,'1 ' ",'"f'

Dictated

DEPARTMENT OF THE INTERIOR

UNITED STATES GEOLOGICAL SURVEY

WASHINGTON D c, AotII 11 1894.

My dear Dp. Hollick;-

L«squereux was very sdovenli' In his
references, and I have had my trials vntn them but have not

^^.

/^

DEPARTMENT OF THE INTERIOR.

gjjlES S. MERKIAM

MUNIEU.*" *T LAW.

•1

UK«Ty.T.,NlWY«K. R E P O PI T

OF TirE

raiTED STATES GEOLOGICAL SURVEY

THE TERRITORIES

R V. HAYDEN,

UNITED STATES GEOLOGIST-IN-CHARGE.

LIBRARY
NEW YORK
BOTANIC AU

GARDEN

VOLUME VII.

WASHINGTON:
GOVERNMENT PRINTING OFFICE.

1878.

I

LIBRARY
NEW YORK
BOTANICAL.

QARDEN

LETTER TO THE SECRETARY.

Offick of the United States Geological and

Geographical Survey of the Territories,

Washing fori, D. C, January 1, 1878.

Sir: I have the honor to transmit herewith, for your approval and for
publication, the Seventh Volume of the Final Reports of the Survey under
my charge.

The work consists of a Report on the Tertiary Flora of the West, by
Prof Leo Lesquereux, of Columbus, Ohio, being Part II of ''Contributions
to the Fossil Flora of the Western Territories"— Part I being "The Cre-
taceous Flora", by the same distinguished author, whose long-continued
studies in palseo-botany have placed the subject upon a firm and enduring
basis. Part I formed Vol. VI of this series of Final Reports, and the
present Part constitutes Vol. VII.

The circumstances which led to the preparation of this work may be
briefly reviewed.

When the geological investigation of the Lignitic formations of the
West had reached a certain point, the conclusions derived from such studies
were discussed by certain geologists who dissented from the views then
expressed respecting the age of these strata. It consequently became
desirable, in order to the solution of the questions invoUed, to elaborate
further, and with the greatest care, all available material bearing upon this
interesting problem. To this end, I desired Prof Joseph Lcidy, Prof E.
D. Cope, Mr. F. B. Meek, and Prof Leo Lesquereux, to present in detail
all the evidence they could secure in their respective specialties of Extinct
Vertebrata, Extinct Invertebrata, and Fossil Flora, bearing on the disputed
age of these formation.s, to decide, if possible, whether the strata in question
are Cretaceous or Tertiary. The various reports which these gentlemen
have furnished testify with what zeal, ability, and success these instructions
have been carried out.

UT

IV LETTER TO TOE SECUBTAKY.

If ohjoction is inaile to the use of the term "Lignitic" Group, I would
say that, in this work, it is restricted to a series of coal-bearing strata lying
above the Fox Hills Group, or Upper Crelaceous, and these are embraced
in ihc divisions Laramie and Fort Union Gronjis. It is well known tliat
there are in various parts of the West, especially along the fortieth ])arall('l
and southwestward, very thick beds of coal in the various divisions of tlx;
Cretaceous, extending down even into the ITpper Jurassic. Had this not
been the case, the more general term Lignitic would have been retained l)y
this Survey, in preference to any other.

As far back as 1859 it was my belief, tbunded on what appeared to be
sufficient evidence, (hat the sequence between the well-characterized Creta-
ceous strata and those of the LigniticGroup, as defined at that time, was
continuous, and that the chasm which was supposed to exist between the
Cretaceous and the Tertiary epoch would be found to be bridged over. This
belief was not based on strictly palceonfological evidence, for no well-marked
Cretaceous fossils were then known to pass up into the Lignitic or brackish
beds. But the physical conditions under which the sediments of the upper
strata of the Fox Hills Group were deposited indicated a gradual change,
from deep, quiet marine seas to shallow waters, which became at length
brackish and finally entirely fresh waters, during which time the purely
marine inverteljrate fauna perished, a brackish and purely fresh-water fauna
taking its place. This condition of the Lignitic Group covered a vast area
in 1hc Xorthwest, extending far southward, alonij the eastern slope of the
Rocky Mountains, to Denver, Colorado. As we proceed southward and
westward from the Missouri River, the brackish beds increase in thickness
until along the fortieth j)arallel they become three thousand feet or more,
indicating, so far as can l)c determined, no break in the sequence from the
Fox Hills Group to the purely fresh-water strata of the Wahsatch Grouj).

Dr. C. A. White, Pateontologist to the Survey under my charge, has
made a critical examination of these formations during the past season, and
he says tliat his investigations have fully confirmed the views expressed by
mc some years ago, and in(ricated i)y the pahcontological sludies of Mr. ]\Ieck,
that the Fort Union beds of the Upi)ei; Missouri River are the equivalent of
the Lignitic formation as it exists along the base of the Rocky Mountains in
Colorado. He also testifies to the equivalency of the latter with the Bitter
Creek series west oflhe Rockv i\I()untains. The.-<c views of Dr. Whit(> are con-

LETTER TO THE SECRETARY. V

firmed, not hy tlic discovery merely of one or twodouhtful species common lo
tlic strata of cacli of these regions, but by an identical I\Iollnscan fauna ranging
through the Avliole series in each of the regions named. Tiiis shows that the
strata referred to, all belong to one well-marked period of geological time.
Dr. White arrives at these conclusions, not merely because there is a similarity
of type in the fossils obtained from the various strata of the Laramie Group
willi those tiiat were before in question, but l)y reason of the specific identity
of many fossils tliat range from the l)ase of tiie Laramie Group up into and
through the strata that were formerly referred to the base of the Wahsatch.
Some of these species were found by Dr. AVhite in the Laramie strata on
both sides of the Rock}' Mountains, with a vertical range of not less than
three th(msand feet and a geographical range of more than a thousand miles.
The conclusion, therefore, becomes more and more apparent that while
till' [)rincipal groups of the Mesozoic and Cenozoic formations in the West
have each peculiar characteristics, and are readily recognized by the geologist,
tiiey really form an unbroken series of strata, not separated by sharply
defined planes of demarcation, either stratigraphical or palasontological. The
facts as we understand them at the present time would seem to warrant this
general division, viz.: a marine series. Cretaceous; gradually passing up into
a brackish-water series, Laramie; gradually passing up into a jiurely fresh-
water series, Wahsatch. It is also probable that the brackish-water beds
on the Upper Missouri must be correlated with the Laramie, and that the
Wahsatch Group as now defined and the Fort Union Group are identical
as a wiiole, or in part at least. The plants which are recorded in this volume
began their existence at the base of the Laramie Group, and continued
through the entire series, brackisli and fresh-water. The reason will now
become apparent why I have, in my former reports, called the Laramie
Group a transition series, or beds of passage, not as a distinctive name,
but only as indicating the fact that they seemed to bridge over the chasm
between the j)urely marine Cfetaceous and the purely fresh-water Tertiary.

