

Pecos
Castile

Rustler

Quarter-
master

Double
Mt.

Salt Fork

Shale
and
undif-
feren-
tiated

•H
§

Kiger

Greer

Wreford

Is.
Elmdale

Wood-
»-ard

Blaine

Clear

Fork
Wichita

Redsand-
stone and
shale of
Webster
Co.

Enid

Merom
sand-
stone

Capitan

Dela-
ware

Ralston

Cisco Is.

Coal 7

Coal
VIII

Shale
and
sand-
stone

c
0

Chandler

lola Is.

iCoal6

Coal
VII

Coals
13-18

Round
Knob

Saltsburg

be
3
a
E

= Pitts-
burg Red
Shale

in this movement are not higher than the Russian Gschelian, and so earlier
than the time of the formation of the red beds in the Western and Basin
Provinces. Red beds and their equivalents, the Park City formation, can
be traced into Montana, but the union of the two earth-movements seems
to have raised the land north and west of this above the plane of deposition
before the red-bed conditions (Triassic) reached that far west.

' Pirsson and Schuchert, A Text-Book of Geology, p. 744, 191 5.

190

ENVIRONMENT OF VERTEBRATE LIFE, ETC.

As noted above, the progress of Permo-Carboniferous conditions during
Pennsylvanian and Permo-Carboniferous time is marked by the disappear-
ance of marine and swamp conditions and the appearance of more purely
terrestrial deposits. The change from swamp to terrestrial deposition is
marked more by the chemical condition than the physical character of the
material — ^that is, the change was more in the climate than in mode of depo-
sition.

As the climatic change was the most significant factor in the devel-
opment of the land vertebrate life, and as its progress is most strikingly
marked, it is taken as the key for the interpretation of the whole environ-
mental migration.

As has been shown in the summary of the stratigraphy of the different
provinces, in the correlation tables i and 3, and in the accompanying schem-

FlG. 5. — Diagram illustrating the record of a progressive wave of definite conditions advanc-
ing across an area in which continuous sedimentation is in progress. The upper and
lower limits of the sediments deposited during the continuance of the definite condition
will lie obliquely to the bedding planes. The upper and lower limits may be marked by un-
conformities or disconformities, but not necessarily.

mmWi

^^^^^^Mllllillld

^^Uii^j^

-iiJijl c

^^^^^MilUiUlD

Fig. 6. — Diagram illustrating the record of a progressive extension of conditions which persist
at the point of origin. The lower edges of the deposit formed during the definite condi-
tions will be oblique to the bedding planes, but the upper edge may be straight, forming a
wedge of the particular sediments, or the upper edge may be above the plane of erosion.
Such a wedge will be formed only if sedimentation is continuous at the point of origin of the
progressive conditions.

atic diagrams (figs. 5 and 6), the climatic change is recorded in conglomer-
ates, glacial deposits, and red beds and can be clearly traced from east to
west in a^ line ascending obliquely across the stratigraphic series from the
middle of the Conemaugh in the Eastern Province to the base of the so-
called Permian in the western provinces. The base of this series of red beds
and conglomerates marks the beginning in each region of Permo-Carbon-
iferous conditions, and the upper limit is lost by erosion or lack of deposi-
tion, or merges into the very similar Triassic deposits.

Permo-Carboniferous conditions in the Eastern Province are not coinci-
dent with Permian strata of current classification, but occur in a part of the
Pennsylvanian strata as well; they encompass a definite period, however,

INTERPRETATION OF ENVIRONMENTAL CONDITIONS 191

marked by distinct environmental conditions, and the correlation of the
strata within this time interval depends largely upon the recognition of the
results of the operation of climatic factors.

The ideas here suggested are in decided variance with accepted methods
of correlation, especially in the conception of the limitation and retardation
of animal and plant migration by unfavorable conditions.

Ulrich says:^

"I have strong convictions respecting the great possibilities of correlation
by a judicious application of organic criteria. Their greatest value in this connec-
tion arises from the demonstrable fact that, as a rule, the migration, and to a con-
siderable extent also the evolution, of species, however slow, is yet relatively
rapid as compared to the inconceivable length of geologic time.

"As to marine faunas, with which the student of Paleozoic stratigraphy is
chiefly concerned, their migrations, when not prohibited by physical barriers,
usually proceeded with such rapidity that their progress can not be expressed in
recognizable units of the geologic time scale. Hence, unquestionable correlations
by fossil evidence, fully checked by physical criteria, may be said to establish, so
far as the practical purposes of geology are concerned, the essential contem-
poraneity of the beds so identified."

Ulrich's remarks are based upon the action of marine invertebrates
almost entirely, but it will be noticed that the only influence that he recog-
nizes as able to deter the rapid spread of such forms is the effect of "physical
barriers." That there is an abundance of other barriers to migration and
evolution will be evident at once to anyone who looks at the fossils as forms
of life, subject in their time to influences strictly similar to those affecting
living forms to-day.

The discovery of vertebrate fossils belonging to identical or closely
related genera, and the evidence of fossil plants, has led to the suggested
correlation of the red beds of Kansas, Oklahoma, Texas, and New Mexico
with the Dunkard series of Ohio and Pennsylvania and with the isolated
deposits carrying vertebrate fossils near Danville, Vermillion County,
Illinois. Such suggestions of correlation, however, do violence to the
probabilities indicated by the stratigraphic position of the beds in which the
fossils are found. Though the correlation of widely separated areas must
be largely accomplished by fossil evidence, it is becoming increasingly
evident to all workers in stratigraphy, as well as paleobiology, that fossils
must be regarded and interpreted as once-living things, and the problem of
their distribution is inextricably associated with the problem of their living
conditions.

The method of evolution is as yet unknown, but all biologists concede
the directive influence of environment, once a line is started. In other
words, the evolution of life follows and responds to the changed conditions

* Ulrich, E. 0., Revision of the Paleozoic Systems, Bull. Geol. Soc. Amer., vol. 22, p. 507, 191 1.

192

ENVIRONMENT OF VERTEBRATE LIFE, ETC.

in the inorganic environment. If this be true, the beginning of a new
geological interval of time is marked by the change in the inorganic world
which will lead by slow degrees and a multiplicity of processes to the de-
velopment or immigration into a definite area, of new forms of life. The
interval begins with the establishment of new conditions permitting the
establishment of a new life in sufficient abundance to be recognized as con-
stituting a new fauna, faunule, or flora. On the other hand, it is very pos-
sible that the establishment of new conditions might be followed by the
immediate introduction of a new fauna or flora by immigration. These
ideas are in strict consonance with the determination of geological intervals
on the principles of diastrophism. If any progressive criteria can be de-
tected and traced which reveal such a change in the inorganic world, then
the evidence from life may be better understood and even in some measure
anticipated.

The commonly obscure and neglected evidence of climatic change induced
by the slow and unimpressive uplift of the continent is the one here used,
and, as shown above, the progress of the changed climate is recorded in the
line which marks the beginning of red sediments and which cuts obliquely
across the stratigraphic column. The persistence of the climatic conditions
on the east while they migrated west gives the resultant deposits a wedge-
shape (see figs. 5 and 6).

Texas, Okla. Kans. Mo.

111. Ind. Ohio Ohio.W.Va. Penn.

Oklahoma^
Missourian |

Duokard
Mbnmgahela ,
ConeiDaugh

Fig. 7. — Diagram illustrating in a schematic way the relative position of the sediments formed
under Permo-Carboniferous conditions. The land was rising from east to west, but there
was continuous sedimentation in the eastern region at the western edge of the rising land of
Appalachia. As the land rose slowly the red beds spread toward the west, occupying rela-
tively higher positions in the stratigraphic column. It is difficult to illustrate the actual
conditions in the diagram, because the " red beds conditions " were advancing, but the wavy
lines indicate the surface of the ground relative to these conditions. In Pennsylvania and
West Virginia deposition was continuous during the conditions. In Illinois and Indiana
deposition had ceased by the time the conditions reached that far west; in Kansas, Oklahoma,
and Texas " red-bed conditions " reached the region in time to affect only the uppermost
Paleozoic deposits. The upper limit of the red-bed conditions is not known, and so the
upper limit of the wedge is indicated by a dotted line.

By all the commonly accepted canons of correlation and by its strati-
graphic position the Dunkard, with its typical Permo-Carboniferous flora,
fauna, of invertebrates, and single reported vertebrate (Edaphosaurus) , is
the very approximate equivalent of the Wichita-Clear Fork beds of Texas,
but both red beds and Permo-Carboniferous vertebrates are found far below
the Dunkard in West Virginia and Pennsylvania. They are, however, well
within the limits of "Permo-Carboniferous conditions" in those States.

INTERPRETATION OF ENVIRONMENTAL CONDITIONS 193

The occurrence of Permo-Carboniferous reptiles and amphibians as low as
middle Conemaugh is no longer a puzzle; the animals appeared with the
environment and migrated with it; they are strictly within the limits of
their proper environment wherever they occur.

The sequence in the evidence for the progressive advance of the climatic
change which induced the deposition of red beds is broken in two places —
at the Cincinnati anticline and at the elevation in Missouri. All efforts to
trace the change around these elevations have been only partially successful.
The breaks are in part due to the eflfects of erosion removing all trace of
Permo-Carboniferous deposition and in part to the fact that these areas
were above the plane of deposition before the migrating climatic change
had reached that far west.

One apparent conclusion from the premises here stated is that the Permo-
Carboniferous vertebrate fauna originated in the eastern part of North
America and migrated westward. This the author is not yet entirely
ready to accept, but he is strongly impelled toward the conclusion by the
facts that the earliest known reptile, Eosauravus, was discovered in the
Allegheny series, at Linton, Ohio; that typical Permo-Carboniferous verte-
brates appeared in middle Conemaugh time in Pennsylvania and West
Virginia; and also that typical pelycosaurs occur in the red beds of Prince
Edward Island at a stratigraphic level much lower than those of Oklahoma
and Texas.

B. INTERPRETATION OF CONDITIONS IN ALLEGHENY AND
LOWER CONEMAUGH TIME.

As already mentioned, the conditions during pre-Conemaugh Penn-
sylvanian time were radically different from those in the late Paleozoic.
The clastic sediments, coal beds, mode of deposition, etc., are all those of
an inclosed basin of singularly monotonous character. Perhaps the best
picture that has been given of the various minor basins of the Eastern
Province is that by David White :^

"Terrestrial Cokditioks.
"Base-Level Basins.

"The examination of the strata intervening between the coal beds in the great
coal fields of the earth and the inspection of the coal show conclusively that
nearly everywhere the vegetal matter was deposited and transformed to peat
beneath a water surface. Furthermore, in the great majority of the basins,
including nearly all the great coal areas,* the coal (peat) was laid down not far
from tide-level, marine beds being intercalated at various horizons in the coal-
bearing series of rocks, and not rarely in the beds immediately overlying the
coal itself. In certain basins brackish-water moUusks are found in many of the

• White, Da\-id, The Origin of Coal, Bureau of Mines Bull. 38, pp. 52-60, 1913.

* The very extensive basins of the Fort Union coal in the northern Rocky Mountains region

are regarded as fresh-water basins,
14

194 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

shales or limestones, although no distinctly marine faunas have been observed,
thus indicating for those basins estuarine waters, which at times may have been
completely isolated, though still in close geographic relation to the maiine.
Brackish-water deposits are present at one or more horizons in most of the
extensive coal-bearing formations, whether younger or older, even those which
for the most part contain only fresh-water types of life.' Usually, however,
the formations are apt also at certain levels to carry distinctly marine faunas.
The parallelism of the strata and the frequency of the salt-water invasions show
the constant nearness to tide-level, whereas the horizontal extent of the strata
indicates the areal extent of the region of deposition. It may therefore be
confidently stated that in most of the great coal fields of the world the coal was
formed on a nearly level surface at or near sea or lake level.

"Coastal Swamps.

"The enormous horizontal extent of many of the coal groups as indicated
by the remnants now found in basins detached as the result of folding and erosion,
the demonstrated continuity of some of the individual coal beds over areas some
of amazing size, the high degree of parallelism of the beds, and the recognition
that the coal beds were laid down beneath a water cover, join in predicating the
existence, at the time the coals were being formed, of vast swamps and broad
but shallow lagoonal areas subject to subsidence at a generally slow rate. Such
coastal or lacustrine swamps, of magnitudes unknown in the world of to-day,
appear to have resulted either from the partial submergence of a very mature
and broadly extended surface of erosion (peneplain) or from so extensive and so
nearly a complete filling of great parts of a basin as to develop, under the action
of waves and currents, extremely shallow littoral flats many miles in width.
Following periods of deeper subsidence and the deposition of varying thicknesses
of other materials, such as clays, sands, and calcareous muds, there was recurrence
of proximate filling of the loaded areas of the basin, and occasionally there was
slight uplift of the region and consequent withdrawal of the water, so that the
level of the subaqueous deposits was so near to the surface of the water cover as
again to favor swamp conditions. Sometimes there was sufficient elevation of
the region and retreat of the water to bring the old bottom some distance above
the water level of the basin; but it is probable that actual elevation to any con-
siderable distance above the sea was exceptional and confined in most cases to
the intervals of more marked earth-movements separating many of the geologic
formations. Conditions of relative quiescence and of partial exposure must
have taken place many times during the deposition of series of strata embracing
many coal beds. The periods of such proximation of sea-level and land-level
were times of extension of the vascular plant cover to occupy the shallow regions,
thus initiating the formation of a new coal bed. If the slope of the exposed land
was reduced to a sufficiently low gradient and the rainfall was heavy and well
distributed, the swamp conditions would have been extended far inland, the
drainage of the water being held back, obstructed, by the most luxuriant and
fecund plant growth, so that the state of partial submergence of the land would
have been carried for indefinite distances along the nearly flat land surface. At
such times the true coast-line was doubtless locally obliterated, except in so far
as barrier beaches marked the zone of wave-action and bordered the expanses
of open water.

* For example consult the stratigraphic sections of the great coal fields of Europe and America.

INTERPRETATION OF ENVIRONMENTAL CONDITIONS 195

"Varying Subsides-ce.

"The thickness, and even the survival, of the peat bed thus begun depended
upon several conditions, important among which are contemporaneous subsidence
and its rate. It is probable that at times the water body was greatly contracted,
receding to deeper and perhaps relatively small areas of the bcisins; and at other
times accelerated regional subsidence caused the reextension of the water-level
over the great border zone (littoral). Variation in the rates of subsidence are
w^ell established and can be observed even in the present epoch. At times of more
rapid depression, when the water became too deep to be interrupted by bars or
shoals or when the water-level in the basins was too high and too expansive for
subaerial plants to root and grow on the bottom, the water was open to movement
and clay and sand were distributed by the waves and currents, or perhaps lime-
stone was formed or calcareous mud was laid dow^n. The effect of these oscillatory
movements accompanied by the leveling of the sediments in the basin, as shown
by the studies of such extensive areas of continuous horizontal beds as are seen
in the Appalachian and the Interior coal fields, seems to have been the production
at various times of enormous expanses of broad, shallow pans or l^oons, over the
greater part of whose areas the water was not too deep for occupation by the
dense and luxuriant vegetation of the time. The examination of the stratigraphy
of the Appalachian coal fields and of the Interior basins shows that during the
periods of deposition of the coal-bearing groups the filling of the broad, shallow
Carboniferous basins nearly kept pace, on the whole, with the subsidence, which
obviously varied in rapidity; but the water was never really deep (probably less
than 200 feet in the deepest axes, except at rare periods of greatest subsidence),
even in the relatively restricted central areas of the basin, far from the varying
coast-lines.

"Formation of Bars and Barriers.

"At Other times great areas, probably embracing most of the depositional
region, were either above water or so near the surface that sand barriers or bars,
possibly in series, and probably more or less irregular in plan, developed far out
toward the edge of the submerged shelf, constituting series of lagoonal or land-
locked shallows of enormous aggregate extent. \Miere not too deep or subject to
vigorous incursions of the sea these were occupied by the coal-forming vegetation.
It is also apparent that in parts of the Appalachian trough large areas of the
submerged flats were, for variable periods, isolated from other parts of the basins
by differential movements or warpings of the basin, which produced barriers or
low arches of great linear extent, these barriers having greater magnitude and
permanency than the bars or other lesser barriers just mentioned. Against the
bars, barriers, and shoals that appear to have separated the pans or great swamp
expanses, the coal beds usually pinch out, though it is frequently observed that,
as the result of continued subsidence, or, in cases, as the result of the impounding
of water by the dense vegetal growth, which, in a humid climate, raised the
ground-water level, the peat bridged many of the shoals and bars.

"It is more than probable that during these periods of coal formation in the
broadly extended swamps and Icigoons of fresh water the very luxuriant and
intricate tangle of peat-producing vegetation, much of it of large size, that occu-
pied the shallow and protected areas, effectually obstructed the penetration to
any great distance into the swamps, of salt water, and rendered the peat-forming
areas practically nonmarine a few miles back of their seaward margin, though
portions of these areas may have been at sea-level. Such obstruction would be

196 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

comparable to the resistance ofTered by the same vegetation on a nearly flat land,
where the water of a heavy rainfall is held back sometimes to a considerable
depth by the dense growth and tangle of fallen vegetation, thus extending the
swamp conditions far inland. Many coal swamps were in vast deltas.

"Semiaquatic Vascular Plant Growth.

"The examination of the structure, development, and affinities of the principal
plant types found associated with the Paleozoic coals points to the conclusion
that many of them, especially among the lepidophytes,^ grew under conditions
characteristic of swamps, a large number of them being adapted to growth
in standing water. Pontonie has pointed out that the enlarged bases of the boles
of many of the old Sigillariae and Lepidodendreae appear precisely comparable
to the basal enlargements of the black gum, Jussiaa, the bald cypress, and many
other types growing in water-covered swamps of tropical and warm temperate
climates at the present time. The lateral traces or "parichnos" of the leaf-
scars were long ago recognized by Renault as points of air contact with the tracts
of lacunose assimilation parenchyma (aerenchyma) in the thick cortices of the
trunks. Subsequently Gothan and Pontonie have pointed out that these twin
leaf-scars, often so conspicuously enlarged on the old trunks, where they sometimes
measure nearly a centimeter in length, are in effect lenticels, or breathing surfaces,
developed to afford a better air-supply to tissues where the water cover of a
swamp deprives the root system of access to the atmosphere. Potonie- has shown
also that Sigillaria developed giant "knees," so-called pneumatophores, similar
to those developed by the bald cypress when growing in water, and similarly
provided with breathing-pores. The height to which some of the trunks of
Sigillaria are enlarged justifies the belief they may have grown in some instances
in a normal water cover nearly 4 feet in depth. It has also been conclusively
shown by Grand 'Eury' that most of the larger types were adapted to survive
rapid deposition about their trunks or rhizomes. Evidence that the commonly
associated plants also were suited to an aquatic environment is found in the
shallow root system of Stigmaria, which extended far out horizontally on or
near the surface of the peat in order to avoid the toxic air-exhausted deeper
layers; the occurrence of gum-canals in the rootlets of the fern trunk, Psaronius,
the hollow interior of the Stigmaria appendages and the absence of root-hairs,
distinctly pointing to an aqueous habitat; the presence of air-chambers in the
Calamarian roots, and the development of subaerial roots in many types; the
presence of mycorrhiza, as described by Osbom, in the roots of Cordaites, indicat-
ing a peat substratum for the plant; the air-chambers provided for the flotation
of many of the seeds ; and the production by many of the lycopods and Calamarian
types of two kinds of spores, requiring an aqueous environment in which the two
kinds could more certainly come together in a place sufficiently wet to insure
fertilization.*

* See Grand 'Eury, C, Du Basin de La Loire, Compt. Rend., 8th Cong. Gaol. Internat.,

Paris, 1901, pp. 521-538; Pontonie, H., Entstehung d. Steinkohle, 5th ed., 1910, p. 173.

* Potonid, H., op. cit., pp. 176, 181.

' Grand 'Eury, C, Sur le caract^re paludeen des plantes qui ont forme les combustibles
fossiles de tout age, Compt. Rend., vol. 138, 1904, p. 666.

* The rapid decrease, almost amounting to disappearance, of the greater number of the

very large spored lycopods during Conemaugh time (early Stephanian) was no doubt
due to failure of fructification caused by periods of relative drought and reduction of
the water-surface, such withdrawal of the water being plainly indicated by the preva-
lent pseudoxerophytic characters observed in the swamp plants of the period.

INTERPRETATION OF ENVIRONMENTAL CONDITIONS 197

"The prevalence of thick, smooth barks on the trees is another indication
of a humid swamp environment. * * *

"Proof that the Stigmaria roots (which are known to have belonged to Sigil-
laria and several other related lycopodialean types) have in most cases grown
in the clays where they are now found beneath the coal or in the clay partings,
may generally on close examination be found in (c) the radial position of the inter-
lacing rootlets passing outw^ard from the parent root in all directions and oblique
to the bedding; (6) the penetration of buried pieces of stem, bark, or other partly
decayed roots by the rootlets of later growth, a common occurrence; (c) the fact
that the roots extend outward from the base in a normal manner and are "right
side up," not having been disturbed or overturned.

"Coal Formation in Bkackish or Marine Waters.

"The question arises as to the extent to which coal of the common types may
have been laid in brackish or marine waters. This problem is . particularly
prominent in certain Paleozoic coal fields like those of the Interior basins, in which
many coal beds are capped by black shales, generally containing marine shells.
The occurrence of Stigmaria roots in place of growth in the underclays of the
Carboniferous coal of Missouri, Indiana, and Illinois, even where, as in many
places in Missouri, the underclays rest on thin limestones of marine origin, or the
occurrence of roots of other kinds beneath the Cretaceous and Tertiary coaJ,
including most of those interbedded in marine formations, must be interpreted as
indicating either subaerial conditions previous 'to an initial stage of inundation
of the old soil or a situation at the initial stage of coal formation in which the
water was shallow enough to permit rooting and growth of a dense coal-forming
vegetation in place. In the latter case it becomes probable that here and there
more elevated parts of the bottom, such as shoals, bars, or barrier beaches and
broader barriers, the latter often the result of slight warping of the strata of the
region, rose above the water-level and excluded the sea from free access during
most, if not quite all, of the period of peat deposition. It may further be as-
sumed— and observations of the bedding of the ordinary coal fully support this
assumption — that barriers of these kinds were numerous, and that they divided
the coal field into smaller areas, which in most cases were very numerous and
irregular in form. The latter feature is, of course, normal to the tide-level pene-
plain surface.

"The physiographic conditions, including the advanced degree of filling of the
basins, the perfection of the base-level, the somewhat unequal loading, and the
recognized slight warping of the basin floors, all support the hypothesis that
during the times of coal formation the areas of broad and unbroken sea expanse
were very restricted, being probably confined in most cases to the deeper parts
of the basin,' now usually concealed. That, however, the sea probably did in
some cases break over the coal-formation swamp is indicated by the intercalation
of over- washed muds, silts, or sands, forming partings few of which contain
brackish or salt water shells such as Lingula and Orbiculoides. Other overwashes

' It is not improbable that such an area of deeper water, largely without appreciable coal
deposition, occupies the axis of the Appalachian trough, as defined in Monongahela and
Dunkard time, in southwestern Pennsylvania and northern West Virginia. In the
first-mentioned State the absence of the coal and the reduction in thickness of the
lower coal formations is proven by abundant drillings. Farther south, on the West
Virginia side of the valley, esjiecially where the younger Pennsylvanian formations rise
to daylight, the changed character of the formations is well recognized.

198 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

originated on the land side and in fresh water. The occurrence of marine mollusks
in the roofs of coal beds is far more common than in the partings of the coal.
The generally local occurrence and variable character of the partings may be
construed as showing that the inroads of the sea were restricted. In this connec-
tion it may be noted that the No. 6 coal of Illinois, which extends with unusual
regularity of thickness and bedding structure over several counties in the southern
part of the State, carries a remarkable persistent argillaceous or slightly sandy
and often 'pyritic' parting known as the 'blue band,' about 0.75 inch to 3 inches
thick, over nearly the whole area.

"The usually rather high ash-content of the coal of the eastern Interior basin,
the generally high percentage of sulphur, and the deposition of silica in the
organic mass, possibly as the result of precipitation of colloids by the overflow of
alkaline waters, appear to the writer to be more or less directly associated with the
conception of relative freedom of access of sea or brackish waters, and thus to
accord with the occurrence and character of the black carbonaceous roof muds
with their marine invertebrate contents which mark the permanent inundation
of the peat deposit by the sea. It is not impossible that in these regions of
most intimate relation of the level of the peat formation to tide-level, some
incursions of salt water, probably local in extent, may have taken place, and that
the effects of such immersions are causally connected with the structural char-
acters as well as the chemical qualities of the fuel.

" It is, of course, well known that in many regions peat is in process of forma-
tion in salt marshes, that is, in coastal-inlet or estuarine swamps covered by salt
water at high tide. It does not, however, seem clear to the writer that coal with
so large a percentage of mother of coal, jetlike wood, etc., and with such pure
carbonaceous matter, that is, containing so moderate a percentage of ash, as the
coal in the Carboniferous of Illinois and Indiana, or that interbedded with marine
or brackish-water beds in Wyoming, was laid down in estuaries flooded by sea-
water at high tides or even at times of ordinary storms. Unfortunately, adequate
analyses for the determination of the ash in tidal-marsh peats from many localities
seem to be wanting. In this connection it must, however, be remembered that
the coal-forming jungle itself, by its great fecundity and profusion of growth,
and its correspondingly rapid contribution of refuse and fallen trees, may have
constituted an effective barrier into which the salt-water overwash of extra-
ordinary tides or storms would be able to penetrate only a relatively short distance
(perhaps a few miles) compared with the great area, often scores of miles in extent,
of the arborescent vegetal growth. Such an almost impenetrable vegetal barrier
would not only retard or dam the inrush of salt water pending its neutralization,
but also would quickly arrest the sediments and, according to the force of the rush,
retain them near the periphery of the swamps. It is not improbable that the
higher ash-content that is apt to characterize coal beds immediately overlain or
interbedded with marine sediments may result from occasional invasions of
brackish water into the swamps, the consequent death and decay of the fresh-
water types and, for short intervals, the deposition of peaty sediments higher in
ash, as is characteristic of brackish or salt-water peats at the present day.

"Within the mouths of the estuaries and outlets of the lagoons, as well as
at the border of some of the swamps, there must have been transition zones in
which the water was at times more or less saline, but just which of the Paleozoic
plant types served in these frontier positions, maintaining their stand on the
border (brackish) zone of a salt-water habitat, is not at present fully known,

INTERPRETATION OF ENVIRONMENTAL CONDITIONS 199

though it is very probable that some of the types of vegetation were adapted to
this habitat. On the other hand, the absence of marine shells from the floor
of the coal bed, and the anatomical characters of the fossil plants and condition
of life, at the present time, of the descendants and relatives of the coal-forming
plant types, show almost conclusively that they were not adapted to live in a
habitat of salt-water submergence. Professor Weiss' points out that of the
lixang pteridophytes, only a single fern grows where it is subject to marine expo-
sure. On the other hand, he calls attention to the fact that the fungi found in
fossil Lepidodendron wood and the parasites discovered on the Stigmariae are of
fresh-water t>'pes, as are also the insect orders to which the eggs found fossil in
the bored woods belong. The large size of some of the Calatnites, whose trunks
sometimes attain a diameter of nearly a foot and a height of 30 feet or more, with
the strength and rigidity afforded by thick developments of exogenous wood,
should have enabled them, when their bases were embedded in the muds, to offer
a relatively strong resistance to such wave-action as may have occasionally
been encountered in regions only slightly exposed, but it is not probable that the
ancient relatives of our strictly fresh-water horsetail family were able to grow
in soil affected by salt water. The large trunks of SigiUaria and Lepidodendron
may have offered an effective reinforcement to the calamarian tyf>es. Most,
possibly all, of the fragments of Carboniferous plants found so rarely in actually
marine deposits may well have been brought by drift from terrestrial or fresh-
water habitats.' On the other hand, it remains most highly probable that the
common types of coal of all ages were laid down in fresh or nearly fresh water."

The beds of the Allegheny series show frequent changes of material due
to minor but repeated and rapid fluctuations of level, but, as shown by the
quotation from David Wliite just given, they maintain on the whole a
homogeneity' which speaks of wdespread and long-continued uniformity
in the general aspect of the land and water. The effect of such an environ-
ment upon the vertebrate fauna has been discussed by the author in an
account of the amphibian fauna of Linton, Ohio.' A small portion of that
paper may be repeated here:

"The Cldiatic Environment of Fauna.

"The flora of the region around Linton has been reported uf>on by David WTiite.
His list* of the plants of the Freeport group contains no forms differing especially
from those of the whole Allegheny series, and aill indicate the existence of a
'singularly equable and humid but not tropical or even semitropical climate.'
There is no e\-idence either in the woody growth, foliage, florescence, or fruition
of any seasonal changes, either of temperature or of humidity. In other words,
the amimals lived in a period characterized by the extreme monotony of the
climatic environment.