The lack of animal remains in the Tjower Lignitic ]\[casures, especially
those of Colorado, is remarkal)h;. On the other iiaiid, all tlie coal-bearing
strata above the Cretaceous Fo.x Hills Group al)ound in well-preserved vege-
table remains. The comparatively few specimens of fossil j)lants obtained
by the Survey in Colorado up to the yc^ar 1870 pointed to the conclusions

VI LETTER TO THE SECRETARY.

which arc now reached, but the evidence they represented was deemed
inconclusive. Prof. Lesquereux was, therefore, desired to take the field
under the direction of the Survey, to study the Fossil Flora of these l)eds at
all the more interesting localities, from New Mexico through Colorado into
Wyoming and Utah. Some results of his researches, conducted during two
seasons, have already been given to the public in the Annual Reports of the
Survey, where the many new species discovered were named and briefly
characterized. Such description of fossil remains of plants, often represented
by mere fragments, was found to be inadequate to the full exposition of the
subject which science demands. It became necessary, therefore, to figure
these fossil plants with great care, in order that their characters might be
fully appreciated, and to compare them closely with those already known
from the different geological formations of Europe. By such representation
and examination alone, could safe conclusions be drawn respecting their true
geological relations.

The carefully drawn plates which illustrate the subject, prepared by
Prof. Lesquereux himself, or under his immediate supervision, place the
characters of these remains in the clearest light. The greater portion of
the text of this volume, forming Part Second, is devoted to the determination,
description, and discussion of the several species. The First and the Third
parts treat mainly of the geological bearing of the fossil plants and animals
upon the main question of the age of the Lignitic formations of the West,
and represent the conclusions derived from the study of the remains here
figured and described in connection with consideration of the evidence
afforded by the fossil animals.

The author states that his final conclusions do not differ materially from
those already advanced by myself, and he regards the evidence as conclusive
that the Lignitic Group is of Tertiary age. This result is gratifying, not only
as settling the question at issue, but as silencing criticism of the value and
reliability of the general work accomplished by the Survey under my direction.

Apart from the technical aspects of the scientific problem here solved,
the Lignitic formations of the West have an economic importance that cannot
easily be overestimated. Their wide extent and the number and thickness
of the beds of coal distributed through these strata confer a value not less
than that of the true Coal-Measures of tiie East, from the Mississippi to

LETTER TO THE SECRETARY. VH

Massachusetts. Their inipoituuce and practical value are enhanced by their
presence in a country otherwise almost destitute of fuel. These AVestern
Coal-Measures render rail communication between opposite sides of the
continent not only practicable, but easy; they make possible the settlement
of an otherwise scarcely inhabitable country, and are invaluable in the prose-
cution of the mining and manufacturing industries of the Rocky Mountain
region. The plants which afford this valuable combustible material merit
close study, no less from an economic than from a purely scientific point
of view.

Other scientific deductions than those already presented are derived
from such investigations. To the study of the plants of the older Coal-
Measures we owe not only our knowledge of the vegetation of the several
geological epochs, but also our recognition of the diverse climatic conditions
which marked successive periods during the slow formation of the continent.
Until recently, the physical influences prevailing during the progressive
modification of the earth's surface from the earliest periods to the present
time have been considered in this connection only by the European palaeo-
botanists. Europe has seen the appearance of many works upon the fossil
plants of all her formations; but these records, however rich and interesting,
are incomplete without comparison with those of other continents. Deduc-
tions respecting the possible uniformity of the climatic conditions of any
one period over the whole hemisphere, or regarding the origin and distribu-
tion of plants, and the actual character of vegetable life, remain unreliable
and wholly unsatisfactory so long as they rest merely upon local observations.
The scientists of Europe, fully aware of this, have regarded the study of the
fossil botany of America as of the utmost importance, and have received
with evident satisfaction the first contributions to the knowledge of the
subject from investigations conducted in North America. A number of
memoirs have already appeared upon the Fossil Flora of the true Carbonife-
rous or Coal-Measures of the United States. The publication of Professor
Lesquereux's Cretaceous Flora of tlie Dakota Group, forming Vol. VI of this
series of Reports, awakened great interest in the whole subject, and incited
fruitful discussions respecting the European formations of the same epoch.
The present volume, on the Tertiary Flora, opens a page of no less interest
and one still more important — one on which are traced the characters of a

VUI LETTEK TO THE SECRETARY.

geologic record whicli too long remained blank, while llie spirit of scientific
incjuiry was moving in other lines of research with such ellbctive energy.
The lively and widespread interest manifested by the people of the United
States in the progress of science would ensure the favorable reception of a
work upon a hitherto unknown subject, even though it did not relate, as this
one does, to one of the most practically valuable as well as scientifically
remarkable geological formations of the continent.

The study of palseo-botany acquires its highest interest when considered
in connection witli the plant-life of the present time. Fossil plants are
records of the past, engraven on the rocks — the legible documents which
enable the student to discern whence and how the Flora of to-day has
acquired its character. The study of recent vegetation is linked with that
of the long past as indissolubly as are the plants themselves related by
descent with modification from preexisting forms; and its rational interpre-
tation is possible only when the subject is viewed in the reflected light of
geological succession.

But the study of Fossil Floras may be brought to bear upon questions
of still greater magnitude and importance, even those of the origination of
continental land-areas as at present existing, and of their connection or sepa-
ration at certain periods of geologic time. To recognize, for example, that
the present American Flora includes types traceable back to the oldest
geologic formations, and that the continent has preserved certain peculiar
types, not found in Europe or elsewhere, through all the mutations of its
surface-features, would autliorize the deduction that these land-areas were
separated for a corresponding length of time. Such studies, again, bear
upon the problem, whether, as some believe, the North American Flora was
derived hy migration across intermediate land, either from Europe or Asia,
or whether, as others maintain, the Flora was indigenous and consequently
peculiar. Such considerations bring us face to face with one of the greatest
and gravest problems that the human intellect may aspire to solve, namely,
the origin and development of sj)ecies.

The value which attaches to the study of Fossil Floras as furnishing data
for general geological purposes has been often (lis[)uted. It is well under-
stood that palccontological or palajo-botanical material is more or less valuable
and reliable in proportion to the abundance and state of preservation of

LETTER TO THE SECRETARY. ix

specimeDs. In general terms, it may be said, that, in llie determination of
marine formations, the remains of fossil plants have little valne in comparison
with those afforded by fossil animals. But the conclusion of this volume, it
is hoped, slious that tlic study of fossil plants gives no less reliable data than
those afforded by animal remains ia the investigation of land formations.

During the whole course of his researches npon tlie Mesozoic and
Cenozoic Floras of the "West, the author has been in constant communication
with Ileer, Schimper, Saporta, and other eminent paIa3o-botauists of Europe,
who have commented npon the progress of his labors in the most flivorable
terms, and have unequivocably confirmed his conclusions. AVhatever differ-
ence of opinion may continue respecting the age of the formation from which
the plants treated in this volume have been derived, the memoir will ever
remain an eloquent witness to tlie learning and ability of its illustrious
author, and a monument to the science he has for years cultivated with the
most gratifying success. If he may not be said to have created palaeo-botany
in America, he has been foremost in fostering it, and has brought it to the
point of advancement that is matched only by the standard of excellence
that the most eminent of his European compeers have attained. -

F. V. HAYDEN,

United Slates Geologist.

Hon. Gael Scuuez,

Secretary of the Inlcrior, Wasliington, V. C.

T F — H

UNITED STATES GEOLOGICAL SURVEY OF THE TERRITORIES.

CONTRIBUTIONS

TO

THE FOSSIL FLORA

WESTEEN TEERITOIIIES.

P^RT II,

r.LW Yo; K

BOTANICAL

THE TERTIARY FLORA.

By LEO LESQUEREUX.

WASHINGTON:

OOVEENMENT FEINTING OFFICE.

1878.

XI

TABLE OF CONTENTS.

PART I.

Pagt.
The LiGNiTic Foumations ok Noktii Amkhica:

^ 1. Areal Distiibutiou I

^ 2. Stratigraphy of the Ligiiitic ami its Capacity for Combustiblt^ MiniTal In

§ 3. The Agi! of the Lignitic. indicated by its Geological Distribution and its Fauna '^I

PART II.

Description ok the Tertiary Fossil Plants ;i:{

CRYPTOGAMS.

Fiingi '.V.'i

Lichenes : 'M>

Algae :iT

Characeie 43

Mnaci 43

Lycopodiaceae 44

Filices 49

EhizocarpiiB (i4

CALAMARI^.

Equi8et<ace{e 07

PHSNOGAMS.

(iYMNOSPERMJi.

CycadinesB 70

ZamiesB 70

CONIFERS.

Cupressiuea; 7'^

AI)iotiueae 75

TiiJtiueie 84

MONOCOTYLEDONES.

FLUMACE^

Gramineaj 66

Cy peraceiB 92

COliONARIiE.

SmilaceiB 93

SCITAMINK.*;.

ZiDgiberacete %

Muaaceio 06

Mil

XIV TABLE OF CONTENTS.

ENSAT^.

Hydrocharidea) , _ _ gg

POTAME^.

Najadeo) gg

IXUVIALES.

Lemnaceo; 102

SPADICIFLOK^.

AraeeiB 103

Aroidea; IO5

MonocotyhiloncD iiicetta: sedis jq(j

PRINCIPE8.
PalmEB 107

DICOTYLEDONES.
Characters of haves I23

APETAL^.

AMENTACEJ3.

Myricaceaj j26

Betulaccx _ I3-

Ciipuliferaj 1^2

Salicineaj Igg

Plataneaj jgj^

Balsamilluie Igg

URTICINE^.

Ulmacca; Igy

Celtideai jgi

Moreas jgj^

OLERACE^.

PolygoueJB 208

Kyctiaginea) 209

PROTEINEJB.

ProteesB ___ 211

Laarioece 211

GAMOPETAL^.

Loniceroas 222

ASCLEPIADINEiE.

Oleaceae 228

DIOSPYRINE^.

KbenacetB 230

ERICINEiE.

EricaceiB 2:i3

POLYPETAL^.

UMBELLIFLOR^.

Araliace^B 235

Anipelidea) 238

Coriieaa 242

NysseiB 245

TABLE OF CONTENTS. XV

Page,
COnNlCULACE^.

Saxif rageaj 24b

rOLYCARPICiE.

Magiioliacca) 247

AnonaceiB 250

NYMPIIKINEJ!.

Nelumbono9 252

MALVOIDE^.

Buttneriaces 254

Tiliaceao •^■>''

ACERINEiE.

AceracejB 260

Sapiuilacea) 263

FltANGULACEiE.

Staphylaccaj 267

Celastro;i3 268

Ilicraj 269

Rbanine® 272

TEREBINTHINE^.

JuglandesB 283

Aiiacarcliaccie 291

Zanthoxjleaj 294

CAMCIFLORJi:.

Halorajjere 295

MYUTIFLOK^.

Myrtaceio 296

ROSIFLOK^.

Pomaceaj 297

Legum inosa: 298 .

Incerta; sedis 301

PART III.

AOE or THF- LiGNITIC FORMATIONS DETKRMINED BY THE CHARACTERS OF THE FOSSIL PlaXTS 309

Table or Distribution of Speciks 314

Table of Distribution of the Species of Point op Kocks 343

LIST OF ERRATA.

Id indicating the uumbcr of liiic8, 1 do not count the. heading of the p.nges.

Page 25, line 7 from top, read "Oslrea " foi- " Onlivo ".

I'age 41, line 5 from l)ottom, read " air) " for " '2A4 ".

Page 42, line 7 from top, read " 373 " for " 273 ".

I'age 47, line 12 from top, erase " stem ", which is re])eated.

Page 47, line 10 from hottom, read " laciiiia;" for ''^locinwi ".

Page 53, line 9 from bottom, add " Ung. in" before " Ueer".

Page 07, line 13 from bottom, read '• cellular " for " medullar "

Page 74, line 15 from bottom, read " fig. 2 " for " fig. 6 ".

Page 86, line B from top, read "103" for " 113".

P:tge 93, line 16 from top, add " c" after " tig. 45 ".

Page 115, Hue 13 from bottom, read " Plate IX" for " IV ".

Page lie, Hue 14 from bottom, read •' Plate XXXVIII " for "XVIIl ".

Page 124, line 15 from bottom, read " brachidodrome " for" brachiodroiiic ".

Page 133, line 17 from bottom, the quotation "Lci^qx.", etc., goes above, afier "p. 545. Mijrka Liidiviyii".

Page 134, line 9 from bottom, read " 412" for " 413 ".

Page 139, line 16 from top, read " 29 " for " 30 ".

Page 151, line 14 from top, read " 378 " for " 373 ".

Page 167, line 3 from top, add " PI. XIX " after " p. 25".

Page 171, line 15 from bottom, insert a comma in place of " that".

Page 178, line 12 from bottom, read " fig. 14 " for " 15 a ".

Page 183, line 17 from top, read " Oeynhausiana " for "(Eninghauaiaiia ".

Pago 184, line 19 from top, read " Oe!j:ih(^usiana" for " CEiiinghausiaiia".

P.ige 203, line 17 from bottom, add " II " after " Palx-out ".

Page 216, line 13 from top, read " Plate XXXVI" for " XXVI ".

Page 216, line 3 from bottom, read " camptodronie " for " brachiodrome ".

Page 219, line 7 from bottom, read " 1873 " for " 1874 ".

Page 220, line 8 from bottom, read "85 " for " 35 ".

Page 220, Une 7 from bottom, read " 29 " for " 294 ".

Page 234, line 19 from bottom, read " shallow " for " hollow ".

Page 277, line 4 from top, add " from" after " dentate ".