"The Organic Environiient.

"The organic environment of any animal or group of animads may be defined
as the group of contacts of that animal with other forms of life. Normally, the

* Weiss, F. E., Address to the Botanical Section, British Association for the Advancement

of Science, Science, vol. 34, 191 1, p. 476.

* White, Da\-id, X'alue of Floral Evidence in Marine Strata as Indicative of Nearness of

Shores, Bull. Geol. Soc. Amer., vol. 22, 191 1, p. 221.

* Case, E. C, The Environment of the Amphibian Fauna at Linton, Ohio, Amer. Jour. Sci.,

vol. 44, pp. 124-136, 1917.

* White, David, Bull. Geol. Soc. Amer., vol. 1, p. 154, 1900.

200 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

organic environment comprises both the flora and fauna, but in this instance the
animals were not, so far as we can see, influenced by the vegetation more than
that they profited by the shade of the umbrageous growths, sought refuge in the
interstices of submerged roots, or possibly fed upon some forms of the algae in the
pool. None of these factors would have left any readable record in the morphol-
ogy of the animals. The list of flora occurring in the shales accompanying the
coals of the Freeport group has been cited above.

"The Character of the Contacts within the Fauna.

"The list of known amphibians from Linton as given by Moodie' includes
51 species. The genera are as follows:

Brachydectes. Erpetosaurus. Molgophis. Ptyonius.

Cercariomorphus. Eurythorax. Odonterpeton. Saurerpeton.

Cocytinus. Hyphasma. (Estocephalus. Sauropleura.

Ctenerpetom. Ichthycanthus. Pelion. Stegops.

Diceratosaurus. Leptophractus. Phlegethontia. Thyrsidium.

Eoserpeton. Macrerpeton. Pleuroptyx. Tuditanus.

"If we examine the animals as described and illustrated in Moodie's excellent
monograph, we find that they were, one and all, provided with sharp, conical
teeth, suitable only for a carnivorous or an insectivorous diet. This eliminates
the vegetation of the period from consideration as a possible source, at least as
an immediate source, of food, but introduces a most effective element of stress
in the competition between the animals themselves, on the one hand to capture
prey and on the other to escape the attack of predatory forms.

"The possible sources of food were fishes, the amphibia, and very probably
the abundant arthropods, molluscs, and insects, though practically no traces of
invertebrates have been found with the remains of the amphibians, except the
casts of spirorbis-like forms. While there can be little doubt that some of the
amphibians were carrion-eaters and scavengers, the ultimate food-supply must
have been the invertebrate fauna of the waters and banks, and the very meager-
ness of the remains of such a fauna speaks eloquently of the crowded habitat
and the eager search for every edible particle. Beyond this the diet was of flesh
and the fauna was self-devouring.

"From the description given it seems fairly certain that the amphibian fauna
was isolated in a pool of clear water surrounded by a great stretch of swamp.
The ordinary factors of environment which influenced the development of a
fauna were absent or ineffective, the physiography and the climate were monot-
onous in the extreme; the vegetation had only an indirect effect. The main stress
upon the life was competition within the fauna. This stress became very high
with the crowding of the pool, but as the monotonous environment afforded
but limited possibilities for the formation of new habits, adoption of new habitats
or the assumption of a new group of contacts in any form, it was not relieved by
any overspecialization either in structure or habit. A study of the amphibia
reveals only a very normal group of animals. They varied in size from 10 feet
to 6 inches in length, some were squat and sluggish, others lithe and serpentiform,
some even so snake-like that they had lost their limbs. Some hid for safety in
dark holes and corners, others lurked in the slime, feeding on carrion or the less-
active and well-protected forms; still others flashed through the water in active
pursuit of prey and dared give battle in their conscious strength. It was a

* Moodie, Roy L., Carnegie Inst. Wash. Pub. 238, p. 18, 1916.

INTERPRETATION OF ENVIRONMENTAL CONDITIONS 201

fauna whose elements occupied all the possibilities of the pool to preserve their
lives and propagate their Idnd, but there is an almost total lack of bizarre and
overspecialized forms, none heavily armored and none vnth an excessive develop-
ment of tusk or talon or spine, and none that could be called giants of their
kind. There was a full occupation of all the reasonable possibilities of life, but
nothing that would indicate an extreme adaptation, either for offense or defense,
to limited paths of life such as occur in other places and in other geological forma-
tions where the members of the faunas were very perfectly adjusted to each other.
There was only the healthy growth induced by competition in a fauna which still
retained all the resilience of its juvenile stage.

"Such an assemblage existing under very powerful stress, if even from a single
source, was full of possibilities of development; rip>e for the rapid and wide radia-
tion in habits emd structures long denied them by the monotony of their environ-
ment. For the animals in such a pool there were but two f)ossible endings.
Either the pool would become choked by the growang vegetation of the surround-
ing swamp, or in the many fluctuations of the land channels would open whereby
the animals could escape into other habitats and encounter a new environment.
It was apparently the first of these fates which came to the Linton fauna. It
was overcome in its full vigor before the ultimate adjustments of life to life had
produced the extreme development of armor and weapons of attack seen in more
mature or in senile faunas. Elsewhere in the same region similar faunas were
released to expend in morphological advances and various adaptations to new
conditions the stored-up stresses of similar periods of isolation."

Studies similar to the one made upon the Linton fauna were made upon
the faunae found at Mazon Creek, Illinois, and at the Joggins quarries in
Nova Scotia. While they resulted in similar general conclusions, it was
impossible to determine the limits of the habitat and the life conditions so
closely. It is sufficient to say that from these three localities have come the
great majority of the known amphibian remains from the Pennsylvanian,
and all bear witness to the monotony of the environment and the accumu-
lating stresses which only awaited the great change of environment which
came with the advent of the red-bed conditions to burst into the great
radiation of reptilian and amphibian life of the late Paleozoic.

In the repxjrt by Stevenson' on the Carboniferous beds of the Appalachian
Basin we have a summary of the conditions during the upper Paleozoic from
which parts are quoted :

"The Allegheny is a thin formation, but its variations in thickness are con-
siderable. * * * There seems to be no reason for supposing that the Allegheny
becomes thicker southward in Kentucky, and at present there is little ground
for supposing that it ever reached much farther south than northern Tennessee.

"The sandstones of the Allegheny contrast greatly with those of the Rock-
castle, even with those of the Beaver. They are persistent only as narrow bands,
and in any given area are apt to be replaced for considerable distances by sandy or
even clayey shale. Along the eastern outcrop from Kentucky northeastwardly
into Randolph and Upshur counties of West Virginia the sandstones are very

• Stevenson, J. J., Carboniferous of the Appalachian Basin, Bull. Geol. Soc Amer., vol. 18,
p. 150, 1907.

202 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

conspicuous, very coarse, and at times for miles almost continuous from bottom
to top of the formation, composing in part Mr. Campbell's Charleston sandstone.
Farther north, in the Potomac area, the sandstones are differentiated, broken by
beds of shale; yet even there the Butler and Clarion are massive, the former at
times pebbly. The sandstones are irregular in Broad Top and the pebbles are
few. Within western Pennsylvania the Butler and Freeport sandstones appear
to be most nearly persistent, and each of them occasionally shows some pebbles;
but they vary greatly in thickness and each of them is often replaced by shale
in tracts containing hundreds of square miles. Well records in the deep portion
of Ohio and West Virginia usually show more or less sandstone in one or more
of the intervals, but many show so little aside from shale that the sandstone must
be due merely to local sorting of material. Pebbles are repoited only from Wirt
County of West Virginia. The great sandstones of the eastern outcrop in West
Virginia break within a few miles toward the northwest; thin shales appear,
which soon increase in thickness, and the sandstones become unimportant.
Along the western outcrop in Ohio, sandstone is most nearly persistent in the
Butler and Freeport intervals. Ordinarily fine in grain, the latter shows pebbly
streaks in Stark, Carroll, Harrison, Wayne, Tuscarawas, and Muskingum Coun-
ties— that is, along the northwestern side; yet in all of these counties not a few
sections show only shale. The Clarion (Hecla) sandstone becomes very con-
spicuous in southern Ohio and is equally so farther south and southwest in
Kentucky. It is noteworthy that a conglomerate is present in parts of Kentucky,
near the horizon of the Vanport limestone, and that at one locality the ore
associated with that limestone is so crowded with quartz pebbles as to be worth-
less.

"The character and distribution of the sandstones show sufficiently a great
advance of the shore-line or a considerable elevation of land at the southwest.
The former condition seems the more probable, and the Allegheny deposits can
have extended hardly so far in that direction as did those of the Beaver. The
shore-line at the east-southeast must have been at only a short distance from the
present outcrop, as the strip of sandstone is very narrow. Coarse material
could be pushed only a little way in the shallow water of that time. There is
much to suggest a similar advance of the shore at the northwest, not only in the
unexpected coarseness of the sandstone, but also in distribution of the limestones.
The presence and great predominance of sandstone in Kentucky, on the southern
and southwestern borders, is equally suggestive of land encroachment in that
direction."

The Putnam Hill limestone of Ohio is said by Stevenson to be very
fossiliferous and to mark an invasion of the sea from the west ; similarly the
Black Flint on the Kanawha River, at the same horizon, is said to mark an
invasion from the Atlantic in the form of a branching bay.

"The Vanport (ferriferous) limestone marks a still greater inroad of the
interior, or Mississippian, sea, reaching in northwest Pennsylvania almost to the
New York line. * * * On the Kanawha, in West Virginia, Professor W. B. Rogers
found a bed crowded with marine forms at 140 feet above the black flint, too
high for the Vanport horizon, but of interest as proving access to the Atlantic
at more than one time during the Allegheny. * * * No later important inroad
of the sea occurred. The fossiliferous shale over lying the middle Kittanning is
found only as far north as central Ohio, while the lower and upper Freeport lime-

INTERPRETATION OF ENVIRONMENTAL CONDITIONS 203

Stones, though extending over a great part of the basin at the north, are either non-
fossiliferous or contain only fresh- water forms; but south from the Ohio River in
Kentucky the upper Freeport limestone carries a characteristic Carboniferous
fauna.

"It is wholly probable that the Appalachian and the Indiana-Illinois fields
were not united during the Allegheny, though they may have been during the
Rockcastle, as they were during the Mississippian."

C. INTERPRETATION OF CONDITIONS IN CONEMAUGH
AND DUNKARD TIME.

Mr. Stevenson here (page 154) notes the possible occurrence of persis-
tently aggrading streams from the Homewood sandstone stage to the, in places,
base of the Lower Freeport :

"Toward the close of the Allegheny a small area in west central West Virginia
near the Ohio River received deposits of red mud, more or less calcareous, accom-
panied often by greenish muds; and, somewhat earlier, similar deposits were
made in northeastern Kentucky. This is the beginning of a condition which in
gradually enlarging or contracting area was to continue until the close of Czu-bon-
iferous time, always predominating, however, within a small area in West Virginia
and the adjoining part of Ohio.

"In many respects the Conemaugh is but the continuation of the Allegheny;
the variations in thickness are, geographically, very similar in both. * * * Except
in a very narrow strip along the southeasterly border in West Virginia, the Cone-
maugh sandstones are more irregular than are those of the Allegheny. One
generally finds some sandstone of some sort in the sandstone inter\^als, but shales
predominate in by far the greater part of the area. * * * Away from the south-
eastern border, pebbles are extremely rare, except along a narrow rudely east-
and-west strip across Indiana, Armstrong, Butler, LawTence, and Beaver counties
of Pennsylvania. This lies many miles south from the northern outcrop and
south from the similar strip in the Beaver formation; its variations are such as
one finds in the gravels of the upper Ohio River. Many similar valleys filled
with standstone during the long subsidence are recognizable in various parts of
the area, and occasionally one is found along an anticlinal crest which seems to
have been made by subaerial erosion. The sandstones for the most part are
indefinite within Ohio, but in Tuscarawas County the Lower Mahoning interval
filled with conglomerate and farther south the Buffcilo interval is filled with very
coarse sandstone at many places. * * *

"While in a general way the conditions were similar to those of the All^heny,
showing a gradually contracting area, yet the subsidence was such as to admit
sea-water to a much greater space. At the very beginning one finds at somewhat
widely separated localities in West Virginia a marine fauna in the Uffington shale
which rests directly on the Upper Freeport coal bed, while at most exposures the
shale y-ields only impressions of land plants. Not enough information is avail-
able to justify any suggestion respecting the relations of the marine localities,
which are confined to the easterly side of the great basin." [It is here suggested
by Mr. Stevenson that the Brush Creek limestone is due to an invasion from the
east, while the Cambridge limestone is due to an invasion by the Mississippian
sea from the west.]

204 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

"With the Ames limestone, inroads of the sea practically ceased. Marine
conditions unquestionably were repeated, but never for periods long enough for
good development of animal invertebrate life. Limestones appear frequently
during the upper half of Conemaugh, several of them widely, though irregularly,
distributed, but in no case are they distinctly marine. Some are crowded with
minute univalves of undetermined relations; others are associated with carbon-
aceous shales, filled with fragments of plants and fishes, which point rather to
fresh-water conditions.

"The most notable feature of the Conemaugh is the red and green shales, in
color resembling those of the Catskill and Shenango, but deeper. The greater
development is in west central West Virginia and the adjacent part of Ohio,
where at times nearly the whole section is red shale. The greatest geographical
expansion was just preceding the deposition of the Ames limestone, when the reds
reached southeast nearly to the outcrop and northward to the outcrop in Penn-
sylvania; but they did not reach into northern Ohio and they are practically
wanting east from the line of Chestnut Hill in Pennsylvania. From that time
to the end of Conemaugh the area contracted and reds occur in irregular patches.
These beds frequently contain nodules of limestone, and are usually fossiliferous.
The red shales in some cases mark horizons elsewhere carrying limestone, and
they may indicate a marine condition.

"The exceeding shallowness of the water and the long periods of quiet during
the Conemaugh are indicated by the coal beds, which, though extremely thin,
have great extent. * * *

"Toward the close of the Conemaugh the streams bringing in materials had
become sluggish, and the deposits, except within limited areas, are fine in grain.
The Monongahela began with a long period of exceedingly slow subsidence,
during which the Pittsburg coal bed gradually extended across the northern part
of the great basin and southward along the east and west sides; but from all
sides it became thinner toward the central part of the basin and it is practically
wanting in a great part of West Virginia and eastern Ohio, where it occurs only
in widely separated patches. The bed may have been almost continuous around
the basin. The singular uniformity of conditions and the extreme slowness of
movement are shown by the structure of this great bed, persisting in such minute
details as partings in tracts of thousands of miles and reappearing even in isolated
patches within West Virginia.

"The area of greatest subsidence during the Monongahela did not coincide
with that of the earlier formations, as appears abundantly from comparison of
sections along several lines. The deepest deposits of Allegheny and Conemaugh
were at the north and east; not so in the Monongahela. * * * The greatest
subsidence was in north-central West Virginia, whence the thickness decreases in
all directions.

"With this change in place of chief subsidence there came clearly a farther
contraction of the basin, while elevation at the north led to spreading out of
sandstone along much of the northern border. This Pittsburgh sandstone is not
present in the eastern localities of Pennsylvania and Maryland, but is persistent
in the Chestnut Ridge area of Fayette and Westmoreland Counties, in that State,
as well as southward along the eastern outcrop in West Virginia to the last
exposure near Charleston, where Doctor White found it 7o feet thick. Evidently
it prevailed along the western outcrop in Ohio, for it is present on the north-
western outcrop and also in the central counties along that line, where one is

INTERPRETATION OF ENVIRONMENTAL CONDITIONS 205

again much farther west than in the intervening counties. This sandstone
becomes more and more indefinite from all sides toward the interior of the basin.
The Sewickley sandstone, underlying the Upper Sewickley coal bed, is fairly
persistent on the east side, but is wholly insignificant in Ohio. There, however,
an important sandstone overlies the upper Sewickley, not pebbly at the north-
west, but coarse and often pebbly in southern Ohio. In Pennsylvania and
northern Ohio a more or less persistent sandstone, the Uniontown, overlies the
Uniontown coal bed, but ordinarily it is unimportant and many sections show
little aside from shale in the interv'al. In West Virginia, however, a strip of
coarse conglomerate, evidently at this horizon, crosses the State from east to
west, passing through Lewis, Gilmer, Doddridge, Tyler, and Pleasants Counties
and extending into Washington, Morgan, euid Athens of Ohio, where it is the 200-
foot conglomerate of Professor Andrews. It is coarser in West Virginia than in
Ohio. The strip is very narrow in the former State and fine-grained rocks
replace the coarse material at a short distance north and south; but in Ohio the
area is broader, as though additional material had been brought in from that side.
This east-and-west line of coarse rock recalls those of the Beaver and Conemaugh
in Pennsylvania and may be explained in the same way. The general distribu-
tion of coarse material indicates a rising borderland and for the southwest a
notable encroachment.

"The limestone [varies] greatly in composition. The Redstone is an impure
limestone, yielding a fair lime when burned carefully; the FishpxDt, when thin,
usually resembles the Redstone, but when thick it is apt to contain some layers
of cement rock; the Benwood has several beds of hydraulic limestone, even of
cement rock, among its most persistent members, while some of the beds are so
impure as to break into small angular fragments after continued exposure; the
Uniontown and Waynesburg are rarely more than slightly magnesian.

"Of the numerous limestones, only the Uniontown can be regarded as really
persistent; it is present in western Pennsylvania and in Ohio at nearly every
locality where its place is shown. The others may be r^arded as confined to
southwest Pennsylvania, the West Virginia Panhandle, and the immediately
adjacent part of Ohio. Their great development is between the Monongahela
River at the east and the Ohio River at the west, where in considerable areas
limestone and calcareous shale fill more than one-half of the interval between the
Redstone and Uniontown coal beds. In all directions from this small area the
limestone diminishes quickly and is replaced by shale and sandstone; toward the
southwest only some thin streaks remain in West Virginia, and in some portions
of that State those streaks seem to be replaced by red shale.

"These limestones are spoken of commonly as merely calcareous muds, and
that explanation of their origin was accepted tentatively on a preceding page.
But it is insufficient." [Mr. Stevenson regards the origin of the limestones as an
unsolved problem.]

"Toward the close of the Monongahela the condition marking the later portion
of the Conemaugh was reached once more. In by far the greater part of the
area the depxDsits are fine in grain, and at the end the Waynesburg coal bed was
formed, in the northern part of the basin, a bed of curiously multiple structure,
which is retained. Like the Pittsburgh, it is wanting in the interior region, but
it seems to have reached irregularly southward to a long distance on each side.

"The Washington opens with a plcuit-bearing shale like that overlying the
Pittsburgh, succeeded by a great sandstone, recalling in some respects the sand-

206 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

stones of the Rockcastle. As the area grows smaller in ascending, it becomes
necessary for comparison to consider separately the lower and the upper portion
of the Washington. * * * The formation thus increases from i6o feet in northern
Washington of Pennsylvania to above 480 feet in the northern counties of West
Virginia, thus showing a continuance of the Monongahela conditions, with the
greatest subsidence in north-central West Virginia.

"The sandstones tell the story of steadily contracting area. The Waynesburg
sandstone is persistent in Maryland, in most of Pennsylvania, as well as southward
in West Virginia for a long distance. It is massive and at times pebbly, though,
like all sandstones of the higher formations, it is sometimes replaced abruptly
by shale. In Ohio, along the northwestern border, it is not a coarse sandstone,
but farther south it becomes coarser and more prominent, being Professor
Andrews's upper sandstone and conglomerate. Thence southeastwardly along
the southern border, in Jackson and Putnam of West Virginia, the rock marking
this horizon is a coarse sandstone, with quartz pebbles sometimes an inch in
diameter. In the interior portion of West Virginia records of oil borings show
sandstone persistent in this interval, except in a small area. The Waynesburg
is the first sandstone of wide extent in the interior region. No notable sandstone
above the Waynesburg appears in Pennsylvania, except that underlying the Upper
Washington limestone, which is confined to the borders of the remaining area
and disappears southwardly. Below this one finds local sandstones, but they are
unimportant. In the southern portion of the basin, on the contrary, the interval
above the Washington coal bed is characterized by great sandstones, the Marietta
of Doctor White, which appear in their greatest development toward the south-
west outcrop, though they are prominent features across West Virginia, extending
northward to midway in the State.

"The limestones of the Washington are quite as perplexing as those of the
Monongahela and they are confined to a smaller area. * * * The limestones of
the Washington bear much more resemblance to calcareous muds than do those of
the Monongahela, but it is difficult to discover the source whence they were derived.
[Mr. Stevenson is inclined to think that these limestones are not marine in origin.]

"During the Washington the crustal movements were sluggish within the
basin of deposition. Thin streaks of coal extend over great areas, many of them
showing complex structure; but toward the close the movements became more
pronounced, and during the early portion of the Greene the deep portion of the
basin was confined to Greene County of Pennsylvania and a narrow strip adjoining
at the west in West Virginia. * * *

"That the area of deposit was contracting rapidly appears also from the
sandstone deposits. * * * All of these [the Nineveh, Fishcreek, and Gilmore
sandstones] are along the middle line of the basin, where during deposition of all
formations prior to the Washington the sandstone intervals were usually filled
with shale. The sources of supply were much nearer than in earlier periods.
But the basin, though rapidly losing in width, still extended for not less than 200
miles in north-northccist to south-southwest direction when the Nineveh limestone
was laid down.

"The limestones, except the Nineveh, are of little importance. * * * There
is no evidence that the sea actually entered the area in which rocks of the Greene
formation remain.

"The Red Beds retained their importance apparently to the end within the
half dozen interior counties of West Virginia and Ohio, and twice during the

INTERPRETATION OF ENVIRONMENTAL CONDITIONS 207

Monongahela the area showed a very considerable expansion, though in neither
case equaling that of the Washington or lower reds of the Conemaugh and in
each very much less than that of Pittsburgh reds of the same formation. After
the deposition of the Uniontown coal bed their area diminished, and during the
Washington and Greene the reds became less and less important, appearing at
least in, for the most part, thin and rather widely separated deposits, though
occasionally, as in Marshall of West Virginia and northern Greene of Penn-
sylvania, they attain considerable local importance."

D. INTERPRETATION OF CONDITIONS IN THE WESTERN
PART OF THE EASTERN PROVINCE.

In eastern Kentucky there are some red and purple shales at the level,
approximatetly, of the upper Conemaugh and the lower Monongahela.
These are probably the leist traces of the more important red beds in West
Virginia and Ohio. In western Kentucky, Illinois, and Indiana the red
beds do not appear, except for a single local patch just above the coal vn
in Illinois. The break in the series of Pennsylvanian deposits caused by the
Cincinnati dome makes the exact correlation of the beds in the two parts of
the Southern Subprovince impossible, but, as shown in the correlation table,
page 48, the deposits in Illinois and Indiana above the probable Mononga-
hela-Dunkard line are unchanged in character from those below. It is
apparent that the uppermost beds preserved were deposited under condi-
tions quite similar to those which prevailed in Pennsylvania and West
Virginia during Allegheny and lower Conemaugh time. The advancing
climatic change had not reached as far west as these localities when the beds
were deposited, although they are at a much higher stratigraphic level
than the middle Conemaugh.

The meaning of the vertebrate-bearing bed near Danville, Illinois, and
the river-channel sandstones of Merom, Indiana, have been discussed by the
author in Publication 207 of the Carnegie Institution, pages 77 to 80; the
evidence shown in this paper of the gradual rise of the land from the east
and the migration of the environment toward the west strengthens the sug-
gestion made long ago that the Permo-Carboniferous vertebrates found near
Danville were embedded in the clays of an excavation of Permo-Carbonifer-
ous time in the exposed deposits of Pennsylvanian age. As an alternate
hypothesis the absence of the characteristic red deposits of Permo-Carboni-
ferous conditions in this region may well be explained b}'' the lack of any
adjacent high land from which such deposits could have been derived and
the fact that in this lower land the increase of aridity and lowering of
temperature were not sufficiently pronounced to destroy an abundant vege-
tation the debris from which would have been deposited wdth the bodies of
the animals whose bones are preserved and have reduced any ferric oxide to
the ferrous condition, with consequent loss of red color. The extensive

208 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

erosion which must have gone on in this region would have removed a very
considerable amount of material, and it is not surprising that any traces of
red sediments have been removed and only the deeper material accumulated
below the level of ground-water left.

E. INTERPRETATION OF CONDITIONS IN THE PLAINS PROVINCE.

The condition of the upland in Missouri, which separated the Eastern
and Plains Provinces, has been well pictured in the quotation given on
page 82, from Hinds and Greene. The red deposits of the Plain sProvince
on its western side are undoubtedly due to climatic conditions very similar
to those which prevailed in the eastern province, but their base is at a
much higher stratigraphic level than middle Conemaugh. The migration
of the climatic change due to the gradual uplift did not make itself felt in
the Plains Province until the beginning of Permo-Carboniferous time, well
above the Missourian. The difference in depositional conditions in the
upper half of the Pennsylvanian in the eastern and western half of the
United States has long been recognized. Marine conditions prevailed much
longer in the West than in the East and such great swamp areas as char-
acterized the eastern basins were never developed. A glance at the typical
Kansas section of the western beds shows the successive series of limestones
and shales which occupy the same intervals as the shales and coals of the
eastern region. In the East the elevation affected a land of swamps and
the climatic change superimposed the deposits of semiarid glacial or subglacial
condition upon one of singular equable conditions of temperature and
humidity; in the west the elevation obliterated or reduced great areas of
epeiric seas and the semiarid and cool conditions came upon the surface and
the flats of a rapidly disappearing sea. It is easy to understand why the
shales and restricted coals of the western area have yielded so few remains of
vertebrate amphibian life in comparison with those of the eastern area.

The source of the material of the red beds of the Plains Province was
largely the highlands of the Missouri region, the Ouachita Uplift in Okla-
homa, and the Rocky Mountain axis to the west. Schuchert is inclined
to think that there must have been a very considerable amount derived
from the great Columbia Positive Element of the extreme southwestern
part of the United States and the adjacent portion of Mexico. The northern
red beds of the Plains Province undoubtedly received a considerable part
of their material from the Rocky Mountain axis, but it is possible that some
portion came from the northern end of the land exposed in Missouri-Iowa.
This suggestion is borne out by the presence of the red deposits near Fort
Dodge, Iowa, which, though lower than the typical red beds of the south-
west, are at the upper surface of the Missourian. The greater part of the
northern portion of the Plains Province is buried under younger deposits
and only by inference can the whole of the region be restored. Something

INTERPRETATION OF ENVIRONMENTAL CONDITIONS 209

of this has been attempted by the author in Publication 207 of the Carnegie
Institution, page 62.

It can scarcely be possible that the highlands mentioned failed to share
in the general elevation of the continent during Pennsylvanian and Permo-
Carboniferous times, and though there is no evidence that at any place they
reached a height or condition sufficient to produce even incipient glaciation,
still the rise was enough to afford an abundance of deposits under red-beds
conditions. Indeed, the source of supply seems inadequate to furnish the
amount of material that has been accumulated, and this is probably why
Schuchert has suggested so remote a source of supply as Columbia.

As has been said, the red beds of the northern part of the Plains Province
are so largely buried that the outer borders are not exposed and it is im-
possible to say how far they extend from the edge of the Rocky Mountain
axis, but that the distance was not excessive is indicated by the discovery in
Wyoming of red beds shading into dark-colored shales and limestones to
the north and east in Wyoming (Carnegie Inst. Wash. Pub. No. 207, p. 62).
WTiat the relation to the land on the eastern side of the northern portion
of the Plains Province may have been we have no means of knowing.

In the southern portion of the Plains Province it is evident from the
data furnished in the summar>' description that the red beds material came
largely from the south and west and that a considerable body of water lay
to the southeast which occasionally spread north and west over portions
of north-central Texas for limited periods; the marine conditions seem to
have lingered longest in Kansas and perhaps Nebraska.

The origin and source of the red sediments have been the cause of con-
siderable discussion. Two writers in particular have drawn pictures of the
Texas-Oklahoma portion of the province which are worth consideration
at this point. Beede,^ in discussing the origin and color of the deposits
in Oklahoma, says:

"In tracing the limestones and shales of the basal Permian beds of Kansas
southward into Oklahoma the relationship of the light-colored sediments to the
red sandstones, red shales, and red limestones of Oklahoma is clearly revealed.
It is shown that some of the heavier ledges of limestone first become sandy
along their outcrops in patches a few rods across. Farther south the sandstone
areas increase in size until the limestone appears only in local areas in the sand-
stones and is finally wanting. Traced farther southward, the sandstones become
deep red or brown, with local areas of white. The decimation of the fauna sets
in as the limestones diminish and the remains of life are not found far beyond the
limits of the limestones. The shales become red very much farther north than
do the sandstones, and are frequentiy moie deeply colored. Some of the lower
limestones become red before they change into sandstones. The sandstone ledges
continue for some distance southward as rather even, uniform beds, but farther
on they are found to thicken and thin in a somewhat systematic manner.