Page 284, line 4 from top, read " whose " for " its ".

Page 286, line 9 from top, erase " Supi)lernent, p. 8 ".

Page 298, lino 9 from bottom, read " PI. LXIII, fig. 8 " for " fig. 5 ".

Page 318, line 4 from toji, read " JErioiaulon " for " Jir'wcolon ",

LETTER TO THE GEOLOGIST-INCH ARGE.

Columbus, Ohio, June 18, 1877.

Dear Sir : I seixl you herewith my report on tl^e Tertiary Flora of the
Territories.

The work has three essential divisions.

In tiie first, the general outlines of the geology of the countries where-
from the specimens of i)lants have been obtained are briefly exposed. This
part is rather yours tlian my own. It could not be omitted, however, in this
volume; for it is advisable, for the understanding of the characters of the
floras, to have for reference an expose of the geographical and stratigraphical
distribution of the groups from which the specimens are derived. The quota-
tions on the subject arc carefully credited to the original authors.

The second part is the description of the species of fossil plants.

The third reviews the evidence aflbrded by the fossil flora to the age and
the relation of the different groups of the Lignitic formations. The conclu-
sions derived from this review may not be generally admitted; they are,
however, confirmed by the careful comparison of the characters of the
vegetable remains. This part is prefaced by a few remarks upon the progress
of the work from the beginning of my connection with your Survey. I have
there mentioned the names of all those who have contributed to the Flora by
researches and communications of specimens. This mention is rightly due
to all, but especially to some friends who have worked hard and given much
time, without any remuneration, to procuring materials often of great value
to American jjalEeontology. You will certainly find that they are all entitled
to a copy of the Flora, as they are also here to the expression of my most
sincere thanks.

This Flora of the North American Lignitic is like a supplement to that
of the Cretaceous Dakota Group. Both together constitute a historical
record not less interesting to Botany than to Geology ; for, beside the evi-
1 T F ^

2 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

tlcnce afforded on tlie relation of the groups of the formations, they expose,

as in a written book, documents illustrative of the origin and the successive

development of some of the predominant and more interesting types of the

present vegetation of this country.

Allow me here to give expression of my gratitude for the assistance

which you have given to my work and the constant interest by which you

have greatly encouraged it.

Very truly and respectfully, yours,

L. LESQUEREUX.
Dr. F. V. Hayden,

United States Geologist, Washington, D. C.

PART I.
INTRODUCTION^

THE LIGNITIC FORMATIONS OF NORTH AMERICA.

^ 1. — Areal distribution.

The country west of the Missouri, and to the base of the Rocky
Mountains, is for nearly six hundrci miles— as from Omaha to Cheyenne,
or from Kansas City to Denver— a vast plain, with a gradual slope, unappre-
ciahle to the eyes, and without any of those land irregularities which gen-
erally, breaking the stratification by upheavals or denudations, expose to view
the rocks composing the crust of the land surface. The ascending grade
from the Missouri River toward tlie mountains does not average more than
ten feet per mile, and as the Cretaceous strata exposed below Omaha above
the Permian Measures, are nearly horizontal, they pass, of course, toward
the west under the different stages of the Tertiary or under more recent
deposits. The great uniformity of the plains and the absence of exposed
rocks prevent the distinct tracing of the line of demarkation between the
Cretaceous and the Tertiary. As far as it is known in the States of Iowa,
Kansas, and Nebraska, the average width of the belt occupied by the Dakola
group, which is in that country the lowest member of this formation, is from
sixty to one hundred miles.* Over this appear the Upper Cretaceous groups,
wiiich, where they have been observed along the Missouri River, have a
thickness of more than two thousand feet, or, for the whole Cretaceous forma-
tion, two thousand five hundred feet. As from O malyi to Cheyenne , which

» Eeport of the United States Geological Survey of the Territories, by Dr. F. V. Haydeu, vol. vi,
Cretaceous Flora, p. 12. 3

4 UNITED STATES GEOLOGICAL SUKVEY— TERTIARY FLORA.

is five thousand feet liiglicr, the distance is five liundred miles, admitting
horizontality of the measures and uniformity of the grade, tlie belt of the
Cretaceous should occupy about half the width of the i^lain between the
Missouri River and the base of the mountains. This estimate is, however,
too high; for, along the Missouri River, Dr. Ilayden fixes the eastern limits
of the Cretaceous, or the appearance of the Tertiary over them, at Fort
Benton;* and considering tlie Lignitic area as marked in his geological map
of the Yellowstone and Missouri Rivers.f that of the Cretaceous would, from
east to west, be about one-third the width of that of the Tertiary.

The first record we have of the area of the Lignitic, or at least of its
wide extent along the Missouri and the Yellowstone Rivers, is obtained from
the narration of the voyage of Lewis and Clarke in 1804. The following
passage is copied from R. C. Taylor's Statistics of Coal, p. 174 : —

"The coal, or lignitic, was first observed twenty miles above the
Mandan village. The bluffs on each side of the river are upward of one
hundred feet high, composed of sand and clay, with many horizontal strata
of carbonated wood, resembling pit-coal, from one to five feet each in thick-
ness, and occurring at various elevations above the river. At fifty miles above
the village, similar coal seams were noted; bnt here they were observed to be
on fire, emitting a quantity of smoke and a strong sulphurous smell. Further
on, the same sulphurous coal continued for eighty miles more ; strata of coal,
frequently in a state of combustion, appearing in all the exposed faces of the
bluffs. The quality of the coal improved as the party advanced, near the
mouth of the White River, eighty-five miles farther, aflTording a hot and
lasting fire, but emitting very little smoke or flame. Thence forty-seven
miles, to the Yellowstone River, and at a bluff eight miles up that stream,
were several strata of coal. For fifty miles above the junction of the Yel-
lowstone and the Missouri, there were greater appearances of coal than had
yet been seen, the seams being in some places six feet thick; and there were
also strata of burnt earth, which were always on the same level with those
of coal. The explorers had thus far traced this lignite formation along the
banks of the Missouri for a distance of three hundred and thirty miles. The
horizontal formation of clay, loam, and sand, with fragments of coal in the
drift of the river, extended three hundred miles more, to Muscleshell River,

• Dr. F. V. H.iyden, Annual Ucport, 18G9, p. 48.

t United Stales War Department Map of thu YellowHtone and Misbouri Eivers, e.xplored by Capt.
W. ¥. Raynolda and Lienl. 11. l',, Miiynadier, 1850-60.

AREAL DISTRIBUTION OF THE LIGNITIC FORMATIONS. 5

or six huiidred and twenty miles from Mandau Village. Even above this
point, wasiiod coal continually appeared ou the shores of the river, and at
Elk llajjids, eight hundred miles iVom Fort Mandan, the liigh horderiug
lilufTs were still composed of horizontal beds of clay, brown and white sand,
soft, yellowish-white sandstone, hard, dark-brown freestone, and large, round,
or kidney-shaped nodules of clay iron ore. Coal, or carbonated wood, similar
to tiiat previously observed, was also seen, and was accompanied with burnt
earth, prol)ably the result of the si)ontaneous combustion of the coal, as was
noticed for hundreds of miles below. After reaching the Grand Fork of the
Missouri, and ascending two or three days' journey up Maria's River north-
ward, it was remarked that j)rccisely the same geological character and coal
strata prevailed for more than sixty miles. So far, therefore, the exj)loring
party had been traveling through or over a ligneous deposit of singularly
uniform character for no less than nine hundred and eighty miles, following
the windings of the river. Pursuing the South Fork toward the Great Falls
of the Missouri, coal was still observed in Idulfs of dark and yellow clay at a
distance of two thousand four hundred and fifty-four miles up that mighty
river, and it was not until near the base of the Rocky Mountains, and after
one thousand miles of traveling across it, that this great region of coal-beds
and lignites was passed."

"On the return, Captain Clarke descended the Yellowstone from about
north latitude 45*^ to its mouth, 48° 20', and everywhere found the same
series of coal and variously colored clays and soft sandstones as was traversed
in ascending the Mis.souri. Below the Big Horn is a large stream falling in
from the south, whose Indian name implies the Coal Creek, from the great
quantity of this mineral upon its border. The same coal series continued to
the confluence of the Missouri, exhibiting uninterruptedly for seven hundred
miles, in addition to the thousand previously traversed, the vast persistence of
this formation. The enormous area of similar strata is further shown by
the decoloration of all the tributaries that enter the Missouri from both the
south and the north, from the forty-second to the forty-ninth degree of north
latitude."

It is from the records of those celebrated explorers especially, also from
those of Audubon and Harris, Sublette, Frdmont, Emory, etc., for the United
States, from the explorations in British America by Dr. Richardson, Drum-
mond, and Captain Franklin, that Taylor obtained the data for the delineation

6 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

of the area of the Lignitic in the map of his Statistics of Coal, 1848.* As
an introduction to it, he remarks (p. 23) on that enormous range of brown
coal, apparently of the Tertiary period, which follows the eastern flank of
the Rocky Mountains, from near Mexico even to the Polar Sea: — " Nature", he
says, "has indeed worked on a truly gigantic scale. We see here a deposit of
brown coal so extensive that the magnitude of its proportions is far from
being defined ; yet enough is known to show that it exceeds in longitudinal
range and breadth all others of the present surface of our planet. So far
seems to be established, that, allowing li])erally for interruptions in continuity,
supposing that any such exist, it occupies thirty-five degrees of latitude, or
near two thousand five hundred miles, following the oblique range, and has a
maximum breadth on north latitude 48° of four hundred miles; the whole
area, as near as we can venture to compute, being two hundred and fifty thou-
sand square miles, or one hundred and sixty millions of acres, more than
twice the size of Great Britain. Compared with this, the largest coal-fields
in the world are absolutely small."

Audubon and Harris ascended the Missouri to the mouth of the Yel-
lowstone River. In the account of their voyage, they give, on thd Tertiary
strata of the country, details in accordance with those recorded by Lewis and
Clarke.f The whole series of strata, for many hundred miles prior to reach-
ing this formation, is described as perfectly horizontal; the upper part of each
bed or rock being successively intersected by the angle of descent to the
river. The Tertiary group is indicated by the remarkable strata which form
the picturesque hills noticed by travelers, and called Mauvaises-Terres by the
trappers and voyageurs. Mr. Plarris counted in one place eight seams of coal
between the river bank and the top of the bluff, varying from six inches to
four feet in thickness. This coal, he observes, is very light, and ignites with
difficulty, emitting a very unpleasant odor while burning. Fossilized wood
is very abundant, occasionally much flattened by the pressure of overlying
strata. Mr. Bell was the only one of the party who had an opportunity of
witnessing the burning of the cliffs about thirty miles above the Yellowstone,
on the northern bank of the Missouri, and all agree in attributing their burning
to the spontaneous combustion of the coal. Mr. Harris states that tlie coal-
seams commence in the upper part of Nicollet's great Cretaceous clay bed,

* Chart showing the imsition of thu coal-fields ou the surface of the k'"^**; ^y Kichard Taylor.

♦ Proceedings of the Academy of Natural Sciences of Philadelphia, May, 184.5.

AREAL DISTRIBUTION OF THE LIGNITIC FORMATIONS. 7

and further, that there occurs in the same formation a substance like petro-
leum in color and consistence, but without odor; that from the specimens
brouglit home by the last-named traveler from the vicinity of Fort Union,
near the confluence of the Yellowstone and Missouri Rivers, we derive incon-
testable proofs of a fresh-water formation. Among other strata exposed in a
cliiFncar the fort are thin beds of clay and argillaceous rock, l)olh conlaining
three or four species of fresh-water univalve shells. There is besides a rock
twenty or thirty feet thick, which also contains proofs of fresh-water origin
in bivalve shells, leaves of deciduous trees, and bones apparently of mam-
miferous animals.

Details in accordance with those given above are reported from the belt
of the Lignitic surveyed north of the limit of the United States and British
America. They extend to Vancouver, even to the Arctic land of Disco,
Greenland, and southward along the Pacific slope to the southern extremity
of the continent. They are, however, still less precise, and evidently Mr.
Taylor refers to the Tertiary coal deposits of different geological ages.
Hence, we have as yet nothing definite in regard to those mentioned coal
strata. Even we may say that scarcely anything positive was known of the
great North American Lignitic when Dr. F. V. Hayden undertook the work
of exploration and began liis researches, in 1854. It is therefore from the
numerous publications of reports and memoirs of the celebrated geologist
that I have to take most of the reliable facts exposed in this introduction.

I cannot enter into the examination of Dr. Hayden's researches without
remarking on the accuracy of the data whicii he has exposed in his numer-
ous Reports and Memoirs on the Geology of the Western Territories. Begin-
ning in Kansas and Nebraska, he has followed the explorations foot by foot,
so to say, not omitting a single fact worth the attention of the geologist.
Collecting specimens of ore, of minerals, of animals, of plants, he has by and
by traced the outlines of the present and ancient history of" these Western
Territories ; and calling to his assistance all the specialists who might ren-
der his work more complete, he has fdled the pages of a truly invaluable
record. For now, the natural history of those western regions, mostly
unknown a few years ago, is exposed as distinctly and precisely as may be
that of any of the oldest States of the Union. The agricultural and mineral
resources, the geographical and stratigraj)hical distribution, the fauna and flora
of the present epoch, those of tiic former geological periods, even the phys-

8 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

ical circumstances influencing the character of the countries surveyed, all have
been considered and studied by Hayden. His researches show tlie constant
vigilance and circumspection of a master attending to the performance of a
great work, the building of a monument whose plan has l)een prepared by
serious scientific studies. I speak here by experience, for in the part assigned
to me I had to follow, so to say, the footsteps of the master, and found that
even the matters of the least importance had been already recorded by him,
and outside of my specialty, the study of the paleobotany of the Cretaceous
and the Lignitic, I could scarcely find anything worth mentioning as new.