• Beede, J. W., Origin of the Sediments and Coloring Matter of the Red Beds of Oklahoma,
Science, vol. xxxv, pp. 348-350, 1912.

210 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

"Several ledges of sandstone frequently occur in a single section, and where
one of these ledges is found thickened the others are apt to be thicker than normal.
Likewise they are all found to be thin over certain areas. The regions of thicken-
ing and thinning were found to be parallel belts lying north and south at right
angles to the major drainage-lines. Two of these belts, together with an inter-
vening region about 8 miles across, were studied. The sandstones thicken at the
expense of the shales, sometimes eliminating them. In one instance a thin lime-
stone was traced southwest into one of these zones. A sandstone 20 feet or more
beneath the limestone thickens and rises above the limestone and practically
unites with the sandstone some distance above it. The limestone seems to die
out a few feet from the sandstone, but farther west the latter shrinks to its
normal thickness and the limestone is present in its proper position with its usual
characteristics.

"In these zones of thickening, which are frequently several miles wide, the
sandstones are very irregularly cross-bedded and frequently ripple-marked, while
the thickening is uneven. It would seem that these zones are opposite the mouths
of streams which brought sediment into the sea, where the coarser materials
were carried farther from the shore than opposite the interstream spaces. The
irregular thickening of the individual beds may be due to current work, wave-
action and heaping into local dunes by the wind, though the action of the last
factor is uncertain. The irregular bedding and ripple-marks indicate a sort of
littoral or very shoal condition for the deposition of the sandstones and shales.

"As this interesting transition of sediments is traced still farther southward,
we find, before reaching the latitude of Shawnee, that the sandstones become
more abundant over the whole area, more lenticular, more irregularly cross-
bedded, and imperfectly lithified. In a single railroad cutting a thick lens of
sandstone may fade into a soft sandy clay shale with the same bedding and
structure as the stone itself and change back into a sandstone a few rods away.
Most of the sandstones are so incoherent when freshly quarried that pieces 2 or
3 inches in diameter crush readily under foot. In many of the wells of the region
the water is obtained in 'quicksand.' Most of the shales contain much fine
sand and offer little resistance to weathering.

"At their southern limits these red sandstones and shales are found to dovetail
into the Permian conglomerates on the southern side of the Arbuckle Mountains,
while similar conditions obtain among the higher beds farther west, where similar
conglomerates occur on the flanks of the Wichita Mountains. These conglomer-
ates are largely composed of the fragments of the pre-Ciarboniferous limestones
aggregating 8,000 to 10,000 feet in thickness flanking the mountains and at one
time covering them. The solution of these limestones produces a red clay wher-
ever the insoluble residue happens to remain undisturbed below the vegetable
mold, and the disintegrating limestone conglomerates produce a more or less
sandy clay indistinguishable from some of the red sediments. Thus it seems
not improbable that much of the material of the red beds in the region studied
was derived from these thick limestones.

"Considering all these phenomena, it is apparent that the transition of
deposits from the Arbuckle Mountains to the Kansas line is such as would be
expected in passing from the mountains out into a shallow epicontinental sea.

"That the solution of limestone produces red residual clays is well known.
It is exhibited in the residual soils and clays of the limestone regions of the
unglaciated part of the Mississippi Valley, Cuba, southern Europe, and elsewhere.

INTERPRETATION OF ENVIRONMENTAL CONDITIONS 211

The clays thus derived and their coloring matter — the red oxides of iron — are
minutely divided and when in suspension settle slowly, but little movement of
the water being sufficient to keep them in suspension. This characteristic
adapts them to long transportation. The great thickness of the Arbuckle and
associated limestones, and their former extent, over thousands of square miles of
country where they are now removed or represented only by their upturned
edges surrounding the mountains, seem to furnish an ample source of the coloring
matter and a considerable amount of clays of these low Oklahoma red beds.
The gabbros, red granites, and red porphyries of the Arbuckle-Wichita region
also contributed their share of sediment to the red beds.

"From these observations it would appeau- that the sediments of the lower
red beds of Oklahoma were derived largely from the Arbuckle-Wichita Permian
land-mass and the coloring matter mainly from the solution of the limestones
known to have been removed from it. It also seems probable that the sediments
of the region studied, especially those some distance from the mountains, were
deposited in very shallow turbulent water, or vast tidal beaches, inimical to life
of all kinds, since they are void of fossils or even carbonaceous matter."

Baker,^ writing more particularly of the Texas beds, says:

"In Texas, Oklahoma, and Arkansas, early Pennsylvanian marine sedimenta-
tion was followed by mountain-making movements in the Ouachita Mountains
region of Arkansas and southeastern Oklahoma, and in the Central Mineral
(Llano-Burnet) region and the trans-Pecos country (Marathon, Van Horn, and
El Paso regions) of Texas. The newly-formed mountains were rapidly eroded
and a large part of the mountain region resubmerged beneath the sea in later
Pennsylvanian time. In the western trans-Pecos region a later Pennsylvanian
hmestone nearly a mile in thickness was deposited and in the Marathon region
shales and limestone covered the much-eroded, closely folded earlier Pennsyl-
vanian and early Paleozoic rocks. In north-central Texas later Pennsylvanian
sedimentation began with sandstones, conglomerates, emd shales, and was fol-
lowed by shales and limestone. The land-derived sediments seem to have been
derived from lands to the east and southeast, and for this reason it is believed
that the mountains of the Central Mineral region were then more or less con-
tinuous with tliose of southeastern Oklahoma and west-central Arkansas, and
perhaps stretched westward to the Marathon Mountains. * * *"

"The Pennsylvanian sea of north-central Texas was never very deep and its
waters were seldom free from sand jmd mud brought to it from land areas on
the south and southeast. It was only near the close of the period, and then only
in the southwestern part of the region, that the sea-waters became fairly clear
from land-derived sediments. The coal beds, found in the Strawn and Cisco
formations, were probably formed in regions of coastal swamps, the surfaces of
which lay very close to sea-level. Comparatively rapid oscillations of sea-level
must have sometimes taken place, because we find beds of coal directly overlain
by limestones containing abundant marine fossils.

"We may draw for ourselves a fairly vivid picture of later Pennsylvanian times
in north-central and west Texas. To the westwau^d lay a great sea with clear
waters abundantly teeming with marine animals. On the south and southeast
was the land of mountain ranges which came into existence earlier in the Penn-
syhanian. Between this land and the western sea was a low foreland £md

* Baker, C. L., Origin of Texas Red Beds, University of Texas Bull. 29, 1916.

212 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

shoreline, now submerged beneath a shallow sea, now a marshy land covered by
forests of the strange plants of the coal period,

"Near the end of Pennsylvanian time there was another period of mountain
formation^ in west-central Arkansas and southern Oklahoma. The Cisco forma-
tion of north-central Texas was laid down after this period of mountain-building.
South and southwest of the Arbuckle and Wichita Mountains of southern Okla-
homa the Cisco sediments are red sandstones, conglomerates, and shales, showing
by their structures and vertebrate fossils that they were deposited as land sedi-
ments by rivers flowing southward and southwestward from the mountains on
a flattish plain much like the country along the shore of the present Gulf of Mexico.
Farther to the southwest the Cisco sediments become marine shales and lime-
stones, indicating rather clear sea-waters in that direction.

"The beginning of Permian time in north-central Texas was a continuation,
without marked interruption, of the later Pennsylvanian. The earliest Permian
formation, the Wichita, is very like the Cisco, in the northeast river and shore-
line deposits of red color, and in the southwest marine limestones and clays. In
trans- Pecos Texas the lower Permian is mainly marine limestone with a smaller
amount of shale, and the sediments here have a thickness of about 8,000 feet.
Here again, although the exact relations between the Pennsylvanian and the
Permian are not yet known, it is probable that there was no great change in
conditions between the later Pennsylvanian and the earlier Permian. The
clearer and deeper sea, as before, lay to the westward.

"There was a notable change in later Permian time. The upper Permian
sediments consist of red clays, and beds of limestone, frequently dolomitic,
gypsum, and rock salt. The gypsum and rock salt were deposited from the sub-
stances carried in solution by the sea-water upon the drying up of the sea. The
upper Permian basin of the Southwest centered somewhere beneath the region of
the Llano Estacado. It is very noteworthy that there are no coarse terrigenous
sediments in the upper Permian. The land-derived materials are mainly fine
clays. When sands occur, they are fine-grained.

"We have in upper Permian time the condition of a nearly or quite land-
locked sea gradually shrinking through the drying-up of its waters. The sedi-
ments contributed to this sea were fine clays and sands derived partly from the
red sediments of the Wichita formation and partly from the maturely-weathered
residual soils formed during later Pennsylvanian and earlier Permian times.
Transportation of these flocculent clays and fine sands would not remove the
thin coating of iron oxide by attrition. Even if it did so remove it, the already
highly saline waters would not be likely to dissolve it. And even if they did dis-
solve the coating, the iron oxide would again be deposited before evaporation of
the sea- waters had reached a concentration high enough to deposit the gypsum
and salt. But it is most probable that the iron oxide coating was never removed
by attrition or solution. Fine sediments derived from residual soils are trans-
ported great distances by rivers of the present day without the removal of the
red coating. * * *

" It is not necessary to assume that all red residual soils are formed from the
weathering of limestones. The Tertiary sediments of the southeast Texas Gulf
Coastal Plain are not limestones, yet their residual soils are red. * * *

' Taff, J. A., Preliminary Report on the Geology of the Arbuckle and Wichita Mountains
in Indian Territory and Oklahoma, U. S. Geological Survey, Professional Paper No.
31, 1904.

INTERPRETATION OF EN^^RONMENTAL CONDITIONS 213

"In conclusion, it seems evident that the Texas 'red beds' were originally
maturely decomposed red residual soils formed under warm and moist climatic
conditions. In the older 'red beds' there is no evidence of arid conditions;
in the later Permian 'red beds' the residual soils were transported and deposited
in arid basins without loss of their color. It is probable that the true origin of
all the ' red beds ' in the western interior of North America is from residual soils,
or the erosion and redeposition without change of color, of older 'red beds,'"

Tomlinson* in a study of the origin of red beds concludes that the red
color is original and that the sediments are residual soils derived from
elevations formed at the close of the Paleozoic. Specifically he suggests
the derivation of the Cutler and Dolores sediments of Colorado from the
Uncompahgre Plateau and of the red beds of Arizona and New Mexico
from high lands in Mexico, Arizona, and southern California. The Rustler
limestone and Castile gypsum he suggests may be deposits in the clear
water of inclosed basins. These give place to red beds on the edges of the
basins in shallower water.

For the northern portion of the Plains Province, Richardson* reaches
the same general conclusion as to original color and suggests the origin of
the material from the Rocky Mountains, washed into a sea which covered
the site of the Black Hills. Two possible areas of supply, the Sioux quartzite
area of Algonkian age to the east, and the uplifted Pennsylvanian limestones
to the east and southeast, are considered, but not r^arded as probable
sources. However, these may have supplied much material now deeply
buried or removed by erosion.

F. INTERPRETATION OF CONDITIONS IN THE BASIN PROVINCE.

As is shown in the summary description of the Permo-Carboniferous
deposits of the Basin Province, the red beds are confined very largely to
the southern portion, occurring in northwestern New Mexico, southwestern
and western Colorado, northeastern Arizona, and southeastern and southern
Utah; to the north the equivalent horizon is marked by shales, impure
limestones, and phosphate-bearing beds. Only in Wyoming do red beds of
Permo-Carboniferous age occur in the northern part of the Basin Province
The source of the material in the southern portion seems to have been in the
elevations now forming the southern ends of the Rocky Mountains, for to
the south lay the seas in which were deposited the limestones and shales
of the trans-Pecos region in Texas, marine conditions which were apparently
continued to the west. As noted above (page 152), Schuchert would place
the Kaibab limestone and its equivalents in the Permian or Permo-Car-
boniferous. If this suggestion should finally be accepted, the only eflfect
upon the argument of this w-ork would be to indicate that the climatic
change arrived in the southern portion of the Basin Province at a later
date than is here assumed.

* Tomlinson, C. \V., Origin of Red Beds, Jour. Geol., vol. 24, pp. 153 and 283, 1916.

' Richardson, G. B., Upper Red Beds of the Black Hills, Jour. Geol., vol. 11, p. 365, 1903.

214 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

In the northern portion of the province the deposits must have come
from the adjacent elevations of the Rockies and the Bighorn Mountains.
The conditions which produced red beds in the south must have been much
the same as those which produced the similar beds of the Plains Province
and, as has been shown, the mountain barrier was probably low at the
southern end. The reason for the replacement of red beSs by phosphate-
bearing shales and limestones and of semiarid terrestrial conditions by
marine conditions farther north is less easily understood.

Blackwelder,^ in discussing the deposits of phosphate rock in the northern
portion of the Basin Province and of phosphate rock in general, says, after
reviewing other theories :

"Nevertheless, we find among the rocks derived from oceanic sediments in
many parts of the world, beds several feet thick which are rich in lime-phosphate
and extend rather uniformly over thousands of square miles. They contain
marine fossils which indicate that they have accumulated upon the sea bottom.
It is therefore evident that locally there must be conditions which cause the
fixation of the phosphoric acid among the bottom sediments. Some students of
these deposits have ascribed them to direct deposition of phosphatic shells, bones,
and teeth, and others have made appeal to the agency of mineral springs. Gen-
erally they have sought an explanation for the abundance of the phosphorus.
As the writer has already shown, however, the quantity of phosphorus dissolved
in sea-water is always sufficient to produce in a few thousand years even the
thickest known phosphate beds; and hence we need only to account for the
special conditions which cause it to be precipitated on the sea floor. There is
excellent reason to think that the immediately controlling conditions are chemical
or biochemical, but these chemical conditions in turn depend upon physiographic
and climatic factors difficult to analyze and estimate. The study of the latter
is a task for the geologist.

"In its simpler aspects, the chemistry of the marine deposition of phosphates
has been plausibly interpreted by a number of European students of the question,
even as far back as 1870. The following is a modification of their views, based
on modern information: The process and results of bacterial decomposition of
organic matter vary according to the conditions as well as the particular class of
bacteria that are at work. In air and aerated water, decay is generally complete,
resulting in the production of carbon dioxide, water, soluble nitrates, sulphates,
phosphates, etc. In the absence of oxygen, however, the anaerobic bacteria
somewhat more slowly break down the organic compounds and produce a dif-
ferent series of end-products, of which the most important are various hydro-
carbons, nitrogen, ammonia, and hydrogen sulphide, with only so much of the
carbonic oxides as the available oxygen in combinations permits. In so far as
free oxygen is present in only small quantities, there should be a compromise
between the two processes.

"Some of the most obvious characteristics of our marine phosphatic rocks
show that they have been associated in origin with the anaerobic phase of bacterial
action. Almost invariably they are black in color and, owing to the fact that they
contain noteworthy quantities of hydrocarbon oils, tars, and gases, they are

* Blackwelder, Eliot, The Geologic Role of Phosphorus, Amer. Jour. Sci., vol. XLii, p. 291, 1916.

INTERPRETATION OF ENVIRONMENTAL CONDITIONS 215

famous for their bad odor. In central Wyoming, phosphate rocks of this kind
contain so much oily matter that they are being successfully exploited for
petroleum. Although such phosphates contain a few fossils such as fish teeth,
brachiopods, and lar\'al gastropods, they are invariably devoid of sessile bottom-
inhabiting organisms, a fact which suggests that the bottom layer of sea-water
lacked the oxygen necessary to support life.

"The deficiency of ox>'gen is, therefore, the controlling chemical condition,
for it not only determines that the bacterial decay shall be of the anaerobic type,
but also prexents animal scavengers from devouring such organic matter as may
fall to the sea bottom, for no ainimal can be active in an oxygen-free medium.
* * *

"As was long ago pointed out by Bonnet, under ordinary circumstances all
of the products of decay are likely to either remain in solution or escape as gases
rather than to be precipitated. Under special conditions, however, most of them
remain in solid form and others react with the sediments of the bottom or with
materials in solution, in such a way as to form insoluble products. For example,
hydrogen sulphide, interacting with the iron compounds, forms the mineral
pyrite, which is common in certain types of black shales. In a similar way,
phosphoric acid in the presence of ammonia reacts with various substances, and
especially lime carbonates, in such a way as to produce phosphatic minerals, of
which the commonest is coUophanite, said to be hydrous calcium carbophosphate.
These changes have been carried out experimentally in the laboratory by several
investigators, and the necessary conditions are such as may readily occur on the
sea bottom where organic decomposition is in progress. The calcareous shells
and fragments lying on the ocean floor thus become phosphatized, and even such
organic materials as excretory pellets and pieces of wood are known to have been
altered in the same way. Bones, which initially contained about 58 f>er cent
tricalcium phosphate, have their organic matter completely replaced by phos-
phatic minerals, thus raising the ratio to 85 per cent or more. In addition,
coUophanite is precipitated in concentric layers around particles of sand or any
solids, forming round or elliptical granules which resemble the oolitic grains in
certain limestones. By the enlargement of these coatings, the granules, shells,
teeth, and other objects are cemented into hard nodules or even into continuous
beds of phosphatic rock. Such nodules have been dredged up from the bottom
of all the oceans in moderate depths, and are not uncommon in certain kinds of
marine limestones and sheiles now on land."

It is shown by Clark* that apatite and calcium phosphate are soluble
in carbonated waters and waters containing humic acid. The phosphate
is deposited, however, in the presence of calcium carbonate. The reaction
Ca3(P04)2 + 2H2CO3 -» 2CaHP04 + Ca(HC03)2 is reversible, but will only
be complete from left to right when all the carbonic acid is neutralized.

As the phosphates of the Basin Province are very largely in the petroleum-
bearing shale, it is probable that the precipitation of the phosphate was
due rather to the real absence of much CO2 than the contact of the dissolved
phosphate with limestone. This is a decided contributory proof of the
stagnant condition of the sea and the action of anaerobic bacteria.

• Clark, F. \V., Data of Geochemistry, U. S. Geological Survey Bull. 616, p. 519, 1916.

216 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

It does not seem at all probable that the phosphates of the Permo-
Carboniferous may be referred to accumulations of guano, as the character
and, so far as we know, the habits of the animals of the time were not such
as to permit such accumulations. Nor does there seem any way in which
the phosphate may have accumulated by secondary enrichment to the
extent in which they now exist. The suggestion by Blackwelder of original
accumulation in a stagnant sea seems by far the most reasonable suggestion.
If this be so then we must add to our picture of the surface of the North
American continent at the end of the Paleozoic a great, shallow, stagnant
sea covering the greater part of the northern end of the Basin Province
within the limits of the United States and becoming gradually shallower
toward the north. As shown by the various sections given, the sea was not
stagnant through all of its history, for at intervals it received normal lime-
stone deposits and many series of normal shales and sandstones, but at least
twice it was reduced to this state. What its borders were to the north we
may not know, as the exposed material there is not yet exactly placed in
the geological series, and if, as seems probable, the deposits are older than
the phosphate beds, the record has been removed by erosion. To the south
the sea evidently terminated in shallower water and great flats and shores
upon which accumulated the typical red beds.

The difference in the conditions of deposition on the two sides of the
northern end of the barrier which separated the Plains and the Basin
Provinces is not clear. It would appear that the red sediments of western
Wyoming gradually shaded into a sea, possibly a portion of the extension
from the Pacific Ocean which at times assumed the character of a relict sea.
The gradual elevation which was in progress from north to south, from
Alaska to northern California, extended its influence eastward and may
have been of large influence in cutting off and confining portions of this sea,
converting them at times into inclosed bodies of water which became
stagnant.

G. INTERPRETATION OF CONDITIONS IN BRITISH COLUMBIA

AND ALASKA.

In attempting an interpretation of the condition at the close of the
Paleozoic in British Columbia and Alaska, it will be well to summarize
briefly certain conclusions expressed by Daly^ in his memoir on the geology
of the forty-ninth parallel. On page 6 he divides the entire cordillera into
an Eastern geosynclinal belt and a Western geosynclinal belt. The two
overlap in the vicinity of the Columbia River. The eastern belt extends
from the summit of the Selkirk Range, just east of the Columbia River, to
the Great Plains. The formations are almost entirely sedimentary and
are included in one general structure which Daly refers to as the Rocky

• Daly, R. A., Geology of the North American Cordillera at the Forty-ninth Parallel,
Canadian Geological Survey, Memoir 38, 1912.

INTERPRETATION OF ENVIRONMENTAL CONDITIONS 217

Mountain geosynclinal prism. This prism may be traced from Alaska
through the Great Basin to Arizona.

The Western Belt is similarly largely made up of sedimentaries and can
be traced from Alaska to southern California.

On page 547 is given a statement of the principles upon which his correla-
tions have been based. Most important of these, for the purposes of this
paper, is the statement that many of the beds, unfossiliferous in themselves,
have been placed in the geological column by tracing them north or south
until they can be connected with fossiliferous beds of known age. The
accepted correlations in the summary description of the stratigraphy of
British Columbia and portions of Alaska will seem far more reasonable if
the reader has the principles set forth by Daly in mind.

In table 35, page 559, he places as equal in position (Carboniferous)
the following:

Southeastern Alaska: Western British Columbia Western Geosynclinal Belt:

Ketchikan series. and Yukon: Pend d'Oreille, Att-

Central Washington: Cache Creek group. wood. Anarchist, Ho-

Peshastin series, Haw- Oregon and northern Call- zomeen, Chilliwack se-

kins formation. Eastern fomia : ries.

formation. Nosoni formation, Mc- Middle California:

Cloud limestone, Baird Robinson formation,

formation, Bragdon Calaveras formation,

formation.

In table 36 he states that all these formations were deposited as marine
sediments accompanied by vulcanism and are terminated above by un-
conformities or by contact with bathylithic intrusions. In table 37 he
states that the Pennsylvanian in the western belt was a time of marine
sedimentation with very widespread vulcanism (general?). Following this
there was probably widespread though not energetic movements and local
unconformity.

On page 565 it is stated :

"The Western belt is in deep contrast with the Eastern belt and in a large
way the tvvo are in reciprocal relations. The area covered by the Western belt
has furnished most of the clastic material in the principal geosynclinal of the
Eastern belt; the Eastern belt has furnished most of the clastic material com-
posing the principal geosynclinal of the Western belt."

On page 568 is given, in the summary of the geological history, a statement
concerning the late Paleozoic:

"At or near the close of the Mississippian period the Western cordilleran belt
was certainly submerged, and the Eastern geosyiiclinal belt was broadly up-
warped, without other general deformation of the Rocky Mountain geosynclinal.
The main Pacific geosyncline was thus initiated or else deepened, so as to receive
a great load of Pennsylvanian sediments. Fossiliferous beds belonging to this
period have been found at inter\-als in the Western belt from the Columbia
River to Vancouver Island. So far as they are clcistic their materials seem to

218 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

have been derived from the newly emerged Eastern belt. The ancient relation
of the two belts was thus reversed, except for local, temporary embayments on
the east. This movement was, apparently, felt from Alaska to northern Utah at
least; farther south, in the region of the fortieth parallel, the reversal of relations
was postponed to the close of the Pennsylvanian period. Otherwise the Eastern
and Western belts have respectively behaved as units in the momentous change.
The larger part of the Eastern belt was to remain as land through the Permian,
Triassic, and most of the Jurassic periods; and even in the later periods to undergo
only partial submergence.

"The new relation between the two cordilleran belts was so similar to that
which obtained on the line of the fortieth parallel at the close of the Upper
Carboniferous period that it is instructive to review King's statement, published
on pages 536-537 of the volume on Systematic Geology, Fortieth Parallel Survey:

"'After the close of this great conformable Paleozoic deposition widespread
mechanical disturbance occurred, by which the land area west of the Nevada
Paleozoic shore became depressed, while all the thickest part of the Paleozoic
deposits from the Nevada shore eastward to and including the Wasatch rose
above the ocean and became a land area. Between the new continent and the
old one which went down to the west, there was a complete change of condition.
The land became ocean; the ocean became land. * * *

"'Upon the western side of the new land-mass, the Archaean continent,
having gone down, made a new ocean-bottom, and upon this immediately began
to accumulate all the disintegration-products of the new land-mass which the
westward-draining rivers and the ocean waves were able to deliver. Throughout
the Triassic and Jurassic periods the western ocean was accumulating its enor-
mously thick group of conformable sediments upon the Archaean floor, * * *
until, at the close of the Jurassic age, there had accumulated in the western sea
20,000 feet * * * of Triassic and Jurassic material.'

"During the Pennsylvanian period the main Pacific geosyncline was the
scene of heavy sedimentation with accompanying powerful vulcanism. The rock
exposures at the forty-ninth parallel do not suffice to show clearly the dynamic
events leading to the Triassic, but from Dawson's work in Vancouver Island, as
well as on the mainland, it appears that there was local deformation of the
Pennsylvanian beds in that part of the cordillera, followed by erosion of the
upturned strata, before these were buried beneath Triassic deposits. It is likely
that the same crustal movement affected the Forty-ninth Parallel section; and,
further, that it is to be correlated with the beginning of the Sierra Nevada down-
warp described, as above, by King. How long or how extensive was this tem-
porary return to land conditions in the Western Belt can not be declared. It is
known, however, that the Triassic period saw, at the forty-ninth parallel, a
resumption of marine sedimentation on the Pacific side of the belt. Argillites,
sandstones, and limestones, together with great piles of basic volcanic material
were then laid upon the Pennsylvanian formations in this region."

It is apparent from the summary descriptions of the Alaska and British
Columbia regions and the conclusions given by Daly that at the latest date
in the Paleozoic from which we have sediments preserved the Pacific Coast
region, even far south of Alaska, was submerged and probably receiving
sediments from the Eastern geosynclinal belt now raised above the level
of the sea. It is not so apparent that the northern end of the Eastern belt

INTERPRETATION OF EmaRONMENTAL CONDITIONS 219

was greatly elevated; indeed, the apparent equivalence of the limestones
in the Yukon Territory with those of Alaska seem to indicate that these
regions were submerged at the same time, but there is little doubt that the
progressive elevation of these lands was from north to south and as little
that the Eastern belt was land at least in the Permo-Carboniferous. This
elevation was sufficient to furnish much sediment to the west and also to
terminate the basins of the Plains and Basin Provinces not far south of the
present international boundary. The elevation furnished also the probable
route of migration of the Asiatic plants represented by the fern Gigantopteris,
which reached as far south as Texas.

CHAPTER VIII.

PALEOBOTANICAL EVIDENCE AS TO THE EQUIVALENCE OF
THE BEDS IN THE EASTERN AND THE PLAINS PROVINCES.

As evidence of the climatic conditions and the equivalence of the beds
in the two widely separated regions where plant remains occur, the fossil
plants are perhaps the most accurate that can be used. The identification
of the several genera and species, while not final, has reached a stage where
dependence can be placed upon the lists as a whole. David White has
given a summary of the plants occurring in the Eastern Province which
may be taken as the standard for comparison with the other regions and
beds.' His list is as follows:

Plants ReccH'ded from the Gsnemaugh Formation (Penns^vanian) :

Cheilanthites (Sphenopteris) solidus (Lesq.)

D. White.
Cheilanthites obtusilobus (Brogn.) Gopp.
Cheilanthites squamosus (Lesq.) D. White.
Mariopteris sillimanni (Brogn.) D. White.
Mariopteris nervosa (Brogn.) Zeill.
Sphenopteris minutisecta Font, and I. C.

White.
Sphenopteris (Crossotheca) ophioglossoides

Lesq.
Alethopteris serlii (Brogn.) Gopp.
Pecopteris unita Brogn.
P'ecoj>teris villosa Brogn.?

Pecopiteris cf. jenneyi D. White.
Pecopteris oreopteridia (Scloth.) Stemb.
Pecopteris miltoni (Artis) Stemb.
Pecopteris poI>-morp^ Brogn.
Pecojjteris sp. D. \\'hite.
Neuroprteris ovata Hoffm.
Neuropteris fimbriata Lesq.
Neuropteris scheuchzeri Hoffm.
^henophyllum ma jus Bronn.
Lycopxxiites pendulus Lesq.
Sigillaria fissa Lesq.
Lepidocystis \'esicularis Lesq.
CoixbicarpcMi gutbieri (Gein.) Gr. 'Ery.

Plants recorded from the Monongahela formation (Pennsylvanian):

Maripoteris ? spinulosa (Lesq.) D. White.
Alethopteris aquilina (Schloth.) Goepp.
Pecopteris unita Brogn.
Pecopteris villosa Brogn. ?
Pecopyteris cf. jenne)"! D. WTiite.
Pecopteris notata Lesq.
Pecopteris nodosa (Goepp.) Schimp.
Dicksonites pluckeneti (Schloth.).