The first explorations of Dr. Hayden over the western coal regions, to
which the name of Great Lignitic is generally and appropriately given, were
extended first up the Missouri River from the first appearance of the Tertiary
strata near Fort Clarke to the mouth of the Yellowstone, and thence up that
river to a point near the mouth of the Big Horn for a distance of about six
hundred miles. He considers that the area of the Lignitic formations cannot
be, on the Upper Missouri, less than one hundred thousand square miles,
without taking into account the belt which extends far north across the
boundary of the United States into the British Possessions.*

On the geological map of the Yellowstone and the Missouri Rivers,
prepared for the explorations of Capt. W. Raynolds and Lieut. H. E. Mayna-
dier for 1859-60, the part colored as Tertiary Lignitic by Dr. Hayden, who
had charge of the geological researches, indicates a wider area, not less than
one hundred and twenty-five thousand square miles, and this only from the
boundary of the British Provinces to the Black Hills. Between these and
the Rocky Mountains, south to the Nebraska River, the Tertiary belt is still
continued over a surface of about sixteen to seventeen thousand square miles.
Farther south we have not as yet any map exposing the distribution of the
Tertiary. Prof Hayden, considering this part of the area occupied by the
Lignitic, says:t — "We may trace it southward in a broad continuous belt across
the Yellowstone River, between the Black Hills and the Big Horn Mountains,
until it is overlapped by the White River group about sixty miles north of
Fort Laramie. If we continue southward along the base of the Laramie
Range, we find that the Lignitic group reappears about ten miles south of the
Union Pacific Railroad ; that where the White River group and the Lignitic

' Report of the United States Geological and Geographical Survey of the Territories, 1874, p. 20.
t Loc. cit., p. 2G.

AREAL DISTRIBUTIONS' OF THE LIGNITIC FORMATIONS. 9

come in contact, the former is superimposed to the latter; and that really
the White River group formed a vast basin subsequent to the existence ot the
great lake on whicli the Lignitic sediments were deposited." He adds:
"We find also, by examining the White River group along the base of the
mountains, that the Laramie Range for.ncd a l>arrier that prevented it from
extending into the Laramie Plains; but the evidence is clear that at the tune
of the existence of the great Lignitie lake or sea, this barrier did not pre-
vent the water communication with the Laramie Plains. Lideed, the evi-
dence seems quite elear that, with the exception perhaps of some isolated
peaks rising above the waters, there was no mountain barrier where we have
n<,w the Laramie Range. Therefore, with the exception of the Bear River
aud Coalville group, we may connect the coal-bearing beds of the Laramie
Plains and Colorado with the vast group in the Northwest."

The southern l)asin, generally named tlie Colorado Basin, is followed,
nearly without interruption, from a (.^^ miles south of Cheyenne to New
Mexico It is continuous to the South Platte below Denver, where it is
covered by a ridge of hills, the Monument Creek group, and then reappears
near Colorado City. On the Arkansas River, near Canon City, outlayers of
the Licuitic have been left upon the Cretaceous, which by denu.lation is
exposed over nearly the whole valley; and south of the Arkansas, or from the
Spiuish Peak the belt becomes continuous again to the Raton Mountains, m
New Mexico, with outlayers or isolated patches appearing as far south as

Albuquerque.

The southern Lignitic covers, therefore, an extensive area. It cannot
be estbnated, however, for the reason that it is cut by more recent deposits
at some places, as south of Denver, and by erosions along the Arkansas River,
and especially because its width from the mountains to the east is unknown.
The upheaval of tlie mountains has exposed the edges of the Tertiary strata
with those of the underlying formations, throwing them up into a series ot
hogbacks, which pass very abruptly from an inclined, even vertical position,
in the proximity of the mountains, to a horizontal direction toward the plains.
All along the mountains, the Lignitic is at the upper stage, and therefore it is
covered merely in passing to the plains by the more recent deposits of the
surface But how far it extends, or it is accessible for coal, has not ye been
ascertained. Shafts have been sunk east of Denver about ten miles, and thick
beds of coal or lignite have been reached at a moderate depth. Other

10 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

shafts, near Plattevillc, north of Denver, have also reached coal quite near the
surface, showing that the belt of the Lignitic extends, locally at least, to a
great distance eastward from the base of the mountains.

§ 2. — Slratigraphi/ of the Lignitic and i/s capacihj for combustible mineral.

On this subject we have documents more precise than for the former,
though they are not complete as yet ; for the amount, thickness, and chemi-
cal value of the coal or Lignitic beds is tar from being exactly known, or
even far from being possibly estimated. Where the Lignitic has been recog-
nized from its base, it has been seen overlying the Upper Cretaceous strata,
whose section is exposed in the Annual Report of Dr. Hayden for 1870, p.
87. The two upper groups, the Fort Pierre group. No. 4, and the Fox Hill
bed. No. 5, have generally an abundance of invertebrate fossil remains, and a
peculiar lithological composition, which makes them easily recognizable. In
the North Basin, or the Fort Union group, the superposition of the Lignitic
to the Cretaceous is not marked by any definite line of demarkation. Indeed,
this line is seen nowhere, neither in an abrupt change of the compounds, nor
in an unconformable stratification, nor in the character of the faunas. On this
subject, Dr. Hayden remarks :* — "When we bear in mind the fiict that where
the Lignitic has been seen in contact with the last Cretaceous beds, the two
have been found to be conformable, however great the upheavals and the dis-
tortions may be, while at the junction there seems to be a complete mingling of
sediments, one is strongly impressed with the probability that no important
member of either system is wanting between them." And at another pagetf —
"That the passage from the brackish- to the fresh- water beds of the Tertiary
is not marked by any material alterations in the nature of the sediments, nor
have we, as far as it is known, any reason for believing that any climatic or
other important physical changes, beyond the slow rising of the land and the
consequent recession of the salt and brackish water, took place during the
deposition of the whole of th(! oldest members of the Tertiary."

In his Geological Report on the P^xjdoration of the Yellowstone and
Missouri Rivers, IS.oO-fiO, Dr. Hayden remarks upon the Lignitic of the
Yellowstone River (p. 58) : — " Passing up the valley of the Yellowstone, we see
the gray sandstone Tertiary, which we have o])served to cover the Cretaceous
nearly to the foot of the blulis. The junction of these formations is quite

"Annual Kcpoit, 1874, p. '2-1. t Same Rcpml, p. a'>.