Neuropteris callosa Lesq.
Neuropteris crenulata Brogn.
Neuropteris grangeri Brogn.
Neuropteris scheuchzeri Hoffm.
Lescuropteris moorii (Lesq.) Schimp.
Aphlebia filicifonnis (Gutb.)_Sdump.
S^llaria menardi Brc^:n.

List of fossil plants reported from the Dunkard formation (Permian):

and L C.

(Font.

Fc»t. and L C.

Diplothema pachyderma

White) D. White.
Sphenopteris minutisecta

White.
SpJjenopteris (Cym<^k>ssa) breviloba (Font.

and I. C. White) D. White.
Sphenopteris (C>Tnoglossa) formosa (Fwjt. and

I. C. White) D. WTiite.
Sphenopteris (C>'TOOglossa) lobata (Font, and

I. C. White) D. White.

Sphenopteris (Cymoglossa) obtusifolia {Foot.

and L C. White) D. White.
Sphenopteris lescuriana Meek.
Sphenopteris dentata Font, and L C. White.
Sphenopteris auriculata Font, and L C. White.
Sphenofrteris foliosa Font, and L C. White.
Sphenopteris hastata Font, and L C. White.
Sphenopteris acrocarpa Font, and L C. White.
Sphenopteris sp.? Font, and L C. White.
Pecopteris pluckeneti (Schloth.) Stemb.

' White, David, The Fossil Flora of West V'irginia, West Virginia Geological Survey, vol.
V (a), part II, pp. 390-453, 1913.

221

222

ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Pecopteris pluckeneti var. constricta Font, and

I. C. White.
Pecopteris germari (Weiss.) Font, and I. C.

White.
Pecopteris germari var. crassinervis Font, and

I. C. White.
Pecopteris germari var. cuspidata Font, and

I. C. White.
Pecopteris dentata var. crenata Font, and I. C.

White.
Pecopteris dentata var. parva Lesq.
Pecopteris pachypteroides Font, and I. C.

White.
Pecopteris (Goniopteris) emarginata (Gopp.)

D. White.
Pecopteris (Goniopteris) oblonga (Font, and

I. C. White) Miller.
Pecopteris (Goniopteris) newberriana (Font.

and I. C. White) Miller.
Pecopteris (Goniopteris) longifolia (Font, and

I. C. White) D. White.
Pecopteris (Goniopteris) elliptica (Font, and

I. C. White) D. White.
Pecopteris (Goniopteris) sp. ? Font, and I. C.

White.
Pecopteris (Goniopteris) arguta (Brogn.).
Pecopteris goniopteroides Font, and I. C. White.
Pecopteris (Goniopteris) elegans (Germ.).
Pecopteris arborescens (Schloth.) Brogn.
Pecopteris arborescens var. integripinna Font.

and I. C. White.
Pecopteris pennaeformis var. latifolia Font, and

I. C. White.
Pecopteris candolleana Brogn.
Pecopteris oreopteridia (Schloth.) Sternb.
Pecopteris rarinervis Font, and I. C. White.
Pecopteris imbricata Font, and I. C. White.
Pecopteris platynervis Font, and I. C. White.
Pecopteris asplenioides Font, and I. C. White.
Pecopteris rotundiloba Font, and I. C. White.
Pecopteris microphylla Brogn.
Pecopteris angustipinna Font, and I. C. White.
Pecopteris tenuinervis Font, and I. C. White.
Pecopteris subfalcata Font, and I. C. White.
Pecopteris heeriana Font, and I. C. White.
Pecopteris schimperiana Font, and I. C. White.
Pecopteris lanceolata Font, and I. C. White.
Pecopteris inclinata Font, and I. C. White.
Pecopteris merianopteroides Font, and I. C.

White. .
Pecopteris rotundifolia Font, and I. C. White.
Pecopteris sp. ? D. White.
Pecopteris ovoides Font, and I. C. White.
Pecopteris latifolia Font, and I. C. White.
Pecopteris pteroides Brogn.
Pecopteris miltoni (Artis) Sternb.
Pecopteris polymorpha Brogn.
Pecopteris elliptica Bunbury.
Pecopteris (Callipteridium) grandifolia (Font.

and I. C. White) D. White.
Pecopteris (Callipteridium) oblongifolia (Font.

and I. C. White) D. White.
Pecopteris (Callipteridium) odontopteroides

(Font, and I. C. White) D. White.
Pecopteris (Callipteridium) unitum (Font, and

I. C. White) D. White.
Pecopteris (Callipteridium) dawsonianum

(Font, and I. C. White) D. White.

Alethopteris virginiana Font, and I. C. White.
Alethopteris gigas (Gutb.) Gein.
Callipteris conferta (Sternb.) Brogn.
Callipteris lyratifolia (Grand 'Eury) var. coria-

cea (Font, and I. C. White) D. White.
Callipteries curretiensis Zeill. (Font, and I. C.

White) D. White.
Taeniopteris lescuriana Font, and I. C. White.
Tseniopteris newberriana Font, and I. C. White.
Taeniopteris newberriana var. angusta Font.

and I. C. White.
Neuropteris ovata Hoffm. (variety).
Neuropteris gibbosa Lesq.
Neuropteris planchardi Zeill. var. longifolia

(Font, and I. C. White) D. White.
Neuropteris dictyopteroides Font, and I. C.

White.
Neuropteris fimbriata Lesq.
Neuropteris cordata Brogn.
Neuropteris auriculata Brogn.
Lescuropteris adiantoides Lesq.
Odontopteris reichiana Gutb.
Odontopteris obtusiloba var. rarinervis Font.

and I. C. White.
Odontopteris nervosa Font, and I. C. White.
Odontopteris densifolia Font, and I. C. White.
Caulopteris gigantea Font, and L C. White.
Caulopteris elliptica Font, and L C. White.
Aphlebia (Rhacophyllum) lanciatum (Font, and

L C. White) Sellards.
Aphlebia lactuca (Presl.) Sterzel.
Aphlebia (Rhacophyllum) speciosissima

(Schimp.) D. White.
Aphlebia (Rhacophyllum) filiciformis var.

majus (Font, and I. C. White) D. White.
Equisetites rugosus Schimp.
Equisetites striatus Font, and I. C. White.
Equisetites elongatus Font, and I. C. White.
Calamites suckowi Brogn.
Nematophyllum angustum Font, and I. C.

White.
Annularia radiata (Brogn) Sternb.
Annularia sttllata (Schloth.) Wood.
Annularia sphenophylloides (Zenk.) Gutb.
Annularia carinata Gutb.
Annularia minuta Brogn.
Sphenophylliim oblongifolium (Germ, and

Kauff.) Ung.
Sphenophyllum longifolium (Germ.) Gein and

Gutb.
Sphenophyllum thoni Mahr.
Sphenophyllum fontaineanum S. A. Mill.
Sphenophyllum filiculme Lesq.
Sphenophyllum tenuifolium Font, and I. C.

White.
Sphenophyllum angustifolium Gutb.
Sphenophyllum densifolium Font, and I. C.

White.
Sigillaria brardii Brogn.
Sigillaria approximata Font, and I. C. White.
Cordaites crassinervis Font, and I. C. White.
Baiera virginiana Font, and I. C. White.
Saportaea salisburioides Font, and I. C. White.
Rhabdocarpus oblongatus Font, and I. C.

White.
Carpolithes bicarpus Font, and I. C. White.
Carpolithes marginatus Font, and I. C. White.
Gulielmites orbicularis Font, and I. C. White.

EQUIVALENCE OF BEDS IN EASTERN AND PLAINS PROVINCES 223

An annotated list giving the synonymy and exact location of each
species is given on pages 392 to 429 of the same publication. Of this list
D. White says:

"The floras of the Conemaugh have had but little study and their diflferentia-
tion from those of the Monongahela, on the one hand, or from those of the
Allegheny on the other, is therefore at present very incomplete. The composition
and characteristics of the plant life of the Monongahela are also but littie under-
stood, though it is known that the floras contain much that is present in, though
not peculiar to, the Dunkard (basal Permian)."

As noted above (page 66), D. White has reported the occurrence of a
species of Callipteris, a genus diagnostic of the Permian, recently discovered
in the upper half of the Conemaugh.

The best summary of the plants of the Plains Province has been given
by David WTiite^ in his description of the characters and relationships of the
genus Gigantopteris, which he regards as not closely related to any known
Paleozoic type:

"Its nearest, though perhaps very distant, relatives are, I believe, to be
found in the fossils described by Morris as Pecopteris goepperti, really a Callipteris,
from the Permian sandstones near Bielebei in the Urals."

Following are the lists of plants given by WTiite from the beds of Texas,
Oklahoma, Kansas, and Colorado. The forms marked with an asterisk (*)
are characteristic Permian or Permo-Carboniferous species.

" Prelimmary List of the Fossils from the Main Plant Bed (M) [in the breaks of the Little Wichita, 4.5
miles southeast of Fulda) and 'Castle Hollow' (H) near Fulda, Texas:

Diplothema sp. ? M.
Pecopteris arborescens, H.
Pecopteris hemitelioides, H, M.
Pecopteris densifolia?, H.
Pecopteris tenuinen-is, M.
Pecopteris grandifolia, M, H.?
Pecopteris, sp., M.
Aphlebia, sp., H.
*Odontopteris neuropteroides, M.
*Odontopteris fischeri? M.
•Gigantopteris americana, M, H.
Neuropteris cf. lindahli, H.
Neuropteris cordata?, M.
*Taeniopteris multinervis, H, M.
*Taeniopteris abnormis, ^L
*Taeniopteris coriaoea?, ^^.
Taeniopteris, new species, M.
*Annularia spicata, H.
*Annulciria? maxima, M.

•Sphenophyllum obo%'atum, M.
Sphenophyllum?, sp., H.
Sigillaria, sp., M.
*Sigillariostrobus hastatus, H.
Cordaites cf. principalis, M.
•Poacordaites cf . tenuifoUus, M.
*Walchia piniformis, 'Si.
*\\alchia schneideri?, H.
*Gomphostrobus bifidus, H.
*Gomphostrobus? sp., M.
Aspidiopsis, sp., M.
•Araucarites, new species, M, H.
Carpolithes, sp., H.
Insect wings, M.
Anthracosia, M.
Estheria, M. H.
Ostracods, M. H.
Fish scales, M. H.

"FVovisbnal list of fossil plants from Perry (P) and Eddy (E), Oklahoma:

Diplothema pachyderma, E. *Odontopteris ch. permiensis, E.

Pecopteris c>'athea, P. Neuropteris, sp., E.

•Pecopteris geinitzi, P. *Taeniopteris multinervis, P, E.

•Callipteris, sp., E. *Taeniopteris abnormis, P.

•Gigantopteris americana, E. P. *T«eniopteris, sp., E.

' White, David, The Characters of the Fossil Plant Gigantopteris Schenk and Its Occurrence
in North America, Proceedings U. S. National Museum, vol. 41, p. 493, 1915.

224

ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Dolerophyllum?, sp., E.
Equisetites, sp., E.
Annularia stellata, P.
*Sphenophyllum obovatum, E.
*Sphenophyllum cf. latifoliutn, P.
*Sphenophyllum stoukenbergi? P.

Sigillaria, sp. ?, P, E.
*Walchia imbricata? P.
*Walchia cf. gracilis, E.
*Araucarites, sp., P, E.
Carpolithes, E."

" List of species provisionally identified from the Permian of Kansas. (R) Wreford limestone, west of
Reece; (W) shales near the Winfield formation, northeast of Washington; (B) Wellington formation
south of Banner; (C) Wellington formation south of Carlton; (S) Wellington formation east of
Salina.

*Schizopteris cf. trichomanoides, W.
Pecopteris unita, W.
*Pecopteris pinnatifida, W.
*Pecopteris cf. geinitzi, W.
Pecopteris hemiteloides, W.
Pecopteris bucklandi? W.
Pecopteris polymorpha, W.
•Scolecopteris elegans, C.
*Cladophlebis cf. tenuis, C.
*Callipteris conferta, W, C.
*Callipteris subauriculata, S, C, B.
*Callipteris cf. curretiensis, C.
*Callipteris cf. Jutieri, R.
*Callipteris cf. goepperti, R.
*Callipteris oxydata, S.
•Callipteris whitei, B.
*Callipteris lyratifolia? S.
*Callipteris cf. scheibei, B.
Odontopteris brardii W.
Odontopteris minor, W.
•Glenopteris splendens, B, C.
*Glenopteris lineata, B.
•Glenopteris sterling!, B, C.
*Glenopteris lobata, C.

'Provisional list of plants from Fairplay, Colorado:

Neuropteris auriculara?, W.
Neuropteris odontopteroides, W.
Neuropteris scheuchzeri, var., W.
Neuropteris permiana, W.
*Taeniopteris multinervis, W.
Taeniopteris coriacea, B, C.
Tseniopteris coriacea, var. linearis, B, C.
*Sphenophyllum obovatum, C, B.
*Sphenophyllum cf. stoukenbergi, W.
*Sphenophyllum cf. thonii, W.
*Sigillariostrobus hastatus, R.
Noeggerathia? new species, B.
Cycadospadix? sp., C.
Cordaites principalis, R.
*Poacordaites linearis? C.
*Walchia piniformis, R.
*Walchia cf. filiciformis, R.
*Walchia sp., C.
*Voltzia sp. C.
*Ullmannia? sp., C.
*Schutzia? cf. anomala, R.
*Araucarites? sp., C.
Rhabdocarpis, new species, R.
Carpolithes, sp., S. B."

(The forms marked A are from the Lacoe collection examined by Lesquerouz and are probably from a wmewhat higher horizon
than the forma collected by D. White, marked B.]

•Sphenopteris schimperiana?, B.
*Sphenopteris lebachensis Weiss, A.
Sphenopteris dentata F. and L C. W., A.
*Sphenopteris gutzholdi Gutbier, A.
•Pecopteris pinnatifida Gutbier, B.
Pecopteris fceminaeformis (Schlotheim) Zeiller, A.
Pecopteris arborescens (Schlotheim) Brongniart, B.
Pecopteris (Danaeites GSppert), sp. B.
•Scolecopteris elegans Gutbier, B.
•Callipteris cf. hymenophylloides Weiss, A.
•Callipteris cf. lyratifolia (Goppert), B.
Odontopteris subcrenulata Rost, B.
Neuropteris auriculata Germar, B.

•Calamites kutorgse? B.

•Sphenophyllum obovatum Sellards, B.

Sigillaria? sp., B.

•Sigillariostrobus hastatus, A. B.

Poacordaites, sp., A.

•Walchia piniformis (Schlotheim), A, B.

•Walchia hypnoides, A, B.

•Walchia gracilis? A.

•Ullmannia, sp.. A, B.

•Voltzia, sp., A.

•Araucarites, sp., A, B.

•Gomphostrobus bifidus, B."

'List of plants from the Denver and Rio Grande tunnel below Swissvale, Colorado:

•Callipteris sp.

•Psygmophyllum cf. cuneifolium.
Odontopteris subcrenulata Rost?
Macrostachya? sp.

•Sigillariostrobus hastatus.
•Walchia cf. piniformis.
Walchia cf. imbricata.
Rhabdocarpos dyadicus Geinitz?"

Concerning these lists D. White says:^

"Fragmentary and incompletely representative of the several floras as the
lists may be, they yet show some interesting aspects of the distribution of the
Permian species. Thus, the genus Walchia, unknown in th^ Permian of the

* White, David, The Characters of the Fossil Plant Gigantopteris Schenk and Its Occurrence
in North America, Proceedings U. S. National Museum, vol. 41, page 511.

EQUIVALENCE OF BEDS IN EASTERN AND PLAINS PROVINCES 225

Appalachian trough, is present at most of the localities, while CaUipteris, which
is ver>' meagerly represented in eastern North America, is common and highly
differentiated in Kansas and Colorado. Gomphostrobus, another type char-
acteristic of the Permian of western Europe and hitherto unknown in North
America, is present in Kansas, Colorado, Oklahoma, and Texas. The common
t\'pe of simple-leafed Tceniopteris, diagnostic of the western European lower
Permian, is nearly everywhere present, sometimes accompanied by other forms,
one of which, with distant, simple nerves, is of distinctly Mesozoic aspect.

"In addition to the many CaUipteris and Walchia sp)ecies just mentioned, the
provisional lists from the western Permian include a number of other forms near
to, if not identical with, diagnostic Old World Permian types hitherto unknown in
this continent. Among these are Schizopteris cf. trichomanoides, Sphenopteris
lebachensis, Pecopteris geinitzi, Pecopteris pinnatifida, Cladophlebis? cf. tenuis,
Scolecopteris elegans, Odontopteris subcrenulata, Tcsniopteris abnormis, Anntdaria
spicata, Khabdocarpos cf. dyadicus.

"It is probable that several cosmopolitan sp>ecies of Pecopteris and Sphenop-
teris •will be found to have accompanied Tceniopteris mtdtinervis from western
Europe to eastern China.

"The examination of the materials from the Western Interior and Rocky
Mountain basins shows that while the flora is composed mainly of types common
to western Europe which have undoubtedly been distributed along essentially the
same northeastern Arctic-American route by which the Pennsylvanian floras
migrated, it contains also a somewhat unique element unmistakably derived from
eastern Asia. The latter includes the Gigantopteris, the peculiar Anntdaria,
and a Tceniopteris form, to which should possibly be added the representatives
of Araucarites and Neuropteridium. The migration of this land-plant element
was very probably by the north Pacific.

"The most important deduction to be drawn from the occurrence of Gigan-
topteris and its particular associates in North America is the essential continuity
of environmental conditions indicated thereby. The vital conditions under which
the types lived in Oklahoma and Texas can not have been very far different in
their essential respects from those prevciiling in the Chinese habitats of the types.
Environmental conditions sufficiently uniform to enable these plants to thrive
must have attended the route of their land migration. We may therefore con-
clude that a climatic environment essentially similar extended from China to
western North America; that is, that during Gigantopteris time western North
America and portions of eastern Asia were probably included in the same climatic
province. The mingling of the western European flora with the Chinese elements
in Oklahoma and Texas suggests that the latter region may have been on the
eastern border of the pro\dnce.

"Another interesting feature of the western Permian is the presence of fronds
possibly identical with Psygmophyllum cuneifolium, Odontopteris per miensis, Odon-
topteris fischeri, jmd SphenophyUum stoukenbergi, species that seem not to have
been known outside of the Uralian region, from which they were described.
Possibly the remarkable Kansas type described by Sellards* as Glenopteris, which
is unlike any European t>'pe of its period, and which may be nearest related to the
Neuropteris salicifolia of Morris, also is of Uralian or Asiatic descent. The types
of Uralian origin also may have reached western North America by the north
Pacific route.

■ Kansas University Quarterly, volume 9, 1900, p. 179.
16

226 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

"According to their composition and relations the floras of the younger
Carboniferous in Shansi and Sheng-King, or Manchuria, which are either at the
latest Pennsylvanian stage or in the early Permian, may with probable safety
be assumed to have antedated the early Gondwana glaciation and the existence
of the Gangamopteris flora in southern Asia. The question arises, then, whether
the floral peculiarities of the Gigantopteris province are due in part to climatic
changes leading to refrigeration in India, and whether later the climate of the
Gangamopteris province extended over a portion at least of the Gigantopteris
province, and if so, whether it did not cover a part of western North America."

And further, on page 513:

"The very incomplete collections of fossil plants from the Wichita formation
in Texas, from its supposedly approximate equivalents in Oklahoma, from the
Chase and Sumner groups in Kansas, and from the great series of undifferentiated
'red beds' in the Rocky Mountain region of southern Colorado, show a mixed
flora embracing (i) mainly representatives of the Permian flora of western
Europe, and including many types not previously known in North America;
(2) a smaller portion peculiar to the Gigantopteris association in south central
and southwestern China; and (3) several types apparently identical with or very
close to forms hitherto known only in the Permian of the Uralian region.

"The distribution of the floral elements indicates that the western European
or cosmopolitan elements of the flora migrated between North America and
Europe, presumably by the same general northeastern route as that followed by
their Pennsylvanian predecessors, while the distinctly Chinese types must have
come to Texas and Oklahoma by the north Pacific (Alaskan) route, by which the
related Uralian forms may also have migrated. Since the land migration of the
Chinese types could hardly have been accomplished without the aid of essential
continuity of environmental conditions, and since it is probable that the Gigan-
topteris elements lived under climatic conditions mainly similar in both Texas
and China, the conclusion appears justified that the climatic province under
which they thrived in Asia extended to western North America and that it
included the region of noith Pacific migration. The mingling of western Euro-
pean species with Gigantopteris in the southwestern 'red beds' is construed to
indicate that this region was probably on the eastern border of the Gigantopteris
province."

Sellards* says of the flora of the Wellington shales:

"The flora of the Wellington differs in toto, so far as species are concerned,
from that of the Cherokee shales, and contains only a small proportion of species
found in the Douglas formation. Of the species listed from the Wellington only
a few have been positively identified with forms found in the Le Roy and Lawrence
shales. More than two-thirds of the Wellington species are either identical with
or most closely related to species or genera characteristic of the European Permian.
The points which seem to have the most importance as bearing on the correlation
of the Wellington are the following: (i) The complete absence of species in any
way confined to or distinctive of the Coal Measures. (2) The comparatively
small number of species originating as early as Upper Coal Measures time. (3)

* Sellards, E. H., Fossil Plants of the Upper Paleozoic of Kansas, University of Kansas
Geological Survey, volume ix, p. 462, 1908. This is a final paper; preliminary papers
were published in the Kansas University Quarterly, volumes 9 and 10, 1900-1901.

EQUIVALENCE OF BEDS IN EASTERN AND PLAINS PROVINCES 227

The presence of a few species common to and characteristic of the Permian of
Europe. (4) The close relation of the new forms to species characteristic of the
European Permian. (5) The distinctly Permian facies of the flora as a whole
and its marked advance over the flora of the Upper Coal Measures.

"The advance in the flora consists in the number of sp>ecies and abundance
of individuals of callipterid and taeniopterid ferns and of the new genus Glenopteris,
which appears to be related on the one hand to callipterid ferns of Permian types
and on the other to the Triassic genera Cycadopteris and Lomatopleris.

" The evidence derived from the fossil plants seems to assure the reference of
the Wellington to the true Permian in the European sense.

"The flora of the formations intervening between the Douglas formation
and the Wellington shales is much less satisfaictorily known. A good deal of
interest is attached to the discovery of plants in the Wreford limestone, especially
as this formation has been recently regarded as the base of the Permian in Kansas.
Nine species have been obtained from this locality, as follows: Baiera sp., Callip-
teris conferta, Callipteris sp., Cardiocarpon sp., Carpolithes sp., Cordaites sp.,
Rhabdocarpos sp., Sigillaria sp., Walchia pinnifortnis. The collection obtained
from this formation is small and comes from a single locality near Reece, Kansas.
The association of the flora so far as obtained is with the Wellington rather than
with Coal Measures flora. The presence of Walchia in abundance, and of callipn
terid ferns, along with the small species of seeds common to the Wellington,
together with the absence, so far as yet noted, of all of the common Coal Measures
species, gives the flora of the Wreford, £is develop>ed at Reece, a distinctive
Permian facies.

"Coal Measures species, although rare in the collection obtained from the
Wreford limestone at the Reece locality, recur in some abundance in the horizon
at Washington, regarded by Beede as near the top of the Chase formation."

Other lists of the Kansas plant fossils were given by D. White :^

"Ki.MDAi.K Flora at Onaga.

Pecopteris newberriana F. and I. C. W. Xeuropteris auriculata Brongn.?

Pecopteris hemitelioides Brongn. Neuropteris scheuchzeri Hoffm.

Pecopteris oreopteridia (Schloth.) Brongn.? E>aubreeia sp.

Pecopteris cf. polymorpha Brongn. Asterophyllites equisetiformis (Schloth.) Brongn.

Odontopteris brardii Brongn. Annularia stellata (Schloth.) Wood.

Odontopteris moorii (Lx.) D. W. Radidtes capillaoeus (L. & H.) Pot.

Neuropteris plica ta Stemb.

"The 13 species from Onaga communicated by Mr. Cievecoeur are, as com-
pared with the floras of Lansing and Thayer, obviously of much later age. No
species in any way characteristic of the Lower Coal Measures or the Allegheny
formation remains. On the other hand, the ferns, either as individual species or
as phases of species having wide range, are clearly indicative of a stage at least
very high in the Upper Carboniferous (Pennsylvanian). Nearly all the species
have been reported from either the Permian of Europe or the Dunkard formation
of the United States, though, with the possible exception of Pecopteris newberriana,
none are distinctly characteristic of the Permian. Most of the forms present
occur in the Dunkard formation, whose flora was fully treated by Professors
Fontaine and I. C. White.* Yet the small flora from Onaga contains none of the

' White, David, in Adams, Girty and White, Stratigraphy and Paleontology' of the Upper
Carboniferous Rocks of the Kansas Section, U. S. Geological Survey Bull. 211, p.

115. 1903-
' Second Geological Sur\'ey Pennsj'h'ania, Rept. PP, Harrisburg, 1880.

228 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

special types or characteristic Permian forms which are present in the Dunkard,
and on account of which the greater part of the Dunkard is regarded as Permian.

"It would seem, however, that the Onaga flora should be of later date than
the Pittsburgh coal, since the facies presented by several of the species has not
yet been seen at so low an horizon. Thus the very large size of the form referred
to Pecopteris hemilelioides; the form referred tentatively to P. polymorpha, but
which seems hardly to differ unless in size from the Dunkard Callipteridium
grandifolium; the form identified by Lesquereux from the Dunkard as Neuropteris
plicata;^ the dilated heteromorphous N. scheuchzeri, and perhaps the type here
doubtfully listed as N. auriculata, all seem to indicate a stage as high as the roof
of the Pittsburgh coal, while some of these peculiar phases are present above and
are not yet known below the Waynesburg coal, i. e., in the Dunkard. Pecopteris
newberriana, which is possibly characteristic of the Dunkard, appears hardly
distinguishable from the small phase of P. fcemincBjormis, figured by Zeiller,*
from the Permo-Carboniferous of France. The normal form of the latter species
is reported from the roof of the Pittsburgh coal in the Appalachian trough. It
is probable that the apparent absence of many of the Dunkard forms in the lower
beds is due entirely to the lack of study of the plants in the strata between the
roof of the Pittsburgh coal, which forms the base of the Monongahela formation,
and the Waynesburg coal, the top bed of that formation. The absence of lepido-
phytes from the ma,terial in hand constitutes negative, and, under the circum-
stances, scarcely important proof, since their failure to be present may be due
to chance in preservation or collection.

"The evidence presented by this small Onaga flora may, therefore, be con-
strued, so far as it represents the plants of its horizon, as indicating a stage
probably within the Monongahela formation of the Appalachian region, or pos-
sibly as high as the lowest part of the Dunkard formation, although, with the
exception of Pecopteris newberriana, the collection in hand does not contain any
species characteristic of the Permian of the Old World, and does not signify a
Permian age for the Onaga (Elmdale) beds.

"Marion? (Wellington in Part?) Flora of Dickinson County.

Sphenopteris sp. Sell. Glenopteris lineata Sell.

Pecopteris sp. Sell. Glenopteris sterling! Sell.

Callipteris conferta Sternb. Glenopteris? lobata Sell.

Callipteris conferta var. obliqua (Goepp.) Weiss. Odontopteris sp. Sell.

Callipteris conferta var. lanceolata Weiss. Neuropteris sp. Sell.

Callipteris conferta var. vulgaris Weiss. Taeniopteris coriacea Goepp.

Callipteris n. sp. Taeniopteris coriacea var. lineata Sell.

Glenopteris splendens Sell. Taeniopteris newberriana F. and I., C. W.

Glenopteris simplex Sell. Sphenophyllum sp. Sell.

"The above list includes only the material published or communicated to the
National Museum by Mr. E. H. Sellards, by whom the collection of the State
University survey is being elaborated. The specimens are described by him as
coming either from the topmost beds of the Marion formation or possibly from
the base of the Wellington formation, next above the Marion. The flora is
regarded by Mr. Sellards' as of Lower Permian age. I have not had an oppor-
tunity to examine the remaining material at the State University, but if the com-

' Probably specifically different from the older form, which seems to agree with Sternberg's

species and which was placed under the same name by Lesquereux.
' Fl. foss. bassin houill. et perm, de Brive, 1892, pi. iv, fig. 5, 6.
' Trans. Kans. Acad. Sci., vol. xvii, 1900 (1901), p. 208.

EQUIVALENCE OF BEDS IN EASTERN AND PLAINS PROVINCES 229

position of the entire flora proves to be of so young a character as the material
described or placed in my hands by Mr. Sellards, his conclusion that the beds
are of so late date as the Lower Permian will appear to be fully justified. I am
not informed whether any of the gymnospermic species so important in, and so
typically characteristic of, the Permian of Europe or Prince Edward Island are
present in Kansas. However, such pteridophytic material as has come to me
for examination is more nearly typical and characteristic of the Permian than
any flora that I have yet seen from another formation in the United States.

"If the plants preliminarily listed above are representative of the plant life
of the Upper Marion or the Wellington formation, the flora of these beds is
probably of a date fully as late as the eau-lier of the floras generally referred to
the Permian in western Europe. In any event a flora containing these species
can hardly be older than the topmost Carboniferous, or transitional from the
Upper Carboniferous to the Permian."

A. EVIDENCE OF FOSSIL INSECTS AS TO EQUIVALENCE OF THE

PERMO-CARBONIFEROUS BEDS IN THE EASTERN AND

THE PLAINS PROVINCES.

Sellards published the results of his investigations on the insects of the
late Paleozoic of Kansas in a series of papers which appeared in the American
Journal of Science during the years 1906, 1907, and 1909.^

The insects described in these papers came from a locality about 3.5
miles southeast of Banner City, Dickinson County, Kansas. The insects
occur, with fossil plants, in a fine-grained, laminated limestone associated
with a hard concretionary limestone. This limestone belongs in the Welling-
ton horizon and lies directly beneath the Cretaceous in this locality. In-
cluding doubtful forms, there are more than 60 species described, with 35
genera, all of which are new. The cockroaches from this locality, described
elsewhere,^ add 10 species and 2 genera, one of which is new. The larger
groups represented are:

Odonata, i genus and species.

Plectoptera, 10 genera and 13 species. Handlirsch has recognized ephemerids as

occurring sparingly in the Permian of Russia.
Megasecoptera, i specimen.
Oryctoblattinidae, 2 genera.
Protorthoptera, 20 genera, 43 species.
Paleoblattidae, 2 genera, 10 species. The rarity of cockroaches is a peculiarity of

this locality.

From the Birmingham shale of the Conemaugh series from near Steuben-
ville and Richmond, Ohio, there were obtained 22 species, belonging to 3
genera, of cockroaches; no other insects were found at this place. Only
one of the genera from Birmingham shale has been found at the Kansas
locality and not one of the species; the two other genera, however, have

* Sellards, E. H., Types of Permian Insects, Amer. Jour. Sci. .vols. XXin, xxvii, 1906, 1907

1909. Correlation of the Insect-bearing Horizon, in part in, p. 169.

* Sellards, E. H., Cockroaches of the Kansas Coal Measures and the Kansas Permian,

University of Kansas Geological Survey, vol. IX, p. 501, 1908.

230 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

been found in the Upper Pennsylvanian of Kansas. "The insect remains
thus far obtained do not therefore permit a close correlation of the Birming-
ham shales with the Kansas section. It seems probable, however, that the
formation is of a somewhat later age than the Leroy shales with the Kansas
Coal Measures." The genus Spiloblattina found in Ohio is also found in the
deposits at Fairplay, Colorado, regarded by David White from paleobotan-
ical evidence as of Permian or late Pennsylvanian.

The Cassville shale, at the base of the Dunkard, has afforded numerous
specimens from the type locality at Cassville, West Virginia. Scudder
recognized 6 species and 5 genera, all cockroaches. Only one genus, Eto-
blattina, is common to the West Virginia and Kansas localities; no species
are common.

Sellards remarks:

"The predominance of the cockroach fauna, together with the absence of
such advanced types as true ephemerids, leads to the view that the Cassville
locality, although Permian, is much older than the Wellington shales of Kansas.
[And further] The insects of the Wellington are on the average of small size as
compared with Coal Measure insects. This is particularly noticeable among the
cockroaches, all of which are small. This dwarfing of the fauna is of interest
as probably indicating unfavorable climatic conditions."

From the Wichita beds of Archer County, Texas, Sellards has described
two species of the genus Etohlattina}

• Sellards, E. H., Two New Insects from the Permian of Texas, Carnegie Inst. Wash. Pub.
No. 146, p. 151, 1911.

CHAPTER IX.
CLIMATOLOGY OF THE LATE PALEOZOIC.

The suggested causes of climatic change in geologic time which are now
considered as most probable are: the atmospheric theory (carbon-dioxide
content), the deformation theory, the sun-sp>ot theory, and the solar-radiation
theory. The first of these is dismissed by Clements, in his Plant Succession,^
with a brevity that seems hardly commensurate with the attention it has
received in other quarters. WTiatever may have been the local cause of
climatic change in any limited locality, it is hardly to be supposed that the
change over such a large area as North America in the late Paleozoic was
not a part of a world-\\nde effect produced by some cosmic alteration in
which a change in the composition of the atmosphere may well have played
a large part. Certainly such a theory' is far more applicable in its observ-
able data to a time so remote as the Permo-Carboniferous than any which
has to do with the intensity of solar radiation or the number of sun-spots.
Chamberlin has already shown the value of this theory and its applicability
to the great climatic cycle which culminated in the glaciation at the close
of the Paleozoic.

Schuchert,* in a brief critique of the atmospheric theory, states:

"The glacial climates are irregular in their geological appearance, are variable
latitudinally, as is seen in the geographic distribution of the tillites between the
poles and the equatorial region, and finally, that they appear in geologic time as
if suddenly introduced. These differences do not seem to the writer to be condi-
tioned in the main by a greater or smaller amount of carbon dioxide in the
atmosphere, for if this gas is so strong a controlling factor, it would seem that at
least the glacial climates should not be of such quick development. On the other
hand, an enormous amount of carbon dioxide was consumed in the vast limestones
and coals of the Cretacic, with no glacial climate as a result;* though it must
be admitted that the great limestone and the vaster coail accumulations of the
Pennsylvanic were quickly followed by the Permic glaciation. Again, it may be
stated that the Pleistocene cold period was preceded in the Miocene and Pliocene
by far smaller areas of known accumulations of limestone and coal than during
either the Pennsylvanic or Cretacic, and yet a severe glacial climate followed."

• Clements, F. E., Plant Succession, Carnegie Inst. Wash. Pub. No. 242, p. 320, 1916.

' Schuchert, Chas., in Elsworth Huntington, The Climatic Factor, Carnegie Inst. Wash.
Pub. No. 192, p. 289, 1914.

* Professor Schuchert does not seem to take into consideration the fact that the formation

of normal calcium carbonate from the water-soluble acid calcium carbonate liberates an
amount of COj equal to that which it locks up. The reduction of COj in the air occurs
in times of land exposure and weathering rather than in times of limestone formation.

231

232 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Of the other theories only two may be examined with any hope of a
rational application to the late Paleozoic: the effect of volcanic dust in the
upper atmosphere in reducing the amount of solar radiation which reached
the earth, and the effect of deformation.

The first would have world-wide effect if the calculations of Abbot and
Fowle' and Humphreys^ are correct and the continuance of the conditions
cited by them and quoted in Clements, Plant Succession, pages 322-324,
would be sufficient to bring on a glacial period. But, as noted by Clements,
"the only evidence of such continuance in geological time would have to
be sought in the coincidence, or immediate sequence, of cold or cooled
climates with periods of great eruptive activity."

Schuchert^ has opposed the sufficiency of this cause, citing the fact that
the period of violent volcanic activity at the close of the Mesozoic was not
followed by a glacial but only by a "slightly cooled climate." The great
Paleozoic and Cenozoic glaciations were not in coincidence with the moun-
tain-making disturbances of those eras. Smaller disturbances localized
within periods of the eras produced a slight drop in temperature, but not
sufficient to produce glacial conditions.

On the whole Schuchert is inclined to attribute the climatic changes to
deformational causes, perhaps accentuated by volcanic dust. He concludes:'

"We may therefore conclude that volcanic dust in the isothermal region of
the earth does not appear to be a primary factor in bringing on glacial climates.
On the other hand, it can not be denied that such periodically formed blankets
against the sun's radiation may have assisted in cooling the climates during some
of the periods when the continents were highly emergent."

With this idea Huntington is apparently in agreement. The deforma-
tional theory is by far the most obvious and easily applied to the explanation
of most of the climatic changes. Evidence has been cited to show that the
eastern border of the continent of North America was being elevated in the
late Paleozoic, and this movement was but a part of the much greater move-
ment which elevated the Armorican-Variscan chains in Europe and the
Appalachian Mountains in North America. As the European portion of
this movement was accompanied by vigorous volcanic activity, it is very
possible that some late eruptions of great vigor supplied an adequate amount
of ash and that, as Clements suggests, the glacial conditions in North
America were induced by a coincidence of causes, perhaps in the order of
their importance, deformation, volcanic dust, and deficiency of CO2 in the
atmosphere.

* Abbott, C. G., and F. E. Fowle, Volcanoes and Climate, Smiths. Misc. Coll., vol. 60, No.

29. 1913-
^ Humphreys, W. J., Volcanic Dust and Other Factors in the Production of Climatic Changes,
and Their Possible Relation to Ice Ages, Mount Weather Observatory Bull., vol. 6,
No. I, 1913.

• Loc. cit., p. 287.

CLIMATOLOGY OF THE LATE PALEOZOIC 233

Schuchert has stated, as noted above, that the glacial periods of the past
have been somewhat sudden in their onset, but if we give to the expression
"glacial periods" the broader meaning which should be given it, implying
reduction of temperature with, perhaps, accompanying aridity or semi-
aridity, it is not so certain that the climatic change was a sudden one.
Certainly the change in late Paleozoic time in North America was a slow
one, and its slow advent is indicated in the increasing accumulation of red
beds, with their suggestion of alternate seasons of drought and humidity.
The climatic change from the conditions of the first half of the Pennsylvanian
to the Permo-Carboniferous, though slow and marked by local fluctuations,
was a very comprehensive one, both in character and the extent of the area
involved.

A. CLIMATE OF THE LATE PENNSYLVANIAN.

The most complete and dependable description of the climate of Penn-
sylvanian time has been given by David White,* from whose paper the follow-
ing quotations are taken :

"The extreme range of climate and the strong demarcation of the earth's
climatic zones in the present day contrast strongly with the atmospheric conditions
that appear to have prevailed during the deposition of most of the extensive
coal, even in the high latitudes. Such strongly contrasting secular climatic
changes as are shown to ha\e taken place since the deposition of peat began in
many of our actual bogs, and as are cited as arguments for possible corresp>onding
variation of climates and vegetal types during the formation of the vastly thicker
peat beds from which our coals were made, are not in the slightest d^ree indicated
in the great coal beds of the older formations, and probably never occurred unless
in the rarest and most exceptional cases, such as possibly in connection with
some of the older Gondwana coal beds of Australia, India, or South Africa. In
no part of the world to-day are the genetic conditions of the great ancient coal
formations to be found, except locally and on a relatively small scale. Topo-
graphically, climatically, and botanically their nearest semblance is to be found
in coastal and near-tide-level swamps of the South Atlantic and Gulf States, the
great estuarine and lowland swamps of India, and the lagoons cmd swamps of the
Indo-Pacific zone of hea\^ rainfall."*

This subject is further considered in connection -with a review of the
evidence as to the climate attending the great coal formations.

" Climates of the Coal-formation Periods.

"In the following pages is given a brief oudine of evidence and conclusions
as to the climates characteristic in general of the periods of the great coal forma-
tion, with particular reference to the regions of the cocil basins. It will be seen
that during the times of deposition of most of the principal coal groups the climate
has been characterized by (i) general mildness of temperature, approaching in
most cases tropical or subtropical; (2) conspicuous equability or approximation

* White, David, The Origin of Coal, Bureau of Mines Bull. 38, p. 67, 1913.

* Potoni6, H., Ein von der Hollandisch-Indischen-Sumatra Expedition entdecktes Tropen-

moor. Naturwiss. Wochenschr., Jena, Oct. 20, 1907, pp. 657-666.

234 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

to uniformity of climatic conditions, which, with a few exceptions, appear to
have lacked cold winters or severe frosts; (3) a generally high humidity, the rain-
fall being from moderately heavy to very heavy and fairly well distributed,
though in many cases there is evidence of the occurrence of dry periods which,
however, seem ordinarily to have been comparatively short and not severe; (4)
an amazingly wide geographical distribution of these genial and equable climates,
which occurred seemingly in almost uniform development simultaneously in the
high and in the low altitudes of both the northern and the southern hemispheres.
This shows either that the essentially uniform climatic conditions were truly
extraordinary in geographic extent, with little regard to modern climatic zones,
or that the formation of coal was mainly confined to the areas of the above-
prescribed climatic environment.

" Paleobotanical Criteria as to Climate.

"The principal criteria as to climate offered by the fossil plant remains
preserved either in the coal or in the enveloping shales and sandstones, and
serving as a basis for the conclusions stated above, may be summarized as follows,
further particulars being noted in the discussion of the climates of the several
most important periods of the coal formation:

" (i) Relative abundance or luxuriance and large size of terrestrial vegetation —
that is, rankness of growth, indicating favorable conditions of temperature,
humidity, etc,

" (2) Character, condition, and amount of the land-plant material preserved
as coal or carbonized in the rocks. The formation of xyloid coal of the ordinary
type, composed mainly of subaerial vascular-plant remains indicates humidity.
In regions of cool temperature the humidity required for the foimation of peat —
the initial state of coal — is moderate; in the warmer climates, where decay is
more rapid, not only must the humidity be greatly increased, in order to provide
the necessary wetness to retard decomposition, but there must be no long dry
seasons of the year for the too great reduction of the water cover. The observa-
tions of peat formation at the present day in tropical climates show that in order
to permit the decomposition of peat the rainfall must be both very heavy and
fairly well distributed through the entire year.*

" (3) Great radial distribution, seemingly over the greater part of the earth,
and especially over wide ranges of latitude, of identical species and genera in
characteristic association, indicating the extension of approximately uniform
climatic conditions in these regions. Floras identical, or essentially identical,
in remote or detached regions can owe their identity to no other cause than ap-
proximate continuity of the environment, whether that continuity is geographic
or chronographic. Conversely, the migration of a flora without change is possible
only through regions of essentially identical environmental conditions. Illus-
trations are found in the Carboniferous, Triassic, Jurassic, and lower Cretaceous
floras, and even to a remarkable degree in the upper Cretaceous and Tertiary
floras.

" ' Peat formation in the United States is not taking place, according to C. A. Davis (in
conversation with the author), in areas of less than 20 inches of rainfall. Practically
25 inches is the lower limit. Failure of peat in certain districts, such as the Pied-
mont Plateau of the South, would appear to be due in part to the occurrence of long
dry seasons; presumably the total rainfall in that region is also insufficient. Peat is
forming in Florida and in many tropical and subtropical regions of heavy and well-
distributed precipitation.

CLIMATOLOGY OF THE LATE PALEOZOIC 235

" C4) Presence of tyiies known to be adapted to or confined to the warm tem-
peratures or moist climatic conditions of the present day, types that though now
extinct once lived in association with other types of ascertained tropical or humid
habitats, and types whose descendants or nearest survi^^ng relatives are char-
acteristic of warm climates. Examples are cycadalean t>-pes in Carboniferous,
Triassic, Jurassic, Cretaceous, and finally in the Oligocene in association, since
the Trias, with li^^ng tropical and subtropical genera or families; the presence of
tree ferns in nearly all periods of coal formation ; palms, cinnamon trees, climbing
ferns, and many other tropical or subtropical tyfies in the Upper Cretaceous; and
the bread-fruit trees, etc., in the lower Tertiary.

"(5) Structures of the plants themselves. Features showing rapidity of
growth; that is, abundant rainfall, mild or warm temperatures, etc. — conditions
favorable to rapid growth :

" (a) Very leirge size of the cells, many with thin walls and lai^e intercellular
spaces, indicating rapid growth and abundant moisture, noticeable in the woods
found in and with most coal.

"(b) Large size of fronds and leaves, indicating conditions favorable to
growth and, at present, characteristic of moist tropical habitats.

" (c) Frequency of laciniate, or much-dissected, drooping fronds and pendant
branches or twigs seemingly adapted to facilitate the run-off of rain, and protec-
tion of the stomata in grooves on the under sides of many leaves, as in the
lepidophytes of the Carboniferous.'

" (d) Smoothness of bark, which is often thick, pointing toward warm, humid
swamps.

" (e) Absence of growth rings in the woods of the older coal formations,
showing climatic conditions favorable to practically uninterrupted growth, and
the absence of long dry seasons or winter frost. Such absence of rings, when
noted in £ill the associated types, plainly shows the approximation to equability
of climate.

"(/) Wide occurrence in the Paleozoic coal fields of heterospory, requiring
prevalent swamp conditions; and the occurrence of delayed fertilization and of
devices for seed flotation.

" (s) The development of subaerial roots in many of the types.

"(6) A circumstance that may be observ-ed in most coal fields in proof of
abundant rainfall at the time of coal formation is the continuity of many cocd
benches, or strata from one hollow or pan over the interx^ening shoal or sand bar
into the next pan or along the slight gradients of the base-levels, a circumstance
impossible except with sufficient rainfall to saturate the vegetal cover and main-
tain a ground- water table of retarded drainage held by the obstructing vegetation.

"(7) Two other interesting lines of e\ndence for the warm climate of the
Carboniferous are seen, as pointed out by Potonie,* in (c) the development of
more flowers and fruits on the lower parts of the stems and branches, as in Ulo-

" * The interpretation by Davis (in conversation with the author) that the so-called " pseudo-
xerophytic" feature of swamp and bog plants whose roots extend near the surface and
are normal to a wet footing are for the purpose of protection against destructive suf-
fering on occasions of drought when the water-level is usually lowered, and that they
are therefore really xerophj'tic, finds abundant support in the paleobotanical criteria
oflFered by the fossil swamps, as described in an earlier section. The pseudo xerophytic
characters appear to indicate probable subjection to occasional times of unusual evapo-
ration.

' Potonie, H., Entstehung der Steinkohle, 5th ed., 1910, p. 167.

236 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

dendron, Sigillaria, and many Calamariae, a characteristic of dense tropical
forests at the present time, and {h) the presence in many ferns of Aphlebiae,
which to-day are unknown except in tropical types.

"Evenness of Climate in Coal-forming Periods.

"It must be borne in mind that the climatic conditions here described as
having prevailed during the deposition of the groups of coal were not necessarily
conditions persisting without change from one period to another; in fact, the
characters of the successive floras and their changes are found to indicate the
occurrence of many climatic changes, some within the limits of a single peiiod
in the geologic record. To these changes in the element of the environment are
largely due the important steps in the evolution of the higher plant groups, such
as the approximate disappearance of heterospory, and the better protection of
the megaspore observed in the cycadofilices of the Carboniferous; the origin of
the dicotyledonous leaf, and of delayed germination, the former developed to
give a maximum vegetative efficiency in a growing season shortened, as indicated
by concurrent evidence by the occurrence of seasons of winter cold, the other to
enable the plant to survive several seasons that might be unfavorable for the
sprouting and successful start of the plant. The periods of coal formation have,
however, been for the most part confined to the long geologic intervals of relatively
uniform climate, the principal features of which are outlined later.

"The evidence afforded by the presence of coal in thick and extensive beds
in various regions is mostly valuable as indicating that during a long period of
time there were no wide variations of either temperature or, especially, humidity.

"But absence of coal or lignite is far from furnishing a certain basis for
conclusions as to opposite climatic conditions. The recurrent deposition of coal
of large areal extent and thickness postulates a base-level subsidence so adjusted
that at various times the necessary close relationship between water-level and
the peat-formation surface may be maintained for considerable intervals, to
permit peat deposition of the required thickness. The formation of coal (peat)
in extensive deposits (always continental) is rare in regions undergoing erosive
dissection, on the one hand, and, on the other, it is likely to fail when the sub-
sidence is too rapid or the water-level reaches the region of topographic relief
so that the coast is bold. It should be repeated that in successively warmer
climates the formation of peat requires not only a heavier rainfall, but also a
more even distribution of the same, so as to obviate the occurrence of long dry
seasons. Even then its formation is possible only by the great rapidity of plant
growth, which exceeds, under favoring circumstances, such as maintenance of
the water cover, the rate of rapid decay.

" Pennsylvania (' Upper Carboniferous ') Coal Measures.'

"Judged by the criteria outlined above, the climate of the principal coal-
forming intervals of the Pennsylvanian was mild, probably near- tropical or
subtropical, generally humid, and equable. The evidence may be outlined in
summary form as follows:

"Abundant humidity and condensation are shown by: (a) succulency of the
growth, large medullary development, and large intercellular spaces; {b) presence
of many hydathodes or water-pores on the leaves (possibly due to an aquatic
environment), and abundant lacunose tissue; (c) dissected or laciniate forms of

' White, David, loc. cil., page 74.

CLIMATOLOGY OF THE LATE PALEOZOIC 237

the leaves in many species; {d) protection of the stomata in dorsal canals, for
example, SigUlaria, as though to prevent flooding, but possibly 'pseudoxero-
phytic ' in origin ; {e) great differentiation and vast predominance of pteridophy tic
forms, including many widely varied heterosporous types to whose prolific
fertilization a very wet habitat is most essential; (/) smooth, hard, persistent
outer bark; {g) adaptation of nearly all t>'pes to fertilization in a rainy habitat,
such as protection of pollination and probable flotation of immature seeds; (A)
flotation devices possibly peculiar to swamp types; (*') prevalence of great
swamps on coastal or inland base-level peneplains, and of great amounts of un-
decayed vegetal matter remaining either in stratified masses as coal or in the
carbonaceous shales and other terrigenous dep>osits; (j) formation of much
xyloid coal, requiring abundant humidity in a climate of mild temperature.

"That the climate was warm is further shown by : (a) the rank, luxuriant growth
and large size of the plants, especially of cryptogamous types; {b) the rapid, succu-
lent growth, with large cells; (c) the dense, large undergrowth; {d) the many
long climbing or clambering filicoid types, including a considerable number of
membranaceous delicate forms; (e) the present tropical habitats of living repre-
sentatives {e. g., Marratiaceae and Gleicheniacese) nearest related to Carboniferous
types, whereas the habitats of the Coal Measures cycad stock are tropical or
subtropical; (/) the attainment of gigantic size by the equisetales (so common
and highly differentiated in the Coal Measures, and to-day always growing in
moist ground), only under warm or mild equable conditions.

"Additional evidence of high importance is found in the absence of growth
rings; that is, continuity of growth, indicating absence of winter frosts or of
long or se\'erely dry seasons unfavorable for the vegetative process. Equability
also may be predicated on the seemingly almost worldwide range of mildness of
climate.

" Proof of relative uniformity in climate is based mainly on the extraordinary
radial distribution of identical species in all lands, even in high latitudes. Prac-
tically entire floras spread over the earth, crossing the equator, seemingly without,
so far 35 noted, experiencing seriously obstructional climatic differences or seasonal
changes. Minor differences between the floras of certain regions, for example,
the coastal districts and the inland fresh-water basins, and the absence of certain
genera or species from one continent or another, have been noted by Gothan*
and White,' but as between continents, the comparatively uniform distribution
of the floras, although less than is sometimes stated, is a remarkable feature of
the period, being most nearly comparable to that of the mid-Jurassic.

"Many identical Westphalian species, including a number characterized by
short duration, range across the coal fields of North America and Europe to
Persia and China, and a few occur in the Arctic Zone, in South Africa, and in
Argentina. It is probable that during "Lxjwer" and "Middle" Coal Measures
(Pottsville) time, at least, no zone of torrid heat, as contrasted with polar tem-
peratures, existed in the equatorial regions. On the other hamd, it is probable
that over most of the earth, at least outside of the polar circles, the temperatures

* Gothan, W., Weiteres Ober floristische Differenzen (Lokalfarbungen) in der europaischen

Carbonflora, Zeitschr. Deutsch. geol. Gesell., voL 6i, No. 7, 1909, pp. 313-325;
Pflanzen-geographisches aus der paleozoischen Flora, Zeitschr. Deutsch. geol. Gesell.,
vol.59, 1907. P- 150.

* White, David, The Upper Paleozoic Floras, Their Succession and Range, Jour. Geol.,

vol. 17, 1909, p. 328.

238 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

did not vary greatly from region to region, and that they were nowhere torrid,
possibly not even fully tropical as a whole. Yet the action of breezes, so well
demonstrated by the abundant ripple-marks on the sands of the Coal Measures, is
also indicated by the equipment of many seeds and spores with wings (although
the latter may have been for gliding), and, more particularly, by the separation
of the male and female flowers in most of the flowering plants.

"The action of sunlight may be inferred from the presence of palisade cells
to shade the mesophyl, the horizontal attitude of the leaves, and the rapid growth.

"PoTTSviLLE Time.

"Of the Pennsylvanian floras, those of the Pottsville and Allegheny time are
perhaps widest spread in relative entirety, though the flora of the upper Cone-
maugh and Monongahela have nearly equal homogeneity in migration. The
fact that these floras differ markedly, though the changes are somewhat transi-
tional, especially between the last-named stages — the fact that plant life changed,
differentiating, eliminating, and adding types — is probably due not merely to
kinetic evolution and exterminative competition; it was undoubtedly due in a
large part to changes in the climates as well as in other environmental elements.
As to the degree of the climatic change we have little knowledge, but, as is later
suggested, they were probably of relatively small magnitude during this interval.

"Allegheny Time.

"As already suggested, it is possible that the maximum uniformity of climate
occurred in the upper Pottsville. In the Appalachian trough this was perhaps
the period of greatest and most evenly distributed rainfall. In the Rocky Moun-
tains some coal was laid down at this time. The Allegheny, which includes the
topmost Westphalian, is marked by the disappearance of many of the climbing
and clambering types, whereas the membranaceous and laciniate-leaved forms
are much rarer, and the pinnules of the filicoid types are growing larger. Coal
formation, which seems to have occurred wherever the adjustment of topography
and water-level was favorable, appears, however, to have been general. The
woods show no trace whatever of seasonal interruption of growth, and the con-
clusion that there was no winter frost to cause a periodic stage of arrest of growth
seems well founded. That there were, however, times when during certain
seasons, possibly exceptional or extraordinary, the water-level was reduced, is
nevertheless indicated by the increasing development of pseudoxerophytic char-
acters. As they later become more prominent in the Conemaugh and Permian,
they are discussed in connection with the paragraphs referring to those periods.

"Conemaugh Time.

"The Conemaugh (of lower Stephanian age) time witnessed several changes
in the floras which may be of climatic cause. Most prominent among these are
a rapid decrease, approaching extinction, of the colossal lycopods (Lepidoden-
dreae), and the rapid development of the group of gigantic tree ferns, such as
Psaronius, whose supposed fronds, Pecopteris, became highly varied, very large,
and more or less distinctly villous in most species. The evidence therefore
points to the occurrence of short, dry seasons. The reduction in the lepidophytes
is attributable to occasional unusual failures or disappearance of the water in
which their spores must fall in order to insure reproduction of the species. Pro-
vision for spells of unusual evaporation may account also for the tremendously
thick bark of the Psaronii, with their abundant intracortical ramentum; for the

CLIBIATOLOGY OF THE LATE PALEOZOIC 239

protection of the stomata on the leaves of SigiMaria; and for the water-storage
tissue in the trunk of Lepidodendron — all swamp types.

"As the Conemaugh is apt to be marked by the deposition of thick red beds,
especially in the Appalachian trough, it would at first seem that the characters
of the flora only confirm the explanation that redness resulted from aridity.
However, in opposition to such a conclusion it must be noted that (a) the flora
in the red beds has not been observed to differ very markedly from that in the
regions of dark contemf>oraneous sediments, including coal ; (6) the plcmts, though
less varied, are not reduced in size, nor possibly, in number; (c) coal, usually
thin, to be sure, occurs in the midst of the red beds of the Conemaugh both in the
eastern and the Rocky Mountain regions of America, cis well as in Europe, some
of the coal being thick; {d) the e\adence of seasonal growth ('annual rings') in
the Conemaugh woods yet examined is slight, though the rings are a little more
distinct than in the Allegheny woods; (e) the great calamitean growth appears
unimpeded, though many of the giant species are provided with thick xylem
and cortex; (/) the nearest living relatives of the Psaronii, the Marratiaceae,
are now exclusively tropical. In view of these facts it is evident that in the
Conemaugh the climate was still mild and practically free from frost, and that the
rainfall was at times certainly ample for the production of peat under fresh-water
conditions. The absence of coal at other levels may, as in other series, be largely
due to lack of proper adjustment of bottom and water-level.