STRATIGRArnY OF THE LIGNITIC FORMx\TIONS. H

well marked on both sides of the river. For a considerable distance both
above and l)elow Fort Sarpy, a l)cd of sandstone forms nearly vertical bluffs
on l)otli sides of the river, which 1 iiiid it difficult to locate. Cretaceous Nos.
4 and 5, composed of yellowish-brown indurated clay, with concretions con-
taining Baculites ovatiis, Rostellarin, etc, in great abundance, occur, passing
into a dark gray coarse-grained sandstone, containing also Baculites ovatus,
Aricida, like A. Nebrascensis, and an Ostrea, new species. This also passes
into a sandstone having a most ragged front, from atmospheric agencies and
the difference in the consistency of the material composing the bed. It is in
the main a coarse-grained, friable, ferruginous yellow sandstone, but contain-
ing vast numbers of concretions ; some a reddish-yellow arenaceous lime-
stone, others sandstone; some nearly compact, with laminai; others divided
into thin layers, the harder portions projecting out beyond the friable ones.
The harder layers lie in the vertical cut, usually from five to thirty feet long.

"The layers are quite irregular in their horizontal fracture, the wliole
bed exhibiting indications of having been deposited in moving waters. May
it not be the transition bed from the Cretaceous to the Tertiary epoch, the
foreshadowing of the Tertiary period?"

In reviewing the whole of the Reports of Dr. F. V. Hayden and of his
assistants, we find similar descriptions of the same great sandstone forming the
base of the Lignitic Measures. My own section of the sandstone overlying
the Cretaceous No. 4 on the Purgatory River, near Trinidad, New Mexico, is,
as will be seen, like a more detailed repetition of Dr. Hayden's description of
the so-called transition sandstone, and also the other sections of the Lignitic
productive measures overlying it expose the general distribution of the Lig-
nitic beds, as indicated by the numerous sections given in the same Report
of Dr. Hayden of the Upper Missouri, or North Lignitic group, thus record-
ing the same characters of the measures at both extremes of the North
American basin.

As an example of the distribution of the Upper Lignitic, I copy the

section of the Pumpkin liutte, between the Black Hills and the Big Horn

Mountains, in the southern part of the North or Missouri Lignitic. It is

in descending order:* —

Feet.

1. Light yellow friable sandstone, with numerous rusty seams 75

The compact bed of .*;andstone caps all the hills, and gives
them the flat, table-like surface which they present at a distance.

'Report, 1659-60, j.. 73.

12 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

2. Alternate beds of lignite, gray and yellow ferruginous friable

sandstone, with bluish ash-colored, gray and yellow reddish tinged
marls and clays, with three seams, of one or two inches thick, of
impure lignite 125

3. Indurated yellow and ash-colored marls, with three small seams of

impure lignite, with one thin layer, six inches, of reddish-yellow
sandstone 60

4. Thin veins of eight inches of impure lignite, with numerous fine

crystals of selenite and masses of petrified wood.

5. Variegated clays and marls, with much sulphurct of iron and two

small seams of lignite 33

6. Impure chocolate liguite, with clay underneath, and large quanti-

ties of selenitic crystals 2

7. Light gray and bluish ash-colored indurated sandstone, laminated

clay and marls, with one or two seams of chocolate-colored im-
pure lignites 148

This section, recording four hundred and twenty-eight feet of strata of
the upper part of the Lignitic Measures, is like the part overlying the pro-
ductive measures of Canon City coal, as given in my report (1872, p. 324).
Here we have a capping of liard sandstone, two hundred feet, over scarcely
productive measures, formed by an alternation of beds of soft clay or soap-
stone, with an abundance of silicified wood, thin seams of lignite (the outcrop
of one near the top indicating two feet), beds of clay hardened and black-
ened by carbonaceous matter, etc. Most of the sections of the great Lignitic
basin of the north are more generally or mostly of the upper strata. The
thickness of its lower coal-measures is, however, locally very great; for Prof.
Hayden, in his Report (1874, p. 21), says that the lower brackish-water beds
arc more than two hundred feet in thickness, and that those that are purely
fresh-water must reach an aggregate thickness of tliree thousand to five
thousand feet, with from twenty to thirty beds or seams of lignite (not including
thin seams of an inch or two, which are very numerous). The lignite beds
average from six inches to ten feet in thickness.

Though the di.>itril)ution of the strata of the southern basin has been
distinctly and specially exposed in numerous reports of Dr. Hayden and his
assistants, as T have myself carefully surveyed a large part of it — that extend-

STEATIGRAPHY OF TDE LIGNITIC FORMATIONS. 13

ing along the base of Uie inoiinfaiiis iVoin tlic Raton to Cheyenne, and thence
along the Union Pacific Railroad to Evanston— I shall especially quote from
these observations given in detail in Dr. Haydcn's Annual Report for 1872.
Perhaps one of the finest exposures of the Lower Lignitic Measures in
regard to its relation to the Cretaceous is that of the base of the Raton
Mountains, a few miles south of Trinidad, and that of the bluffs on the Pur-
gatory River, opposite this last place, and mentioned above. The l)ase of tlu;
Raton is composed of a series of heavy, mostly whitish, sandstone, which is
conformably superposed to the black shales of the Cretaceous No. 4. This
sandstone is also conformably overlaid by the productive Lignitic. The
whole section, being fully exposed from top to base, is as follows:* —

LIGNITIC.

Ft. In.

1. Sandstone and shale covered with pines ^0

2. Soft shale alternating with soft clay (soapstone) 35

3. Outcrop of lignite, indifferent 2

4. Soft shale and fire-clay -^ ^^

5. Lignite outcrop, thin

6. Hard gray shale with fossil plants at basef 30

7. Shaly hard sandstone in l)ank ^

2

8. Soapstone shale

9. Lignite outcrop, good

10 Fire-clay and shale ^^ ^

11. Lignite bed, exposed

12. Fire-clay ^ ^

13. Soft shale 2^ ^

14. Lignite, opened

15. Fire-clay ^ "

16. Ferruginous and slialy sandstone, covered '^^ ^^

300 G

SANDSTONE.

17. Brown-reddish shaly sandstone, with debris of land vegetables. 37

18. Yellow shaly sandstone full of Fucoids ^ ^

• Annual Report, 187-2, p. 319.

I At a hUoi t (libtancf, tlm sliulo passes* to HanilHtoiif, No. 7.

14 UNITED STATES GEOLOGICAL SUKVEY— TEKTIAKY FLORA.

19. Ferruginous sandstone, barren 11

20. White compact sandstone, in bank, and barren 28

21. Hard white sandstone, in bank, witii Fucoids 10

22. Soft white sandstone, with Fucoids 32

23. Very hard block sandstone, barren 19 6

24. Ferruginous sandy shale, with Fucoids 6 G

25. White sandstone, barren 5 G

2G. Ferruginous sandy shale, willi Fucoids 8

27. Red shaly sandstone, with great abundance of Fucoids , . 3

28. Hard white sandstone, in bank, \vith some Fucoids 12

178
Between the last stratum^No. 27, and the Cretaceous black shale, no
nuiddy or brackish beds are seen. The transition is remarkai)Iy clear, but,
indeed, not more marked than it is between some beds of the Lignilic. The
characters of tiie lower group, one hundred and seventy-eight feet, from No.
17 to 28, are clearly described after the section, as follows: —

1. Its general color is whitish-gray; so white, indeed, sometimes, that
the lower strata, seen from a distance, appear like banks of limestone.