"Granting that the Calamariae were swamp plants, and that many other
types, including the enormous Sigillariae, were also inhabitants of the marsh, it
still remains obvious that aridity was not sufficiently developed to dry up all
the swamps. Bearing in mind also the protection of the stomata in the lycopods
and Calamariae, and the villous development of the Pecopteris species, the water-
storage equipment of many of these t>'pes, including the tree ferns, the develop-
ment of waxy covers in some Neuropteris species, and the presence of resin canals
and other secretory cells in many of the ferns, c>'cadofilices, lycopods, and
Calamariae, or, in fact, that the g>'mnospermous woods were perhaps of kinds less
affected by dry seasons, it nevertheless is evident that in general the rainfall was,
at certain stages at least, sufficient, even at warm temperatures, to permit coal
formation over great areas in nearly every coal field. It seems not improbable
that during most of Conemaugh time the total rainfall was distinctly less than in
Allegheny time, and it is possible that for brief intervals the climate may
have been dry as compared to the latter; but it appears improbable that during
ordinary red-bed deposition — say, in the Appalachian trough — the climate even
closely approached aridity as that term is employed in reference to present-day
conditions. Such intervals must have wrought greater changes, more sweeping
extinctions in the flora. From the paleobotanical standpoint the widely current
belief that aridity in the actual sense is to be assumed as causally and almost
indispensably associated with red-bed deposition is not well founded. The
generalization seems to be too broad and too sweepingly applied. In this connec-
tion it may be noted that although in the Broad Top and the Potomac basins
red beds are almost absent, redness of color extends diagonally downward into
the top of the Allegheny, which is still coal-bearing, near the Kentucky-West
Virginia line in the Kenova quadrangle. Withal it must not be forgotten that
the xerophytic characters are, perhaps, confined to aquatic or swamp plants, and
that the protections are exactly those adopted by bog plants of today to insure
against exposure to too great loss of water, or to increased toxicity of the sub-
stratum resulting from an unusual reduction of the water cover.

240 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

" It may, however, be noted that in the red-bed regions coal formation usually
falls off in the thickness if not in the number of the beds; on the other hand, the
greatest coal formation of the Conemaugh occurs in general in the districts of
least red-bed deposition. We may therefore infer either that the conditions
favorable for redness in such cases involve relations of topography, water-level,
and epirogenic movement that are less favorable to thick and repeated fresh-
water peat formation or that a drier state of the land surrounding the swamp
not only caused a less rank growth of plant life, but also, by reducing the run-off
through jungles formerly tropical in density, permitted smaller burdens of nearly
pure vegetal matter to add to that growing in the somewhat reduced swamps.
It is possible that in certain regions comparative aridity prevailed for restricted
periods during which the floras found a not too distant friendly refuge, from which
they returned with the resumption of favorable conditions without too great loss
or changes.

"MONONGAHELA TiME.

"Monongahela (Upper Stephanian) time is marked in many regions of the
earth by conditions approaching in some respects those of the Allegheny. The
Appalachian, as well as the contemporary beds in western Europe, eastern Asia
(Manchuria and China), and southeastern Africa, are nearly everywhere marked
by heavy deposition of coal.

"Among the notable paleobotanical characters of this stage are (a) the
presence of a waning group of thick-barked Sigillariae, probably confined abso-
lutely to swamps; (b) large and abundant Psaronius tree ferns; (c) increasing
size of Calamites, reinforced by increased wood development; {d) high differentia-
tion of types of seeds, that is, expansion and differentiation of the seed-bearing
habit, as the great spore-bearing types were eliminated, probably in consequence
of occasional seasons of unusual dryness; {e) increase of features, possibly xero-
phyllous, especially in the surviving and new cycadofilices, including Noeggerathia
and Dolerophyllum, though most of these are presumably exclusively swamp
plants; (/) first appearance of fronds of distinctly cycadaceous aspect {Ptero-
phyllum, Sphenozamites, Plagiozamites) . The nearest living relatives of these,
as well as of the tree ferns, are tropical or subtropical, and though many of them
are accustomed to survive dry seasons, they are characteristic evidence against
winter frost. Testimony against winter cold and prolonged seasons of drought,
such as would ordinarily prevent peat (coal) formation, especially in a warm
climate, is also found in the stage of development of seasonal rings in the wood,
the latter being usually slight and sometimes very indistinct, thus showing that
the period of interruption or retardation of growth was usually of short duration.
Both the survival of warm-climate types and the evidence for relatively short
periods unfavorable to growth argue against winter seasons of frost during this time.

"On the whole, the paleobotanical inferences are that during Monongahela
time the climate was mild, probably subtropical, and nearly uniform over the
greater part of the earth, as shown by the geographic distribution of the types
that were able to extend in relative purity of association of identical species^
around the world from east to west, and from the latitude of England and Man-

' Seven of the eight species described by M. Zalessky (Verb. Russ. K. Min. Ges., vol. 42,
1905. pp. 485-508) from the mines at Jantai in Manchuria are also present in western
Europe, six of them being present in America also. All of the eleven species reported
by R. Zeiller (Ann. des Mines, vol. 4, 1883, pp. 594-598) from Tete on the Zambesi are
present in Europe, and nine or ten of them are also found in the Appalachian trough.

CLIMATOLOGY OF THE LATE PALEOZOIC 241

churia on the north to southeastern Africa on the south. Though the total
rainfall seems to have been heaN^y, probably over 50 inches, the climate was
seemingly marked by short seasons of drj-ness, but not of winter frost in any
region that has yet furnished fossil plants. The prev-alence of great swamps in
all areas where littoral deposits are known, and in all continental fresh-water
basins, combats the idea of aridit>'. Though the Monongahela formation con-
tains some red beds in most countries, and though it may have covered short
interxaJs of dryness greater than at other times, it is improbable that desert
conditions prevailed in these regions at any time during the period. The existence
of well-defined climatic provinces at this time may well be doubted.

"Perjoan Coal Measckes.

"The rapid changes in the flora of the Permian indicate corresponding climatic
changes, and these in turn suggest the differentiation of climatic zones, which at
an early stage are reflected in the conspicuous development of at least two great
climatic provinces. In northwestern Europe and eastern America, where coals
are not rare in the Permian, the climatic changes were less marked than in western
America, and particularly in southern Asia and the southern hemisphere, where
for a time, presumably in the early Permian, glacial conditions are known to
have occurred on a scale far greater than that of Pleistocene glaciation in the
northern hemisphere. Consequent to the climatic developments in the broad
regions of refrigeration in southern South America, South Africa, India, and
Australia, there was develof>ed a peculiar flora, consisting of a limited number of
types, which is known as the Gangamopteris (so-called Glossopteris) flora. Al-
though it is practically certain that the extermination of the old mild-climate
cosmop>olitan plant life from the Gangamopteris floral province was due to climatic
rigor, it is not, however, certain that the Gangamopteris flora itself lived under
conditions of actual climatic severity, though probably some of its types were
able to endure marked seasonal changes.^ Hence it does not necessarily follow
that the coal beds dep>osited in the Gangamopteris proxnnce during its occupation
by this flora were formed in a mean low temperature, though the prevailing climate
may have been colder than that in the north at the same time and may have
been marked by colder T^-inters. The maintenance of the distinctions between
the Gangamopteris and the cosmopolitan floral provinces during the lower Permian
was perhaps due to topographic, marine, or other conditions causing isolation."

In connection with this discussion of the climatic conditions of late
Paleozoic by David Wliite, it is of interest to quote some remarks from an
earlier paper by the same author:

"The Stephanian or Ouralian (including the Gschellian) of Europe dates
from the Hercynian uplift. Prior to this movement the sea had reached its
maximum extension in the coal fields of the northern hemisphere. The Hercynian
thrust caused its practical expulsion from the old synclines of western Europe
and the creation, especially to the southward, of new bsisins, mostly of fresh or
brackish water, to which were transferred the scenes of coal-formation. In

* See \Miite, David, The Upper Paleozoic Floras, Their Succession and Range, Jour. Geol.,
vol. 17, p. 320, 1909.

242 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

America the line between the Westphalian and the Stephanian is not yet accu-
rately drawn, the fossil floras being not studied in sufficient detail. In view,
however, of the paleobotanical evidence indicative of a point near the Allegheny-
Conemaugh boundary, I, personally, am inclined to regard the formation of the
Mahoning sandstone (conglomeratic), the changed sedimentation of the Cone-
maugh formation, the probable upwarp of the southern Appalachian region which
later resulted in the exclusion of the sea from the northern area also, and the
consequent climatic changes, as due to the same great orogenic influence. * * *

"It is clear that the new elements of our Stephanian flora are chiefly, at
least, of European origin, the plant life there having been directly influenced by
the important physical changes to which it was immediately subjected. The
various exotic types migrated to North America, probably, along or near the
general route traversed by their Westphalian predecessors. Also, since the
Stephanian flora of the American basins seems to aff^ord no evidence of a rapid
or strongly pronounced climatic alteration, it becomes fairly probable that the
more abrupt plant changes described in western Europe were induced chiefly
by the sweeping orogenic effects of the Hercynian movement, rather than by a
great climatic change of world-wide extent. This does not, however, preclude a
moderate but far-reaching modification of climate, in which changes in the atmos-
pheric composition may have played a subtle if not important part. It seems
hardly possible that the tremendous amounts of carbon then being stored away
in the coal fields as the result of plant extraction from the air could have failed
to produce some effect on the atmospheric content of CO2. * * *

"The coming of the Permian is characterized not only by orogenic movements
in the eastern hemisphere, but also by indications of increasing climatic diff^erences.
The first paleobotanical eff^ect of these is the extinction of nearly all characteristic
Carboniferous types, except in Pecopteris, Cordaites, and Neuropteris, the latter,
however, disappearing nearly completely by the close of the Autunian or lower
stage. They are replaced by varied forms of Callipteris, the Ungulate Odontopteris,
and the ribbon-like Tceniopteris, together with expanding gymnospermous types,
such as Walchia, Dicranophyllum, Doleropteris, Psygmophyllum, and Ginkgo-
phyllum. Later, in the Scixonian, or Middle Permian, Voltzia, with the thick-
leaved Equisetites, appears, while more of the older types go out; and in the Thur-
ingian, or Zechstein (Upper Permian), Rhipidopsis, Araucarites, Gomphostrobus,
Voltzia, and Ullmannia become the characteristic genera, while Pecopteris, domi-
nant in the Stephanian, has nearly vanished. Though lacking the abundant
Cycad and Cladophlehis-Asterocarpus elements, the Upper Permian is in many
respects transitional to the older Mesozoic flora. * * *"

" ' Permo-Carboniferous Climates.'

'^Climate of the Carboniferous. — The climate of the Pennsylvanian ('Upper
Carboniferous') as viewed in perspective was mild and relatively humid, and,
above all, equable over the greater part of the earth. It was moderate in tem-
perature, not tropical, possibly not even subtropical, but, during the Westphalian
at least, always and everywhere equable. It was truly temperate. The criteria
which may be interpreted in support of this generally accepted proposition
include:

"l. The tremendous size and great height of the types, and their rank foliar
development, indicating favorable conditions of environment and vigorous
nutrition.

CLIMATOLOGY OF THE LATE PALEOZOIC 243

"2. The succulent nature of many of the forms, the Izirge size of the vessels
and cells, and the relatively great proportion of soft tissue, ail indicating rapidity
of grow^th in a moist, mild climate.

"3. Spong>' leaves suggestive of a moist atmosphere, and abundant and
large intercellular spaces, as in the lycopods, pointing to rapid moisture-loss;
also water-pores for disposal of excess of moisture.

"4. Stomata placed in grooves, as in the lycopods, as if to prevent obstruction
by falling rain.

"5. Absence of annual rings in the woods; hence absence of marked seasonal
changes.

"6. The analogies of the present day show [that^ aerial roots, so prominent
in many of the Carboniferous types, are characteristic of moist and tropical
climates; that the nutrition — ». e., the decomposition of COj — is most rapid
and the consequent growth also greatest and most rapid where the light is not too
strong; that the ferns and lycopods, so abundant in the Pcileozoic, usually avoid
bright glare. The same types are able to withstand larger amounts of CO* with
benefit to themselves.

"7. The nearest living relatives of the Paleozoic vascular cryptogams reach
their greatest size in humid and mild or warm climates. The successors of the
marratiaceous and gATnnospermous typ)es are now mostly confined to tropical
or subtropical regions. The cycadalean stock, now characteristic of the same
zones, was actually present in the upp>er Coal Measures.

"8. The formation of great amounts of coal indicates a rank growth, but in a
temperature not so warm eis to promote decay beyond the limit of rainfall
protection.

"9. Living nearest representatives of Paleozoic fishes now inhabit the estuaries
of warm countries; while the nearest relatives of the Carboniferous insects are
now found in mild and moist habitats.

" 10. The most forcible argument, after all, for an equable and uniform climate
lies in the extraordinar>' geographical distribution of the floras in relative unity
over the face of the earth. Humidity must naturally have attended such equa-
bility', extending, without distinct terrestrial climatic zones, possibly completely
into the polar regions.

"Some of the criteria above mentioned are susceptible of different inter-
pretations; but taken collectively they appear to admit of but one conclusion.
\\Tiether or not we admit that climatic changes may be caused by reasonable or
practicable changes in the amount of carbonic-acid gas in the air, it is certain that
in geological times the vegetation of the earth must have been more or less in-
fluenced by the constitution of the atmosphere from which the plant derives
so important a part of its real food. * * *

"As has already been indicated, the Westphalian probably witnessed the
greatest extension of uniformity and equabilitj' of climate over the earth. In-
the Stephanian the flora is hardly so homogeneous, though the world-climate
appears still to have been so equable as to allow free migration of the larger part
of the flora from a moderate latitude on one side of the equator to the opposite
without encountering seriously obstructive seasonal changes. In the Permian
the regional distinctions between the floras are much clearer; and presently
climatic zones, and consequently botanical provinces, are recognized. Yet, about
the North Atlantic the climate of the Lower Permian was still relatively uniform,
so that moderately free migration of the floras without the development, so far

244 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

as we know, of pronounced annual rings, took place in the Autunian of France,
the Permian of Prince Edward Island, the Dunkard of southwestern Pennsyl-
vanian, the Chase of Kansas, and the Wichita of Texas."

David White cites the fact that the plants of the Red Beds are not
essentially different from those of the gray shales and the limestone beds in
Pennsylvania and West Virginia and questions whether the red color means
anything in particular as to the aridity of the climate. He says:

"It is probable that there was aridity in certain regions and during certain
intervals of the Permian; but there was evidently enough moisture to produce
most extensive glaciation, and, later, to promote the formation of coals over broad
areas in the great fresh-water Gondwana series laid down on the continents of
South America, Africa, and Asia."

It is well to recall here a note to David White's discussion of the physiog-
raphy of the coal basin :

"The rapid decrease, almost amounting to disappearance,' of the great number
of the very large spored lycopods during Conemaugh time (early Stephanian)
was no doubt due to failure of fructification caused by periods of relative drought
and reduction of the water-surface, such withdrawal of the water being plainly
indicated by the prevalent pseudoxerophytic characters observed in the swamp
plants of the period."

B. CLIMATE OF THE PERMO-CARBONIFEROUS.

The presence of the red beds across the continent, of the Roxbury tillite,
and the New Glasgow conglomerate are all inorganic evidence of a most
decided and extensive alteration in climate. Even as far west and south as
Oklahoma there is some suggestion of a possible decrease in temperature
sufficient to form ice of at least local extent.

TafP in 1909 reported the presence of a bowlder bed in the Caney shale
in the Wichita Mountains, Oklahoma, in which the individual bowlders bear
grooves and striae which he attributed to ice action and the accumulation
of the bowlders to floating ice. Ulrich,' in 1911, stated his concurrence in
this view. "No other competent means of their transportation than ice —
presumably heavy shore ice — has been suggested."

This region was visited by Woodworth later, and he also expressed his
concurrence with Tafif's view,'* that the bowlders and smaller stones of the
Caney shale have been transported by some sort of ice action :

"Floating ice is naturally suggested as the probable agency, notwithstanding
that to have pan-ice at sea-level demands a greater degree of cold in this latitude

' White, David, Origin of Coal, Bureau of Mines Bull. 38, note b, p. 56, 1913.

' Taff, J. A., Ice-borne Boulder Deposits in Mid-Carboniferous Marine Shales, Bull. Geol.

Soc. Amer., vol. 20, p. 701, 1909.
' Ulrich, E. O., Revision of the Paleozoic Systems, Bull. Geol. Soc. Amer., vol. 22, p. 352,

footnote, 191 1.
* Woodworth, J. B., Boulder Beds of the Caney Shales at Talihina, Oklahoma, Bull. Geol.

Soc. Amer., vol. 23, p. 461, 1912.

CLIMATOLOGY OF THE LATE PALEOZOIC 245

than would be demanded for floating detached portions of mountain or plateau
glaciers entering the sea in their zone of melting."

[The abundant evidence of Permian glaciation, etc., make it] "quite as reason-
able to supfKDse that ice formed on the fresh waters of the Carboniferous." [Also,
he says of a bed near the top of the Roxbury conglomerate:] "This presumably
tillite bed is possibly of Permian age, but its association with the underlying con-
glomerates and similar thick, water-worn conglomerates of known Carboniferous
(Allegheny) age in the Xarragansett area points to the correctness of Shaler's
theory of the glacial origin of the conglomerates as a whole."

The western instance is stUl an isolated one and may hardly yet be con-
sidered as definite proof of such low temperatures as are suggested by the
authors cited.

Other evidence of a decided lowering of the temperature in Permo-
Carboniferous time is furnished by the change in life. David WTiite, in the
paper quoted (page 238), has noted the "rapid decrease, approaching
extinction, of the colossal lycopods (Lepidodendriae) , and the rapid develop-
ment of the group of gigantic tree ferns, such as Psaronius" with minor
changes; all of which he regards as insufficient to indicate a great climatic
change, but are clearly the beginning steps of what came later. Similarly
slight but progressive changes are noted in Monongahela time, which,
however, David \Miite holds are still so slight that they indicate a moist
and warm climate with absence of killing frosts or long-continued periods
of aridity. In Dunkard time, however, the change was rapid and resulted
in the formation of distinct climatic provinces. WTiite does not believe
that the eastern portion of the United States was extensively affected because
of the continuation of the coal during the Permo-Carboniferous.

Aside from the change in the composition of the flora, the physiological
adaptations have been used in an interpretation of climatic conditions.
Xerophilous adaptations have been found in many plants of the late Paleo-
zoic, and these have been interpreted as indicating a decided increase in
aridity, but it is evident from the work of the botanists that xeromorphy in
plants is attributable to more than one cause and is still incapable of exact
interpretation. The occurrence of such structures in bog plants is well
known, but its meaning is still in dispute. The matter has been review^ed
b^- Clements,^ and without entering at length into the matter it may be
stated that it has been suggested that the effect of decomposition in stagnant
bogs and swamps is to produce both acidity and toxic products. It is
assumed by some that the presence of one or both of these acts as a deterrent
to root groui;h or functional activity and the xerophytic structure is a
response to the inability of the plants to obtain an adequate supply of
physiologically wholesome water. The literature of this matter may be
followed from Clements's summary and from Dachnowski.*

•Clements, F. E., Plant Succession: An Analysis of the Development of Vegetation,

Carnegie Inst. Wash. Pub. No. 242, p. 90, 1916.
* Dachnowski, A., The Problem of Xeromorphy in the Vegetation of the Carboniferous

Period, Amer. Jour. Sci., 4th series, vol. 32, p. 33, 191 1.

246 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

The interpretation of conditions from the presence of plants showing
xeromorphic structures, in Pennsylvanian or Permo-Carboniferous time, is
thus rendered very uncertain. Especially in the Permo-Carboniferous time,
with its continuous approach to aridity, the interpretation is complicated by
the evident close juxtaposition of a great variety of plant habitats. The need
for caution in this particular is well illustrated in Spalding's discussion of
the distribution of plants in an arid habitat.^ He shows clearly how purely
aquatic plants assembled along water-courses may exist in close association
with a purely desert vegetation, and it is evident that in such chance accumu-
lations as might easily occur the fossilized remains might lead to very con-
fusing and erroneous results if not correctly interpreted.

In this connection, David White says:^

"Irregular temporary reductions or withdrawals of the water cover, possibly
seasonal or perhaps less frequent, are, in the writer's judgment, causally related
to the ordinary type of lamination of much of our coal, and the sheeting of the
latter by fragments of ' mineral charcoal ' (' mother of coal '). To their occurrence
is probably due also the development of xerophytic and water-storage devices for
the protection of so many of the coal plants of the Carboniferous swamps. Such
periods of water reduction and evaporation appear generally to have been attended
by concentration of the hydrocarbon solutes resulting from the putrefaction
process in the form of paste, which now constitutes the jetlike 'binder' of the
coal. Conversely, the alternate periods of rise of the water-level and the attend-
ant dilution of the water cover favored to some extent not only the extraction,
and, in cases of flushing, the removal of some of the putrefaction products from
the upper part of the peat-forming debris, but also promoted the oxygenation
and, consequently, the revival of decay wherever the asepticity was neutralized."

Turning to another phase of biologic evidence, Schuchert^ says:

"A climatic change naturally must affect the land life more quickly and pro-
foundly than that of the marine waters, for the oceanic areas have stored in them-
selves a vast amount of warmth that is carried everywhere by the currents.
The temperature of the ocean is more or less altered by the changes of climate,
be they of latitude or of glaciation. The surface temperatures in the temperate
and tropical regions, however, are the last to be affected, and only change when
all of the oceanic deeps have been filled with the sinking cold waters brought
there by the currents flowing from the glaciated area. We therefore find that
the marine life of earlier Permic time was very much like that of the Coal Meas-
ures, and that it was not profoundly altered even in the temperate zones of Middle
Permic time (Zechstein and Salt Range faunas). Our knowledge of Upper
Permic marine life is as yet very limited and will probably always remain so
because of the world-wide subtraction of the seas from the lands at that time.
It was a period of continued arid climates, and the marginal shallow sea pans

'Spalding, V. M., Present Problems of Plant Ecology: Problems of Local Distribution in
Arid Regions, Amer. Nat., vol. 43, 1909. Reprinted in Annual Report Secretary Smith-
sonian Institution for 1909, p. 453.

' White, David, Origin of Coal, Bureau of Mines Bull. 38, p. 64, 1913.

' Schuchert, Chas., Climates of Geologic Time, in Huntington, The Climatic Factor as
Illustrated in Arid America, Carnegie Inst. Wash. Pub. No. 192, p. 279, 1914.

CXIMATOLOGY OF THE LATE PALEOZOIC 247

were, as a rule, depositing red formations with gypsum, and locally, as In northern
Germany, alternations of salt with anhydrite or polyhalite in thicknesses up to
3,395 feet. In certaiin of these zones there were developed annual rings so
regular in sequence as to lead to the inference that they were the depositions of
warm summers and cold winters, enduring for at least 5,653 years (Gorgey, 191 1)."

With reference to the insect life, Schuchert says:^

"The very large insects of the Coal Measures tell the same climatic story,
for Handlirsch says that the cockroaches of that time were as long as a finger
and the libellids as long as an arm. They were 'brutal robbers' and scavengers
living in a tropical and subtropical climate, or at the very least in a mild climate
devoid of frosts. We therefore conclude that after Middle Devonic time the
climate of the world was as a rule uniformly warm and more or less humid and
that it remained so to the close of Upper Carbonic time. * * *"

[In Permian time] "the grand cosmopolitan swamp floras of the Upper
Carbonic, consisting in the main of spore-bearing plants, such as the horsetails
(equisetales), the running pines, and club-mosses (lycopodiales), and the ferns,
among which were also many broad-leaved evergreens (cordaites) and seed-
bearing ferns (cycadofilices), were very largely exterminated in the southern
hemisphere at the beginning of Permic time. In the northern hemisphere, how-
ever, the older flora maintained itself for a while longer, as best seen in North
America, but finally the full effects of the cooled and glacial climates were felt
everywhere. Then in later Permic time the old floras completely vanished,
except the hardier pecopterids, cycads, and conifers of the northern hemisphere,
and with these latter mingled the migrants from the hardy Gangamopteris flora
originating in the glacial climate of the southern hemisphere. Some of the trees
show distinct annual growth rings, and hence the presence of winters. It was
these woody floras that gave rise to the cosmopoUtan floras of early Mesozoic
time.

"With the vanishing of the cosmopolitan coal floras also went nearly all of the
Paleozoic insect world of large size and direct development, for the insects of late
Permic time were small and prophetic of modem forms. Then, too, they all
passed through a metamorphic stage indicating, according to Handlirsch, that
the insects of earlier Permic time had learned how to hibernate through the winters
in the newly originated lar\-al conditions."

The change in land vertebrate life has been repeatedly demonstrated
and the difficulties of interpretation caused by the discovery of generically
identical forms as low as Middle Conemaugh and as high as the Clear Fork
are removed by the f>osition taken in this paper of a migration of the habitat
from east to west, favorable conditions appearing at later and later intervals
and at stratigraphically higher levels as the development of Permo-Carbon-
iferous conditions is traced to the west.

The significance of "red beds" has been the subject of discussion for a
long time, and it now seems to be fairly well accepted that such deposits,
devoid of accompanying salt and gypsum beds, or vdth only a small per-
centage of such beds, is the result of a climate wath moderate rciinfall occur-

* Schuchert, Chas., Climates of Geologic Time, in Huntington, The Climatic Factors as
Illustrated in Arid America, Carnegie Inst. Wash. Pub. No. 192, p. 278.

248 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

ring in rainy seasons alternating with dry seasons. Barrell says* of the
significance of the appearance of red sediments:

"Turning to the climatic significance of red, it would therefore appear both
from theoretical considerations and geological observations that the chief condi-
tion for the formation of red shales and sandstones is merely the alternation of
seasons of warmth and dryness with seasons of flood, by means of which hydra-
tion, but especially oxidation of the ferruginous material, in the flood-plain
deposits is accomplished. This supplements the decomposition at the source
and that which takes place in the long transportation and great wear to which
the larger rivers subject the detritus rolled along their beds. The annual wetting,
drying, and oxidation not only decompose the original iron minerals, but com-
pletely remove all traces of carbon. If this conclusion be correct, red shales or
sandstones, as distinct from red mud and sand, may originate under inter-
mittently rainy, subarid, or arid climates without any close relation to tempera-
ture and typically as fluvial and pluvial deposits upon the land, though to a
limited extent as fluviatile sediments coming to rest upon the bottom of the
shallow sea. The origin of such sediment is most favored by climates which are
hot and alternately wet and dry as opposed to climates which are either con-
stantly cool or constantly wet or constantly dry."

The question of the greatest climatic significance in the constitution of
the red beds of Permo-Carboniferous time is whether the color is original or
has been produced secondarily by a dehydration of hydrated oxides of iron
by pressure or chemical change. That the color is a primary one due to
deposition of the ferric oxide as such with the sandstones and shales is
fairly certain. Observations by Case and by Baker have been reported
upon this subject.

Case'^ says of the beds in Texas and Oklahoma:

"We may be certain that the red clays of Texas, with their ferric oxide, were
deposited in the sea, or other bodies of water, in the condition in which they now
occur, and are not due to subsequent dehydration or decarbonation, because
(i) the color is uniform throughout; (2) because there is a solidity and density
in the clays, and a lack of filled seams and veins, which would be impossible
after such changes, which involve a decided increase in volume ; and (3) because
the red color transgresses into the limestones and sandstones with marine fossils."

Baker, ^ writing of the same region, found that the following facts
indicated a non-arid condition of ' Red Bed ' origin :

"l. 'Red Beds' are not being formed to-day in any desert region. On the
contrary, they are being formed under conditions of warm, moist climates in the
southern temperate and tropical regions, as maturely weathered residual soils.
They are being formed, for example, in such regions as the southeast Texas Gulf
Coastal Plain, the Great Valley of the Southern Appalachians, and as laterite in
the subtropical and tropical regions.

' Barrell, Joseph, Relations Between Climate and Terrestrial Deposits, Jour. Geol., vol.
XVI, p. 292, 1908.

• Case, E. C, The Permo-Carboniferous Red Beds of North America and their Verte-
brate Fauna, Carnegie Inst. Wash. Pub. No. 207, p. 43, 191 5.

' Baker, C. L., Origin of Red Beds, University of Texas Bull. 29, p. 3, 1916.

CLIMATOLOGY OF THE LATE PALEOZOIC 249

"2. The plant fossils' in the Wichita red beds of Texas and some other 'Red
Beds' show no xerophytic adaptations. On the contrary ' red beds ' are associated
with coal deposits in various parts of the world.

"3. The amphibians and reptiles of Wichita time and the vertebrate fossils
in some other 'red beds' were land animals which lived part of the time in water,
part of the time on land. They did not live in a desert environment.

"But the presence of widespread deposits of salt and gypsum, of practically
incontestable sedimentary origin, contemporaneous with the red clays, seemed
to indicate conditions of aridity in at least later Permian time in Texas and else-
where. Therefore, two working hypotheses were formulated to account for the
conditions: (i) That the 'red bed' sediments were not originally red, but had
in some way been changed to a red color subsequent to their deposition ; (2) that
the 'red beds' associated with the salt and gypsum were derived from old residual
soils of moist warm climates, transported and deposited without change of color
in the arid basins of the later Permian.

"At the time of the original investigation it seemed impossible to make a
definite choice between these two hypotheses, and so the further investigation
was held in abeyance. In more recent years, the examinations of samples from
deep borings in the 'red beds' of Texas by Dr. J. A. Udden,* has demonstrated
the persistence of the red color in depth. So, although in some instances the red
color may be secondary, in the red beds of Texas it is almost certainly primary,
i. e., contemporaneous with the deposition of the sediments. Later, a re-examina-
tion in the light of late evidence of later Pennsylvanian and Permian geologic
history of Texas has shown that the second hypotheses can be consistently
advanced as the solution of the problem."

The lateral transition from light-colored sediments into red beds, typic-
ally along the Kansas-Oklahoma line, so often repeated, is also a point in
evidence.

The very names used by well-drillers for certain horizons in Pennsylvania
and West Virginia, as "deep red" shows that the color holds in the east and
is original. Another convincing evidence of the original red color of the
shales and sandstones has been noted by the author wherever he has seen
red beds of Permo-Carboniferous age, from Prince Edward Island to Arizona.
The red shales are frequently mottled by light blue or green dots, circular in
section and evidently spherical in the undisturbed rocks. These are found
in freshly fractured surfaces of fragments taken from the bottom of excava-
tions so deep as to be beyond the reach of surface-waters. The only explana-
tion for these spots is the presence of small bits of organic matter which
reduced the ferric oxide after deposition. In other specimens from similar
localities the light green or blue color appears in blotches and irregular
patches, but all with sharply defined limits. In masses of hard red shale
or sandstone nearer the surface the edges of cracks, both horizontal and
vertical, have the same light green or blue color, due to the infiltration of
surface-waters carrying organic matter.

• White, David, Jour. Geol., vol. xvii, pp. 320-341.

' Udden, J. A., The Deep Boring at Spur, Bull. Bur. Ec. Geol. and Techn., Univ. of Texas,
1914, No. 363. Potash in the Texas Permian, Ibid., No. 17, 1915.

250 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

In the Eastern Province the deposits of Pennsylvanian time below the
middle Conemaugh are limestones, gray and black shales, and light-colored
sandstones, evidently deposited in swamps or shallow basins subject to
invasions by marine waters. The deposits were derived from low lands at
great distances and laid down in either a submerged area or areas in which
the water-table reached the surface, so that the iron which furnished a large
part of the coloring matter remained in the ferrous condition.

The elevation of the eastern side of the continent was apparently beyond
the limits, to the east, of the Appalachian trough, as is evidenced in the
accumulation of conglomerates to the north and west of the Cobequid
Highlands of Canada and in the Boston Basin. In the latter region the
elevation was sufificient to produce at least local glaciation. To the south
the elevation was less, but still sufficient to expose the igneous rocks of the
pre-Cambrian core of Appalachia to the vicissitudes of a rigorous climate
with wet and dry seasons. It is not necessary to assume that the climate
was glacial or even semiglacial in the latitude of Pennsylvania and West
Virginia; it may have been not unlike that of the present day, for red
deposits are forming now in the Potomac River and the conditions in the
southern Appalachians and upon the Piedmont Plateau are just such as
would permit the formation of red sediments. Such an assumption is not
inconsistent with the conception of local glaciation on higher areas, as, for
instance, southeast of the Boston Basin.

The sudden change in the sediments and the uplift which they reveal,
with the consequent climatic changes, are decidedly inconsistent with current
conceptions of uniform conditions prevailing throughout Pennsylvanian time.
The picture drawn by David White (see page 233) and the idea of a uniform
climate prevailing over the earth from pole to pole' is as correct as may
now be given for the preserved portion of the Pennsylvanian series in the
western part of the Eastern Province, but is applicable in the eastern part
of the province to only those portions which lie below the red beds (middle
Conemaugh) .

As bearing upon this point it is of value to quote from a paper by
Matthew, whose expressions are those of a trained vertebrate paleontologist.
In remarking upon Chamberlin's theories of climatic changes, he says:^

" Chamberlin's theories are to-day well known and are year by year gaining
a wider acceptance. So far as they pertain to the present subject, they differ
from the older prevailing concept of geological climatic conditions chiefly in that
they involve an alternation of climates through the course of geologic time from
extremes of warm, moist tropical and uniform, to extremes of cold, arid zonal
climates. The former are the results of prolonged base-level erosion and the

* White, David, and F. H. Knowlton, Evidences of Paleobotany as to Geological Climate,

Science, vol. 31, p. 760, 1910.

* Matthew, W. D., Climate and Evolution, Annals New York Academy of Science, vol.

XXIV, p. 173, 1915.

CLIMATOLOGY OF THE LATE PALEOZOIC 251

overflow of large continental areas by shallow sea. The latter are the results of
the readjustments needed to bring the continents once more into isostatic balance,
involving the general lifting of the continents, especially of their borders, the expan-
sion of the continental areas to their utmost limits, and the renewal of rapid
erosion.

"These alternations of conditions are marked by alternations of the prevalent
type of formation in the geological series. The uniform base-leveling corresponds
to widespread deposits of limestones and in its waning stages with coal foimations.
The periods of uplift are marked bj' thick barren formations, often red in color,
by indications of arid conditions in salt and gypsum beds, and they finally cul-
minate in great extension of glaciers from boreal and high mountain areas."

C. CAUSE OF THE CLIMATIC CHANGE.

The evidence for a slowly maturing cycle of climatic change which
culminated in excessive aridity in the late Permian or the Triassic has been
presented. The possible causes, atmospheric content, solar heat affected
by volcanic dust, and deformation, have been considered. The last was
the precipitating cause, at least the one which can be rationally tested
with the best hope of permanent results. The following discussion is
limited to that cause alone, though there can be no doubt that the others
very possibly were contributory.

The continuity of the great tectonic line which includes the Paleozoic
Alps (Hercynian, Armorican-Variscan) of Europe and the Appalachian
Mountains of North America is now beyond question and the progress of
the movement from east to west along this line is equally well established.
That there were movements even in the western part of the line at an early
date is attested by the evidence of uplift even at its extreme western limit
in mid-Pennsylvanian time. Blackwelder^ gives the following statement in
regard to their movement:

"Arkansan {mid-Pennsyhanian). — The folded structures underlying the moun-
tains of Arkansas and Oklahoma were made, as nearly as can be inferred from
current correlations, in the latter part of the Pennsylvanian period. Thus, in the
central Arkansas coal field the deformation followed the laying dowm of the lower
Pennsylvanian coal measures,* but no younger strata exist there. In the Arbuckle
Mountains of Oklahoma it occurred after the deposition of the Caney shale (early
Pennsylvanian?) and before that of the Frauiks conglomerate (late Pennsylvanian).
In the Wichita Mountains still farther west in the same State, the folds had been
truncated before the deposition of the Oklahomian (early Permian) red beds.
Thus it is the conclusion of Taff' 'that the Arbuckle uplift (= crumpling) began
near the middle and culminated near the close of the Pennsylvanian time, previous
to the deposition of the red beds. * * *

1 Blackwelder, Eliot, A Summar>' of the Orogenic Epochs in the Geologic History of North

America, Jour. Geol., vol. xxii, p. 641, 1914.
* Collier, A. J., The Arkansas Coal Field, U. S. Geological Sur\'ey Bull. 326, p. 24, 1907.
» TafT, J. A., Geology of the Arbuckle and Wichita Mountains, U. S. Geological Survey,

Professional Paper No. 31, p. 80, 1904.

252 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

"The folding of the Ouachita beds has been referred by Dana and others to
the Appalachian revolution. However, unless published correlations are seriously
in error, we must conclude that the Ouachita folds had been formed and truncated
before the deposition of the latest Pennsylvanian sediments, whereas the Appa-
lachian folds were not begun until after the early Permian strata had been laid
down. It is now generally agreed that the climax of that disturbance came near
the close of the Permian period. If these correlations are correct, we must then
recognize two separate orogenic epochs. There is apparently ground for cor-
relating the Arkansan crumpling with that which produced the Armorican and
Variscian systems of western Europe, which Haug assigns^ to the opening of the
Stephanian (upper Pennsylvanian) epoch."

Disturbances coincident with Blackwelder's Arkansas orogenic move-
ment are not recorded in any folding or disturbance of the rocks on the
east side of North America later than the end of Mississippian in the southern
Appalachians, but that there was an approximately coincident uplift of
the land east of the Appalachian trough is recorded in sedimentary changes
from Prince Edward Island to West Virginia. The descriptions of the
Glasgow conglomerate, Roxbury tillite, and the red beds of West Virginia
and Pennsylvania are given in the stratigraphic section.

The evidence for an uplift to the south and east of the Canadian and
Massachusetts localities is included in the description of the rocks of those
areas, and I. C. White's argument for the meaning of the sudden appearance
of the red beds has been given (pages 65 et seq.).

* Haug, Emile, Traits de Geologic, vol. 11, pt. i, p. 829, 1910.

CHAPTER X.

AREAL GEOGRAPHY OF NORTH AMERICA IN THE LATE

.PALEOZOIC

As suggested in an earlier paper and in other portions of this work, the
geography of North America was undergoing a progressive change during
the latter part of the Paleozoic, and no one place or period can be considered
as t>'pical of the whole. Certain areas, however, are fixed for the whole
time. The Paleozoic continent of Appalachia extended far to the east
of its present exf)osure and for some undetermined distance beyond the
present coast-line of the Atlantic. There can be no doubt that for a portion
of the time some part of this continent was a high and rugged land. To the
west of this land lay the sinking area containing the coal basins of the
Pennsjlvanian period, which was slowly filled in the latter half of that period,
coincident with and dependent upon the elevation of Appalachia.

The same condition of depressed land with accumulation of swamp ma-
terial existed to the northeast through eastern New England and the Mari-
time Provinces of Canada even to Prince Edward Island, but it is possible
that this northeastern area continued to the south, somewhat to the east,
and independent of the basin area in Pennsylvania and West Virginia.
The southern extension of the Boston and Rhode Island Basins was very
probably directly to the south, and such sediments as may now be preserved
are beneath the surface of the ocean.

South of West V^irginia and Kentucky the surface of the eastern part of the
United States was above the plane of deposition and no record is preserved.

The great area including western Pennsylvania and West Virginia
stretched away to the west as far as the eastern border of the uplift in
central Missouri. It was broken by the elevated land around the Cincinnati
uplift, but the series of beds may, in part, be traced through the portion of
Kentucky to the south. The southern peninsula of Michigan was at this
time probably in the last stages of the formation of the upper coal beds of
that State.

The elevated area in Missouri was subjected to partial invasion by local
seas from Illinois and from Kansas in late Pennsylvanian time, but soon
became dry land, and during all of the Permo-Carboniferous was out of
water and undergoing erosion. It is very probable that through all of the
late Paleozoic there was a land area north of Missouri which reached up to
the southern edge of the Canadian shield.

253

254 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

West of the Missouri land there was sea during the first part of the
late Paleozoic which endured well into the Permo-Carboniferous and only
slowly and reluctantly yielded to the prevailing movements of elevation
and the accompanying climatic change which culminated in the desiccation
of the Triassic.

Beyond this western sea was the great barrier where the Rocky Moun-
tains now stand. There can be no doubt that by the last half of the Penn-
sylvanian this barrier was prominent and eflFective, and one is led to suspect
from the amount of material derived from it that it was far larger and broader
than has yet been suggested. This barrier was not complete between the
Plains and Basin Provinces, for marine and terrestrial deposits of the late
Paleozoic may be traced more or less completely around the northern and
the southern edges. Though there are short breaks in the connection
between the beds on the two sides of the barrier in the south and the char-
acter of the sediments changes, it seems very probable that the terrestrial
conditions of the eastern side shaded off into marine conditions toward the
west, in trans-Pecos Texas. The northern limit of the barrier is equally
uncertain. It may have extended north to where the uplift merged into
the highlands of the Canadian shield, but the short distance between the
sediments of the two provinces in Wyoming and north of the Big Horn
Mountains and the probability of continuity in some places indicates that
the barrier was broken in several places. The late Pennsylvanian marine
deposits of the Basin Province merged to the north into the similar deposits
of the great sea of the same age which covered Alaska and western Canada ;
the Permo-Carboniferous deposits of inclosed and stagnant seas, in the
same age, lie upon the southern edge of the land formed by the uplift which
drained this sea toward the close of Pennsylvanian time. In greater detail
the geography of the continent was approximately as follows :

In middle Pennsylvanian time the continent of Appalachia extended far
to the north and south on the eastern side of North America; its surface, as
indicated by the derived sediments, was in general marked by a subdued
topography; west of this upland lay a broad area of very slowly sinking land,
in which the accumulation of fine sediments kept pace with the subsidence,
maintaining a nearly constant but low level of the surface. The rate of
subsidence must have been very slow, for the character of the sediments
accumulated upon it indicate that the land-surface from which they were
derived was so low and gentle that erosion must have been very gradual.
For the most part the surface of the sinking area was marked by the presence
of great fresh-water swamps, alternating with low elevations and stretches
of open water due to local invasions by marine or brackish water.

The subsiding area west of Appalachia was divided into two distinct
parts, approximately the same in position as the Northeastern and the
Southern Subprovinces of the Eastern Province, defined previously. The
northeastern part, including eastern New England, the Maritime Provinces

AREAL GEOGRAPHY OF NORTH AMERICA IN THE LATE PALEOZOIC 255

of Canada, and Prince Edward Island, was made up of isolated and semi-
isolated troughs of varying size, outlined by the pre-Pennsylvanian surface.
The whole outline of this area and the general trend of the individual
troughs indicate a contour parallel to the older elevations of western New
England and of this portion of Canada and to the (probable) western edge
of Appalachia. The distinct character of this area is indicated not only by
the shape of the troughs and the different character of the deposits, depen-
dent largely upon the small size and isolation of the troughs, but upon the
difficulty in correlating the deposits with those of the larger basin to the
southwest. The continuation of the northeastern area to the south beyond
the coast of Rhode Island is strongly suggested and leads to the impression
that a depression existed in the surface of Appalachia which continued
south for an unknown distance, entirely east of the present exposed edge
of that old land. If this be true, Pennsylvanian deposits must lie buried
beneath the coastal waters of the Atlantic and the deposits of the Atlantic
Coastal Plain. The presence of such a depression, even if confined only to
the northern part of Appalachia, would have an important bearing upon the
explanation of the different character of the sediments in the Northeastern
and the Southern Subprovinces.

The southern basin was far larger than the northeastern. It extended
from the western edge of Appalachia far into Ohio and Kentucky and at
its largest reached into Indiana and Illinois. As has been repeatedly
shown, the basin was contracting from originally very wide limits through
all Pennsylvanian time, until in the Conemaugh and Monongahela it was
restricted to western Pennsylvania and West Virginia and eastern Ohio
and Kentucky. The portion west of the Cincinnati anticline, originally
connected with the eastern part, was in late Pennsylvanian time receiving
deposits as a completely or partially isolated area. In middle Conemaugh
time came the first effects of the uplift of the eastern side of North America;
the eastern part of the basin continued sinking under the accumulating
load of sediments, but the western part was gradually raised into the zone
of erosion and purely terrestrial deposition.

The elevation of eastern North America, the initiation of Permo-Carbon-
iferous conditions, produced very different effects in the two subprovinces.

The first deposits of the northeastern basin are coarse conglomerates —
the lower conglomerates of Prince Edward Island, the New Glasgow con-
glomerate, the Roxbury conglomerate, and the Dighton conglomerate.
A part of these, the Squantum tillite member of the Roxbury conglomerate,
is of glacial origin, and glacial conditions lingered for some time, as is shown
by the series of advances and retreats of the ice demonstrated by Mansfield.^

' The author is as fully aware of the uncertainty of the stratigraphic position of the deposits
in the Boston Basin as his readers will be, but believes that the similarity of conditions
there to those of deposits of determined stratigraphic position to the north and south is
a bit of confirmatory evidence of their Permo-Carboniferous age worthy of consideration.

256 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

All of these heavy deposits came from the east (Prince Edward Island) or
southeast (New Glasgow, Boston, and Narragansett Basins), and were
evidently derived from a notably high and rugged land in these directions.
It is probable that they were derived from a distinct range involving the
Cobequid Hills, some pre-Pennsylvanian elevations in New Brunswick,
and a now-submerged portion which occupied a part of the broad conti-
nental shelf which extends as far south as Long Island Sound. Its further
extension is entirely problematical.

The southern basin was evidently much farther removed from any source
of coarse sediments. If the range of uplands postulated above was con-
tinued any farther south than is indicated by the present contour of the
continental shelf, it was separated from the western side of the Appalachia
by a continuation of the depression of the northeastern basin in which was
trapped the coarse material derived from it, or its trend carried it so far
east that only the finer material was transported to the southern basin.
It is, perhaps, more probable that the range merged into the generally lower
surface of the southern portion of Appalachia and that coarse sediments
were not originated in any quantity.

The general elevation which raised the northern or eastern portion of
Appalachia sufficiently to bring it within the possibility of local glaciation
and vigorous erosion affected the southern or western portion only sufficiently
to initiate a milder erosion under a variable climate resulting in the formation
of finer red sediments.

The progress of the elevation was slow. In the northeastern basin the
coarse sediments soon gave place to fine red sands and shales, indicating
the lowering of the rugged heights; and in the southern basin the finer red
sediments only partially replaced the darker shales, light-colored sandstones,
and thin limestones, which shows that the area continued to subside as
fast as it was filled. The western part of the southern basin, receiving a
far smaller load of deposits, was raised more rapidly until western Kentucky,
Indiana, and Illinois were above the plane of deposition; deep valleys were
being cut and purely terrestrial beds accumulating while the eastern half
was still sinking beneath its increasing burden. The lack of adjacent lands
high enough to furnish much sediment under the influence of a variable
climate accounts for the lack of red sediments except in small and local
patches. North of Ohio and Indiana, the surface of the lower peninsula of
Michigan was still covered by coal swamps after sedimentation had ceased
to the south. Northern Illinois and Wisconsin were a part of the low-lying
land area extending south from the Canadian shield. Similar conditions
prevailed to the west, where the highland of Missouri continued to the
north through Iowa to the old land of Canada; here Permo-Carboniferous
conditions left little trace, for, as has been shown by the author, the red shales
and sandstones of Webster County, Iowa, are in all probability a residual
soil of Missourian age.

AREAL GEOGRAPHY OF NORTH AMERICA IN THE LATE PALEOZOIC 257

The topographic changes resulting from the broad uplift initiated on
the eastern side of the continent were accompanied by climatic changes
fully as important. Though the continued subsidence of the eastern half
of the southern basin may in part have maintained the humid and singularly
equable climate, as a local phase, the incursion of the red sediments from
the east show that on the higher land a cooler, variable climate, with alter-
nate periods of drought and humidity, had set in. The vegetation of the
upper Pennsylvanian which is recorded in this basin was very possibly a
persistent phase holding over in a locally favorable environment, while the
general environment had changed to the new " Permo-Carboniferous condi-
tions." Farther west, beyond the limit of red sediments transported from
Appalachia, the land gradually rose and the changed climate is only recorded
in a few and scattered evidences of erosional activity, terrestrial accumula-
tion, and evidences of decreased humidity. Due to the direction of the tilting,
sedimentation continued for a longer time in the West under the conditions
prevailing in Pennsylvanian time, but in Indiana and Illinois ceased before
the "Permo-Carboniferous conditions" had migrated that far west. The
result was that the line of changed sediments marking the advance of the
climatic change rises toward the west across the stratigraphic column and
if continuous would lie above the plane of deposition in Illinois and Indiana.
(See fig. 7, p. 192.)

South of the elevation in Missouri, sedimentation terminated within
the limits of the Pennsylvanian. No evidence of conditions during Permo-
Carboniferous time has been discovered in that region.

West of the Missouri land and surrounding the southwestern extension
of Ozarkia, " Permo-Carboniferous conditions" were coincident with Permo-
Carboniferous time. At the close of the Missourian period, which corre-
sponds to the upper half of the Pennsylvanian in the Eastern Province, the
sea still covered a good part of the Plains Province, and was bordered on
the east by the terrestrial deposits and coal swamps which lay on the
western slope of the Missouri land. The movements of the sea over this
land and its final retreat have been detailed on pages 82-84. To the
south the sea bordered on the northern, western, and eastern sides of the
southwestern extension of Ozarkia, a good portion of which was finally
buried by marine and terrestrial deposits, leaving exposed only the portion
now occupied by the Wichita Mountains and the Arbuckle Hills. The
eastern side of this portion of Ozarkia was in imperfect connection with the
highland of Missouri and northern Arkansas. Coal swamps and narrow
stretches of sea covered the lower land at alternate intervals.

As Permo-Carboniferous time progressed, the sea contracted, and gradu-
ally retreated toward central or north central Kansas, where limestone and
marine shales continued to be deposited long after terrestrial conditions
were established all around the last remnant of the sea. The Plains Province

258 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

consists essentially of the area occupied by this sea and the terrestrial de-
posits formed upon the land laid bare by its contraction.

The uplift of the continent, apparently, did not raise the Missouri land
sufficiently to produce any large accumulation of red sediments; if any were
formed they have been completely removed by erosion on the central-
eastern side of the province, or such small remnants as may persist are
hidden beneath the glacial or younger deposits of soil. On the southern and
western sides the uplift of Ozarkia and the Rocky Mountain barrier was suffi-
cient to furnish enormous quantities of material under the changed climatic
conditions. So great, indeed, is the amount of Permo-Carboniferous "red-
bed" material that the source as now revealed is entirely inadequate, and
some authors, notably Schuchert, are inclined to believe that a considerable
portion of the material in the southern part of the Plains Province must
have been derived from the elevated area in southern Texas and northern
Mexico, the old positive element, Columbia. In this southern portion of
the Plains Province red sediments accumulated until they spread far and
wide, principally to the south, west, and north of the land, and finally
partially buried their source.

In a similar way the Rocky Mountain barrier between the Plains and
Basin Provinces furnished red sediments which were spread out on the
eastern face of the barrier and for an unknown distance out upon the Plains
Province, where they are now hidden by overlying younger deposits. In
the south these sediments can be traced west from the barrier until they
approach very closely to those derived from Ozarkia, but the actual meeting
of the two, if it occurs, is hidden under the southern part of the Staked
Plains. Farther to the north, in the latitude of Tucumcari, New Mexico,
the exposed red beds are Triassic in age; these may be traced to the great
sandstone plateau east of Las Vegas and picked up again on the edge of the
mountains. If any Permo-Carboniferous material occurs in this region, its
exposure is confined to a narrow strip on the eastern slopes of the foot-hills
which continues through northern New Mexico into southern Colorado.
What the extent of the buried Permo-Carboniferous sediments may be we
can not tell, for the similarity between the late Paleozoic and Triassic red
beds is so great that they can not be distinguished in well records.

From Canyon City, Colorado, north to the Black Hills, the red sediments
continue unbroken and unchanged, indicating a similarity of conditions
amounting to identity in climate, height of the barrier, and all inorganic
factors. The total absence of any trace of land vertebrates is inexplicable;
by all evidence of the sediments the environmental conditions were strikingly
similar to those existing in Texas and Oklahoma. So many men have gone
over these beds in the hope of finding vertebrate fossils that some fragments
of bone, at least, would have turned up had the animals existed in any
abundance. Some, as yet not realized, factor of distribution of the sedi-

AREAL GEOGRAPHY OF NORTH AMERICA IN THE LATE PALEOZOIC 259

ments has prevented the occurrence of fossils or some obscure factor pre-
vented the occurrence of animals in this large area.

In this connection it is pertinent to recall that the size and nature of
the barrier between the Plains and Basin Provinces is still little understood.
The widespread marine sediments of late Pennsylvanian time and their
great thickness strongly suggest a suppression of the Rocky Mountain axis
at that time and it is the opinion of some writers that the whole region was
entirely submerged. On the other hand, the amount of Permo-Carbon-
iferous sediment is so great that it can only have come from a very large
area and the nature of the material implies considerable height and vigorous
erosion. The change in the sedimentary record reveals an important uplift
forming an elevation of a height and geographical extent for which we have
no other evidence.

There is some good reason to believe that the barrier between the two
provinces failed in the region of the Black Hills and Bighorn Mountains.
The Permo-Carboniferous red beds of the Black Hills unquestionably belong
to the Plains Province and the equivalent red beds of the western side of the
Bighorn may be traced with little question into the phosphate-bearing
beds of the Basin Province. If the barrier continued to the north it must
have passed between the two elevations, but it is very probable that the
elevations were far less prominent in the late Paleozoic than now and that
they were nearly, if not quite, covered by the sediments of the time. If the
red sediments are continuous between the two, the provinces were united
at the northern end and the source of much of the material was from the
permanent land to the north and east.

At their southern ends the two provinces are separated by the great
mass of late Pennsylvanian limestone, now thrown up into the series of low
mountain ranges which run through central New Mexico. Whatever may
have been the condition in Permo-Carboniferous time, there was clearly an
effective barrier to the migration of vertebrate land animals, for the reptiles
and amphibians found near Soccoro and Abiquiu are distinctly different
from those found in the Plains Province. There can be no doubt that in late
Paleozoic time the western part of trans-Pecos Texas and the adjacent
portions of Mexico and New Mexico were covered by a sea which extended
into Arizona. This sea left the great Guadalupian series of limestones,
the upper part of which is Permo-Carboniferous, possibly true Permian,
and the sea may have extended to the north, constituting a temporary but
effective barrier to land life. It is true that the upper limestone of the
Guadalupian series shades off into red deposits toward the north, implying
land in that direction. In the absence of any definite evidence of a high
mountain barrier, the presence of a temporary arm of the sea would be the
most satisfactory explanation of the separation of the two faunas; it would
constitute an efficient continuation of the land barrier farther north.

260 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

On the west side of the Rocky Mountain barrier the deposits of the
Basin Province resemble those of the Plains Province and the environment
of life was the same, as far north as southwestern Colorado and southern
Utah; beyond, to the north, conditions were radically different through the
most of the Basin Province. Only in central Wyoming do the Permo-
Carboniferous deposits of the Basin Province shade again into red beds
typical of the Plains Province.

The upper Pennsylvanian sea occupied the Basin Province, depositing
heavy limestones and, toward the end, receiving great quantities of sand
in its northern half, the present Weber quartzite and its equivalents. The
limestone may be traced with a fair degree of certainty into Canada and
the quartzite may be followed as far, from its first appearance in central
Utah, until the two apparently merge into the Cache Creek formation.

In the southern part of the Basin Province the elevation permitted red
beds of Permo-Carboniferous age to accumulate directly upon the limestone,
but farther north the elevation of the barrier to the east must have been
somewhat earlier, and the sands of the Weber and its equivalents were
deposited under Pennsylvanian climatic conditions. The Permo-Carbon-
iferous deposits of the northern half of the Basin Province were laid down
in stagnant seas and under climatic conditions not greatly different from
those of the late Pennsylvanian.

The difference in the sediments on the two sides of the barrier reveal
important climatic differences. The source of the Weber sands and the
limestones and fetid shales of the Park City formation must have been the
same as that of the red beds of the eastern side of the barrier, and the
structure and size of grain reveal no great difference in the shape or slope
of the two sides. The main erosion of the western side took place in late
Pennsylvanian under Pennsylvanian climatic conditions and of the eastern
side in Permo-Carboniferous time under Permo-Carboniferous climatic condi-
tions, but from the absence of red beds on the western side of the barrier,
except where it was probably broken down at the northern end, and from
the constant indications of the large size of the barrier, the question naturally
presents itself whether the barrier may not have interrupted prevailing
winds, or otherwise caused a difference in humidity and temperature upon
the two sides.

The sum of accumulating evidence shows that marine conditions pre-
vailed in late Pennsylvanian from Alaska south along the Pacific coast to
northern California; in British Columbia the sea extended as far east as
the west side of the Canadian Rockies and south to the international boun-
dary, where it was continuous with the sea which occupied the Basin

Fig. 8. — Map showing the distribution of late Paleozoic and Permo-Carboniferous deposits in
North America. The portion of Canada and the United States to the west of the heavy
broken line was the site of deposition in Gschelian time and elevation in late Paleozoic.

AREAL GEOGRAPHY OF NORTH AMERICA IN THE LATE PALEOZOIC 261

Late Paleozoic TVpical "Red Beds of the

deposits pi^tns and basin prtwincfrs

Note: Dotted lines jodii^te
probable extension of
Ihe beds bewood their

T^srmo-Garbonrferous
deposits

Scale

X)0 o

present linirts.

262 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

Province. There is little doubt that the deposits along the Pacific coast
were made in a sea which was in more or less open connection with the one
which lay over British Columbia, but all of the sedimentary deposits of the
eastern sea have been so disturbed by earth movements and so seriously
metamorphosed that the correlation can only be made in a very broad way.
One thing is becoming increasingly evident — that the limestones and asso-
ciated clastic sediments of Alaska and western Canada were deposited in a
sea of late Pennsylvanian time, equivalent to the Gschelian.

In late Pennsylvanian an uplift, first apparent in northern Alaska, devel-
oped and spread to the south. This movement was accompanied by much
diastrophism and vulcanism which, with later phenomena of like kind, have
sadly obscured the record. The effect of the whole movement was to raise
the surface above the plane of marine deposition during Permo-Carboniferous
time and no traces of terrestrial deposition or erosion during that time have
as yet been detected.

This uplift was of the first importance and had a two-fold efifect upon the
geography and the environment of life during Permo-Carboniferous time:
(i) The movement of the uplift was progressive from north to south, at
right angles to the progressive movement of uplift of the continent through-
out late Paleozoic. It is very probable that the progressive uplift from
north to south, exerting most of its effect farther west, penned in a part of
the sea which occupied the northern half of the Basin Province in Pennsyl-
vanian time and converted it into the relict seas of Permo-Carboniferous
time in which were deposited the phosphate shales and limestones of the
Park City formation. (2) The uplift converted the northwestern part of
the United States and the western part of Canada into an upland which
joined the northern end of the Rocky Mountain barrier between the two
provinces and prevented the accumulation of Permo-Carboniferous sedi-
ments north of the international boundary. The land area thus formed
furnished a reasonable route of migration for the Gigantopteris flora from
its home in Asia to the location in Texas where it has been found.

The upper limit of Permo-Carboniferous time or "Permo-Carboniferous
conditions" is as yet undeterminable with any exactness. In the regions
where Triassic red beds overlie those of Permo-Carboniferous age, there is
no evidence of earth-movements of any magnitude, but there is constant
evidence of an increasing aridity which became so vigorous as to be a
prominent, if not the dominant, factor in the change of vertebrate life, so
pronounced at the juncture of the two periods.

CHAPTER XI.

DEVELOPMENT AND FATE OF VERTEBRATE LIFE IN THE

PERMO-CARBONIFEROUS IN RELATION TO ITS

ENVIRONMENT.

The study of the development of vertebrate life in North America during
late Paleozoic time emphasizes the changes from a long period of slow
evolution in a singularly monotonous environment through a period of
rapid expansion in a diversified environment to final extinction. As has
been repeatedly intimated in the course of this work, the chief directing
influence in the sudden expansion was a decided climatic change, accom-
panied by physiographic changes, induced by an alteration in the level of
the surface of the continent.

The basis for any study of the vertebrate fauna in relation to the en-
vironment must be a study of the morphology of the forms involved.
Previous publications of the Carnegie Institution of Washington by the
author and others have summarized this as far as our present information
permits. The material in the preceding pages summarizes our knowledge of
the environment in a similar way.

In attempting an analysis of the response of the vertebrates of the late
Paleozoic to the changing environment it is necessary to begin the study
with at least the middle of Pennsylvanian time, in order to understand the
conditions which fixed upon the animals the homoplastic characters which
were developed in the Permo-Carboniferous radiation. There can be no
question that the radiation of the fauna began with the development and
spread of "Permo-Carboniferous conditions" in the middle Conemaugh
time, not at the beginning of Dunkard time as has been commonly assumed.
It is therefore of the utmost importance to keep clearly in mind the great
change in the inorganic environment which came with the development of
Permo-Carboniferous conditions.

Fortunately for the simplification of the work, the conditions during
early Pennsylvanian time were singularly uniform over large areas.^ David
White has repeatedly emphasized the equability of the humid climate.
The work of David White, Stevenson, and others has emphasized the topo-
graphic uniformity of the slowly sinking coal basins, in which the filling
maintained a nearly constant level. Under such conditions the ultimate
food-supply, vegetation, would be fixed in kind though abundant in quan-

* See analysis of an environment on p. 40.
263

264 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

tity. Such an environment would permit an enormous increase in numbers,
but the increase in kinds would reach a definite limit, whether large or small.
A monotonous environment does not imply a small number of genera and
species in a fauna or flora, but it does imply a distinct limit to the number.
The struggle for existence can result in the persistence of new forms only
so far as the new forms can find an isolation or favorable environment;
beyond that point new forms can not survive and a period of stagnation in
development will ensue, the stagnation being more or less complete as the
monotony of the environment is more or less pronounced and long-continued.
This condition will be in the nature of an end-result upon a fauna or flora
and the effect does not necessarily extend to the suppression of variability
in the individual organisms. In the struggle for existence, induced by
increasing numbers and closer adjustment of the inter-relations between
distinct species, the amount of variability might remain the same or even be
accentuated. In the author's opinion the tendency to continue the develop-
ment along definite lines, call it by what name we will, would be seriously
affected by the failure of new forms to develop to maturity. The lack of
mature new forms would prevent the survival of new structures demon-
strating the tendency of evolution, beyond a certain point, but, at the same
time, the constancy of the environmental conditions would tend to fix ever
more firmly the tendency and increase the number of variants in that direc-
tion. Viewed as a whole, then, the end-result would be, on a large scale,
somewhat as discussed in the author's paper on the Linton fauna. The
animals in their environment of limited possibilities of morphological expan-
sions would soon fill all of the available spaces and new forms would cease
to reach maturity, this being largely due to the early extinction of the
variants. The continued and increasing pressure to produce new forms
because of the fixation of the tendencies would impose upon the fauna as a
whole a state of stress to which it would need only the relief afforded by the
possibilities of migration into a new environment or of a change in the
environment to find expression in a rapid development of new types. This
opinion is opposed by my colleague. Dr. A. F. Shull, who suggests that an
inherent tendency to evolution along definite lines would find relief in the
mere production of offspring and that the fauna would not experience an
increase in any tendency to develop along definite lines or any increase
in the number of variants. This suggestion has great force and it may well
be that it is the true situation. The author is well aware of the limitations
of the evidence at the disposal of the paleobiologist, but the evidence is so
conclusive and repeated of long periods of stagnation in evolution followed
by rapid development that he can not rid himself of the impression that
faunas in periods of stagnation go through a period of preparation, in some
form, for their subsequent radiation.

DE\^LOPMENT OF VERTEBRATE LIFE IN PERMO-CARBONIFEROUS 265

Environmental monotony would result in the persistence of older and
simpler types because the variants, possibly being constantly produced, would
not have a chance to develop.

Another result of the long association of the unchanging or only slowly
changing members of the fauna would be a very close adjustment of the
interrelationships of the various elements of the fauna, and of the fauna
with the flora and the inorganic environment. In a region of large possi-
bilities of varied habitat such persistence, amounting to static conditions,
might result in the production of highly specialized tjT^es showing excessive
morphological characters, but in a region of limited possibilities of habitat
the morphological expression of close adjustment would be less obvious.
Close adaptation of the interrelationships is in itself an evidence of the
long association of comparatively fLxed groups, and when this finds expression
in the skeletal structure it can easily be read in the fossils, but if it is expressed
in physiological characters or in habits not revealed in the structure, as pecu-
liarities of feeding, etc., or in other things which may seem to the observer
minute and unimportant but are in reality vital, the record is not decipherable.

Again, as has been shown by Beecher, certain morphological peculiarities
are produced only in the senility of a group and are characteristic of it;
the upper Paleozoic air-breathing vertebrates were still in the best stage of
their development. Though amphibians appeared in the Devonian and
footprints record their occurrence through the Mississippian and early
Pennsylvanian, their progress seems to have been very slow, a condition
which is analogous to that of the mammals in the Mesozoic.

We have as yet no knowledge of the place of origin or the direction of
migration of the Permo-Carboniferous fauna, zmd speculation upon changes
induced by migration into new regions must be based on very meager
evidence. It is altogether possible that in their slow development the
primitive amphibia migrated only into extensions of their original environ-
ment and so experienced little change in their surroundings and received
no stimulus to evolution. The almost purely aquatic character of the
primitive amphibians and the intolerance to salt water, so characteristic of
the living, and presumably of the fossil, forms would tend to restrict their
movements most decidedly. Only when the as yet undetermined influence
which compelled a change appeared or gathered sufficient force to cause a
vigorous development would they burst through the physiological barriers
and begin their radiation.

The upper Pennsylvanian air-breathing vertebrate fauna was yoimg,
very numerous in individuals, possibly in kinds, but restricted in its further
development by the monotony of the environment. It was, however,
accumulating force towards a great radiation to be expressed as soon as the
limitations were removed, even in a partial degree.

266 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

The middle of Conemaugh time saw the beginning of the end of the fixed
environment of the amphibia and the first reptiles. The change began on
the eastern side of the continent and the record as shown by the deposits
has long been recognized. Girty says:^

"The Upper Carboniferous, rather in contrast with the Lower, was a period
of emergence of shores and of shallowed waters, and it presents the variety that
appertains to such conditions. In considering the stratigraphic relations of eht
Pennsylvanian and Permian one can not fail to be struck by the local character
of the phenomena, and the vast amount of detail, from which it is difficult to
disengage facts of broader significance."

" Lithologically the beds of the Upper Carboniferous* and Permian present
the greatest variety, and about the only truth of broad applicability has long
been known. I mean that in eastern North America the sediments of the Upper
Carboniferous are chiefly shales, sandstones, and conglomerates, with some thin
limestones, while in the West the limestones have a much larger development,
and coals, which toward the east play so important a part, if not in thickness at
least economically and significantly in the Carboniferous sediments, are there
practically absent. From this it has been justly inferred that the character of
the eastern Carboniferous indicates shore and estuarine conditions of deposition,
while that of the western indicates marine conditions of deposition."

The variety of sediments emphasized by Girty is, however, rather a
repetition of a relatively few kinds of beds than an evidence of varied
habitat.

The same change is recognized by David White,' who notes that the
beginning of Stephanian time dates from the Hercynian uplift in Europe,
but that the abrupt differences of level and vegetation do not appear in
America :

"In view, however, of the paleobotanical evidence indicative of a point near
the Allegheny-Conemaugh boundary, I, personally, am inclined to regard the
formation of the Mahoning sandstone (conglomeratic), the changed sedimentation
of the Conemaugh formation, the probable upwarp of the southern Appalachian
region which later resulted in the exclusion of the sea from the northern area
also, and the consequent climatic changes, as due to the same great orogenic
influence. Accordingly I would provisionally place the greater part, if not all,
of the Conemaugh together with the Monongahela in the Stephanian.

"The final exclusion of the sea from the Appalachian trough appears to have
occurred soon after the deposition of the Ames limestone, near the middle of the
Conemaugh, since, according to reports, only fresh or possibly brackish water
mollusca occur in the higher terranes. It is probable that the Monongahela
was never deposited in the southern Appalachian region, from portions of which
the Conemaugh may also have been absent, the red oxidized sediments of the
latter being in part derived, I believe, from the eroded unconsolidated older
Pennsylvanian to the southeastward."

' Girty, G. H., Outlines of Geological History, p. 124, 1910.

^ Idem, p. 126.

' White, David, Outlines of Geologic History, p. 148, 1910.

DEVELOPMENT OF VERTEBRATE LIFE IN PERMO-CARBONIFEROUS 267

The same change is noted by I. C. White/ who, however, emphasizes
the change in Hfe which followed the change in environment:

"Viewed from the standpoint of change in physical conditions the proper
place for such a dividing-plane between the Conemaugh and Allegheny beds
would be the first general appearance of red rocks, near the horizon of the Bakers-
town coal, about lOO feet under the Ames or crinoidal limestone horizon. That
a great physical change took place soon after the deposition of the Mahoning
sandstone rocks, the present basal members of the Conemaugh series, must be
conceded, since no red beds whatever are found from the base of the Pottsville
up to the top of the Allegheny, and none worth considering until after the epoch
of the Upper Mahoning sandstone.

"The sudden appearance or disappearance of red sediments after their absence
from a great thickness of strata is always accompanied by a great change in life
forms, and the present one is no exception. In fact, the invasion of red sediments
succeeding the Mahoning sandstone epoch of the Conemaugh may well be
considered as the ' beginning of the end ' of the true Coal Measures, both from a
lithological as well, as a biological standpoint, and hence it is possible that the
best classification, aside from the conveniences of the geologist, would leave the
Mahoning sandstone in the Coal Measures, and place the rest of the Conemaugh,
as well as the Monongahela series above, in the Permo-Carboniferous. This
reference is also confirmed by the character of the fauna and flora, both of which
contain many forms that characterize the Permo-Carboniferous beds of Kansas
and the west as may be seen in the lists published on a subsequent page under the
detailed description of the principal Conemaugh strata."

In previous pages emphasis has been placed upon the physical change
which began on the eastern side of the continent and spread to the west.
The land was gradually elevated east of the coal basins and a cooler and
less humid climate accompanied the elevation. It has been pointed out by
David White, Stevenson, and others that the eastern half of the Southern
Subprovince continued to sink for some time after middle Conemaugh time,
as it received the added load derived from the rising land to the east, but
the change of the sediment and, above all, the new elements in the fauna
and flora, show that the climatic change was having its effect. Beyond the
limits of the basin in all directions the lack of accumulating sediments
permitted the elevation to have full effect.

The fauna, long restrained from any expression of its evolutionary
tendencies, full fed, and in the vigor of its youth, responded at once to the
change, and new forms appeared so suddenly as to be unheralded in the
preserved remains. This is of course more apparent than real, but there was
unquestionably a rapid evolution amounting to vigorous radiation, expressed
especially in those features which were adapted to life upon a drier land or
in aquatic areas of far wider possibilities of varied habitat than in the enor-
mous and monotonous swamps. Either of these conditions, rapidity of evolu-
tion or life away from favorable conditions for the preservation of the remains,
would lessen the probability of the preservation of the connecting forms.

There can be no question, however, that the response to the changed
environment was rapid and in favorable localities very complete. The

' White, I. C, West Virginia Geological Survey, vol. n, p. 226, 1903.

268 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

variants upon the strongly impressed homoplastic characters were now able
to survive and in the multitude of new forms far-reaching readjustments
were necessary to the changed interrelationships. The environment, long
monotonous, became increasingly diversified, until the elevation and the
climatic change had produced their full effect, and new avenues of migration
were constantly opening westward. The number of individuals was far
less relative to the available space, and the cloture of the environment was
removed. In the readjustment many of the animals found an isolation
which permitted rapid development of morphological peculiarities.

The author must here repeat to some extent his conception of isolation
in the sense here used. The environment of any organism is the sum of
all its contacts with the external world, organic and inorganic. If the
organism, by virtue of its structure, habits, acquired immunity, or other
means attains a position in its enviroment where it is relieved from prejudicial
contacts either in part or in toto, it just so far attains an environmental
isolation and is free to develop individual peculiarities. This is decidedly
different from geographical isolation, though the latter may result in the
same benefits to the organism. Environmental isolation may be attained
even in a very thickly inhabited region. A typical case is that of the
modern skunk; another is the Permo-Carboniferous reptile Dimetrodon,
which, by increase in size, agility, and raptorial powers, so far dominated the
fauna in which it lived that it was isolated from many disadvantageous
contacts and developed probably incipiently useful structures to marked
excess.

Such a condition of environmental isolation, attained by only a portion
of the fauna, accompanied by close adaptations in the interrelationships
of the members of the fauna, could only be attained after a really long period
of association; in the early stages of such an association, changes of far-
reaching effect would occur. The individuals would be free in a large
measure from peculiarities, except inherited ones developed under the earlier,
more stable conditions. Such peculiarities would be very apt to be dis-
advantageous, and the animals possessing them would disappear, and the
less-specialized forms would gradually assume peculiarities as adjustment
took place under the new conditions. As intimated above, the extinction
of certain forms and much of the readjustment would take place in a time
of rapid evolution and the record would be very faulty from the lack of
preserved material. It has frequently been shown that the changes from
uniform conditions of the land to elevated and disturbed conditions have
always been relatively rapid, while the return to low lands and a positive
movement of the strand-line have been relatively slow. Thus the air-
breathing life was uniformly subjected to relatively rapid changes effecting
a violent disturbance of the established relations, followed by a long period
of extremely gradual return to monotonous conditions, producing frequently
a static environment, during which readjustment took place.

DEVELOPMENT OF VERTEBRATE LIFE IN PERMO-CARBONIFEROUS 269

As the change which affected the environment of the upper Penn-
sylvanian fauna was an elevation and exposure of the land with little
deposition the record of the life change, in preserved fossils, is very imperfect,
the continued sinking of the land in the eastern part of the Southern Sub-
province maintained the old conditions and archaic types of life in the area
best explored and has given a false impression of the general state of affairs.
The record there preserved is of a relict fauna, while the record of the ad-
vancing development in progress on the higher lands is not preserved or has
not yet been discovered. Matthew, in his essay upon climate and evolution,
has given his ideas of the effect of such a change as occurred in late Penn-
sylvanian time:^

"The periods of continental emergence were periods of arid and markedly
zonal climate, and the faunae must adapt themselves to these conditions. Such
conditions, while favoring the spread and wide distribution of races, would be
unfavorable to abundance of life and the ease with which animals could obtain a
living. The animals subjected to them must maintain themselves against the
inclemency of nature, the scarcity of food, the variations of temperature, as well
as against the competition of rivals and the attacks of enemies. In the moist
tropical climatic phase, animals would find food abundant and temperature rela-
tively constant; but the larger percentage of carbonic acid and probably smaller
percentage of oxygen in the atmosphere during those phases would tend to slug-
gishness.

"We should expect, therefore, to find in the land life adapted to the arid
climatic phase a greater activity and higher development of life, special adapta-
tions to resist violent changes in temperature and specializations fitting them
to the open grassy plains and desert life. In the moist tropical phase of land
life, we should expect to find adaptations to abundant food, to relatively sluggish
life, and to the great expanse of swamp and forest vegetation that should
characterize such a phase of climate."

While these remarks are largely applicable to the mammals with which
Dr. Matthew is chiefly concerned in his essay, similar effects would doubtless
be produced in lower forms of life.

The place of origin of the Permo-Carboniferous fauna is unknown. As
shown previously there are reasons for considering that the North Amer-
ican radiation of the Permo-Carboniferous fauna started in the eastern
part of the continent, but this is far from proven. Until more detailed
work shall establish the true relations of North America to the other con-
tinents of late Paleozoic time the question may not be settled. The
question of the origin of the fauna and its relations to the somewhat similar
fauna of South Africa is discussed in a preliminary way elsewhere (Carnegie
Inst. Wash. Pub. No. 207, p. 117). It is the opinion of the author that
the North American fauna was unique in its adaptations and radiation"
and that the presence of such highly specialized forms as Edaphosaurus in
Saxony and Bohemia was due to later migrations. This opinion may very

* Matthew, W. D., Climate and Evolution, Annals N. Y. Acad. Sci., vol. xxiv, p. 177, 1915.

270 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

possibly be shown to be erroneous. The presence of such highly specialized
forms, so little changed as to be only specifically separable, at such widely
separated localities leads to a consideration of the effect of wide migration
upon a fauna. Matthew, in his paper on climate and evolution, discusses
this effect.^

"Whatever agencies may be assigned as the cause of evolution of a race, it
should be at first most progressive at its point of original dispersal, and it will
continue this progress at that point in response to whatever stimulus originally
caused it and spread out in successive waves of migration, each wave a stage
higher than the previous one. At any one time, therefore, the most advanced
stages should be nearest the center of dispersal, the most conservative stages
farthest from it. It is not in Australia that we should look for the ancestry of
man, but in Asia.

"In the same way, in considering the evidence from extinct species as to
the center of dispersal of a race, it has frequently been assumed that the region
where the most primitive member of a race has been found should be regarded
as the source of the race, although in some instances more advanced species of
the same race were living at the same time in other regions. The discovery of
very primitive sirenians in Egypt, while at the same time much more advanced
sirenians were living in Europe, has been regarded as evidence that Africa was
the center of dispersal of this order. It is to my mind good evidence that it was
not. It is very common to see references to the African fades of the Miocene
or Pliocene mammals of Europe; but it is much more correct to say that the
modern African fauna is of Tertiary aspect and is in large part the late Tertiary
fauna of the northern world, driven southward by climatic change and the com-
petition of higher types.

"The chief arguments advanced in support of the method here criticized
appear to be that the modification of a race is due to the changes in its environ-
ment and that the primitive species are altered more and more as they spread
out or migrate into a new environment; but, assuming that a species is the product
of its environment, the conclusions drawn would only hold true if the environ-
ment remained constant. This is assuredly not the case, and if it were there
would be no cause left for the species to change its range. In fact, it is the
environment itself, biotic as well as physical, that migrates, and the primitive
species are those which have followed it, while those which remained have had
to adapt themselves to a new environment and become altered thereby. Prob-
ably it is never the case that the environment of the marginal species is an
absolute replica of the older environment of the race. In many cases it must be
profoundly modified by its invasion of new regions, and there are many features
in the evolution of a race which appear to be only partly, if at all, dependent on
environmental change. But to assume that the present habitat of the most
generalized members of a group, or the region where it is now most abundant,
is the center from which its migrations took place in former times appears to me
wholly illogical and, if applied to the higher animals, as it has been to fishes and
invertebrates, it would lead to results absolutely at variance with the known
facts of the geologic record."

' Matthew, W. D., Climate and Evolution, Annals N. Y. Acad. Sci., vol. xxiv, p. i8o, 1915.

DEVELOPMENT OF VERTEBRATE LIFE IN PERMO-CARBONIFEROUS 271

From the above it is evident that Matthew considers that peripheral
species of an expanding fauna are the most primitive and that species near
the point of origin show the greatest modification. This might be true under
certain conditions, but only if the migrant species followed a migrating
environment which advanced as an expanding belt without change and if the
original location underwent progressive change. It is obvious that this is
not always the case; in the Permo-Carboniferous radiation the change of
environment starting from the eastern side of the continent spread out, not
as an expanding belt, with other and new conditions arising at the original
source, but rather as an expanding blanket from a source where conditions
remained constant. In this case, certainly, the migrant forms would follow
the edges of the expanding environment, but the point of origin would offer
no stimulus for further change and the fauna would be similar in all parts,
except that at the place of origin where the fauna had lived longest in close
association there would be closer adaptations in the interrelations of the
organic environment, and except as the factors of evolution which are
independent of the environment made themselves felt. In such highly
specialized forms as Edaphosaurus, as widely separated as Bohemia and
New Mexico, it is certain that one place, at least, is far removed from the
place of origin. There is, of course, the possibilitj', less in the reptiles
perhaps than in the mammals, of recurrent migration or of the active mi-
gration of groups during really short intervals of time, within an environment
very similar over large areas; this might cause certain forms to become very
widely spread and to confuse the interpretation of the fossil record.

With the spread of Permo-Carboniferous conditions, the fauna of the
time extended its range over a large part of the United States at least,
if not over a large part of the continent. Whether the fauna occupied all
parts of its extreme range at any one time is uncertain, but it is certain that
in late Paleozoic, true Permo-Carboniferous time, it extended from Penn-
sylvania to New Mexico, and it is very probable that it existed in regions
where no remains have been discovered. The large proportion of exposed
land in the eastern and southern parts of the United States permitted little
preservation of the remains of the animals, if any were present. Beyond
New Mexico we have no record and in the northwest marine conditions
prevailed during Gschelian time and gave place to elevated land, with the
accompaniment of volcanic disturbances and earth-movements which prob-
ably prohibited the presence of the peculiar vertebrate life, though it per-
mitted the passage of the fern Gigantopteris .

Permo-Carboniferous vertebrate land life came completely to an end
in North America and, so far as we can tell, suddenly. Inadequate
history makes it impossible to draw conclusions which approach to finality,
but the chances for the discovery of far-reaching evidence are seemingly
so small that it is well to summarize what can be said from that now
available.

272 ENVIRONMENT OF VERTEBRATE LIFE, ETC.

It is certain from the interpretation of structure that a great preponder-
ance of the fauna was palustrine in habit. This is certainly true of the
Amphibia and equally true of most of the Reptilia. Certain of the reptiles
were adapted to other conditions, as the very swift, perhaps arboreal,
Areoscelis, and Edaphosaurus, whose structure and the method of occurrence
leads to the belief that it was an upland animal. The fauna as a whole
migrated with the extension of its peculiar habitat, following the retreat of
the swamps and marine littoral probably to the westward, and came to an
end with the habitat. The upland members of the fauna, many of which
have undoubtedly left no traces of their existence, lingered upon the higher
lands and may have moved about freely, even penetrating back to the
eastern side of the continent.

It is at least suggestive that Edaphosaurus, an upland form, is found
farthest from the locus of the greatest abundance of the genus, but an equally
significant fact, if we could understand it, is the presence in the same locality
(Bohemia) of Cricotus, or a closely related form, the most aquatic of the
Amphibia.

The only consideration that offers a possibility of reconciling these
apparently contradictory bits of evidence is that Edaphosaurus was capable
of wide migration because of its land habits, and Cricotus was equally capable
because of its power as a swimmer. If Cricotus had the intolerance of salt
water so characteristic of the Amphibia, and there is no evidence to the
contrary, there must have been a peculiarly favorable geographical arrange-
ment that permitted these two forms of such different structure and habits
to reach the same locality after such a long journey. If Cricotus was
tolerant of salt water, it is strange that no remains have been found associated
with the widely distributed Mesosaurus.

The great majority of the fauna seem, so far as we can tell, to have been
greatly favored by their environment, but to have been closely restricted
by it and quite unable to adjust themselves to changing conditions.

The close of Permo-Carboniferous conditions and the beginning of
Triassic conditions was accomplished within the limits of the "red beds,"
and certainly within the limits of any known occurrence of air-breathing
vertebrates, by a steady increase in the aridity of the climate, with a con-
sequent decrease in the area of aqueous habitats and an increase in the
salinity of such as remained, which were largely desiccating pools.

To anyone who has studied the red beds of late Paleozoic and early
Mesozoic time the similarity in the structure of the two is beyond question,
but the evidence of gradually increasing aridity is equally obvious. The
Permo-Carboniferous beds do not contain any large amount of salt or
gypsum or other evidence of extreme aridity in the lower middle portions —
the equivalents of the Wichita and Clear Fork formations — where the fauna
occurs, but such evidence increases in the upper portion and in the Triassic

DEVELOPMENT OF VERTEBRATE LIFE IN PERMO-CARBONIFEROUS 273

beds it reaches a maximum. As is well known, there is no great structural
break at the top of the Paleozoic series in these regions; in only one place,
near Ouray, Colorado, is there known even a considerable unconformity
at the base of the Triassic.

It must be remembered that the deposits of the red beds of late Paleozoic
and early Mesozoic are not in any degree continuous. They consist of
lenses and small irregular bodies of shale, sandstone, and clay, with small
layers of impure limestone, following each other in no persistent sequence.
This is so pronounced that a section made at one place is not to be depended
upon even within the distance of half a mile. A consideration of the many
sections taken in the regions concerned will reveal this condition.

Despite the similarity of the beds revealing similar conditions of deposi-
tion, the beds carrying Permo-Carboniferous vertebrates are uniformly
followed by a barren inter\^al and tHen by the appearance of typical Triassic
forms. Other than the obvious climatic change, there is no suggestion of a
cause for the extinction of the fauna, but the climatic change is itself a
sufficient explanation. The members of the fauna were closely adapted to
ever>' phase of their environment, and, in taking advantage of abundant
possibilities, had developed to a high degree of specialization. They had
passed the zenith of their development and the group as a whole was in a
stage of developmental senility where overdevelopment of certain morpho-
logical characters is the common condition. Flexibility in evolution was so
far lost that with the advent of a new environment the fauna disappeared.
This is quite what would be expected. The only question is whether the
fauna was totally extinguished upon the continent and replaced by a new
fauna developed elsewhere, or whether some of the less-specialized forms
survived to give rise to the Mesozoic types. This question can not be
answered at present, but so far as we know there were no survivals in North
America. WTiat was going on in the portions of the continent from which
we have no record we do not know; it is possible that some of the forms
migrated to the continents of the Old World and there perpetuated the
fauna in the Triassic. One thing is very definite at present: no form has
been found which bridges the gap between the North American fauna and
that of the Triassic. The two occur in the same localities, but are separated
by a barren space in the geological column, and when the new forms appear
they are already well-defined Mesozoic types. So far as we can now tell,
the Permo-Carboniferous land vertebrates became extinct upon the North
American continent and the Mesozoic forms appeared by migration from
an unknown source. Perhaps connecting-links may be found in North
America, but at present there is no suggestion that they exist or of where
to search for them.

THE UNIVERSITY LIBRARY

UNIVERSITY OF CALIFORNIA, SANTA CRUZ

SCIENCE LIBRARY

This book Is due on the last DATE stamped below.

1^/1 JUN lo

N.B.-HOLD

MAR 1 3 1974

mvMARe

REC'D MARli

50m-4,'69(J7948s8 12477

OEtnxasci

liriTrrniiiiiii

3 2106 00249 7276

■^<mm>m

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "The environment of vertebrate life in the late Paleozoic in North America; a paleogeographic study"

See other formats