2. Though generally hard, it weathers by exfoliation under atmosplieric
influences, and its banks are thus molded in round undulations; and as it is
locally hardened by ferruginous infiltrations, it is often, too, concretionary or
grooved in cavities, so diversified in size and forms that sometimes the face
of the cliffs shows like the details of complicated architecture.

3. It is entirely barren of remains of animals.

4. On the contrary, from the lowest stratum to its upper part, it abounds
in well-preserved remains of marine plants or Fucoids, which in some localities
are seen even in the sandstone over lignite beds.

5. In its upper part, the sandstone or the shales of this group are mixed
with broken debris of land vegetation, with which also Fucoidal remains are
found more and more abundant in descending.

The disposition of the strata and their compounds is about the same on

the otiier side of Purgatory River, opposite Trinidad, where the section is

from top downward : * —

Feet.

1. Hard, ferruginous, shaly sandstone, with few remains of Fucoids, but

abundance of debris of land plants 25

• Annunl Report, 1872, p. 320.

STRATIGRAPHY OF THE LIGNITIO FORMATIONS 15

2. Hard, whitish sandstone full of Fucoids 57

3. Shaly sandstone, with abundance of Fucoids 50

4. Soft, laminated, ferruginous, sandy clay, with Fucoids 11

5. Ferruginous shale, with Fucoids 4

6. White block sandstone, barren 5

7. White sandstone, with Fucoids 22

8. Ferruginous shaly sandstone, with Fucoids 33

9. Black shale No. 4, Cretaceous 147

10. Covered space, sandstone and shale, to bed of river 153

In both these sections, the remains of marine plants are remarked in
most of the sandstone strata and their intermediate clay beds, and as abundant
at the base as near the upper part; and, in this last section, they are seen
mixed with fragments of land plants, even to the top of the sandstone,cut like
a tower at the point of the highest hill facing Trinidad.

In passing from the black shale of the Cretaceous No. 4 to this group
of sandstone beds overlying it, the fliiference in the characters is striking,
not only in considering their compounds, but in the class of fossil remains
which they contain; the traces of deep marine life predominating in the black
shale, while here tliey have totally disappeared. The absence of the Upper
Cretaceous formation No. 5 might be taken into account for explaining this
difference; it is not the case, however, tor, as seen above, the Upper Creta-
ceous sandstone beds are as definitely characterized by their fossil remains as
a deep marine formation as the second group No. 4 Now, at the Eaton, in the
sandstone above No. 4, marine life marks its activity only by the abundant
remains of Fucoids, indicating by their growth a comparatively shallow water.
They attest, therefore, a slow upheaval of the bottom of the sea, in wliicii
they appear to have lived, for their stems penetrate the sandstone in every
direction. And this indication is still more manifest in the great abundance
oi dcbria o{ land plants, which, apjiarently ground by the waves, seem to have
been thrown upon the shore and mixed in the sand with Fucoidal remains.
This slow upheaval and its result in tlic fbruiatiun of a new land are read as
in a book in the fossil remains of this group of sandstone, and every observer
should forcibly admit that these memorials of old expo.se the beginning of a
new era, or of what we call a new formation.

It has been seen already that Dr. Ilayden has everywhere remarked the
same distribution, the same confbrmal)ility of stratification, the same charac-

16 UNITED STATKS GEOLOGICAL SURVEY— TERTIARY FLOllA.

ters of the lower sandstone, as well in the northern area of the Lignite as in
the southern, and has come to the same conclusion expressed liere, that it
exposes a gradual change or transition, by the slow upheaval of the land,
merely a passage from a marine Cretaceous formation to a Tertiary land
formation.

In the report above quoted, sections are given of the distriI)ution of the
Lignite at Canon City near Pueblo, Gehring's Coal near Colorado Springs,
Golden, etc., which represent the same general distribution of the strata,
with inere local modifications, which do not affect in any way the general
characters of the group. I can give only a few of the most important sec-
lions, especially those which give a satisfactory representation of the capacity
of the measures for combustible mineral.

At Golden, seventeen miles west of Denver, the Tertiary Measures,
thrown up by the upheaval of the mountains against a basaltic ridge parallel
to their base, have been forced up in a vertical position, and thus, from this
place to Coal Creek, the Lignitic beds are exposed and worked from their
edges, their thickness varying from four to fourteen feet.

Further north, in the Boulder Valley, the measures come to their normal
position, dipping in various degrees from the mountains toward the plains,
and at Marshall a tine exposition of the Lignitic is presented, as seen in the
following section. It is copied from Dr. Ilayden's Report, 18G9, p. 129, and
is scarcely different from that published before, from tiie same locality, by
Dr. John L. Le Conte, and also from that which I received later from the
proprietor of the coal : —

48. Drab clay, with iron ore along the top of the ridge.
47. Sandstone.
46. Drab clay and iron ore.
45. Coal (No. 11), no development.
44. Drab clay.
43. Sandstone 15 to 20 feet.
42. Drab clay and iron ore.
41. Coal (No 10), no development.
40. Yellowish-drab clay, 4 feet.
39. Sandstone, 20 feet.

38. Drab clay, full of the finest quality of iron ore, 15 feet.
37. Tliin layer of" sandstone.

STRATIGRAPnY OF TIIK LIGXITFO FORMATIONS.

17

?,G.
35.
34.
33.

32.
31.

30.
29.

28.
27.
2(5.
2).
24.
23.
22.

21.

20.
19.
18.

17.
16.
15.
14.
13.
12.
11.
10.

9.

8.

7.

6.

5.

4.

o

CO

Coal (No. 9), nearly vertical where it has been worked, 12 feet.

Arenaceous clay, 2 feet.

Drab clay, 3 feet.

Sandstone, 5 feet; then a heavy seam of iron ore; then 3 feel of drab

clay ; then 5 feet of sandstone.
Coal (No. 8), 4 feet.
Drab clay.

Sandstone, 25 to 40 feet.
Drab clay, 6 feet.
Coal (No. 7), 6 feet.
Drab clay, 5 feet.

Sandstone, with a seam of clay 12 to 18 inches intercalated, 25 feet.

Drab clay, 4 feet.

Coal (No. 6), in two seams, 4^ feet.

Drab clay, 3 to 4 feet.
Yellowish, fine-grained sandstone, in thin loose layers, with plants, 5 to

10 feet.
] c' f Drab clay; excellent iron ore. "'j
} ^{ Coal (No. 5), 7 feet. } ]f) feet,

j « [ Drab clay. j

Sandstone, dip 11°. This sandstone has a reddish tinge, and is less

massive than No. 14.
Drab clay. "|

Coal (No. 4). '} 20 feet, obscnre.
Drab clay. J
Sandstone, massive, 60 feet.
Drab clay.
Sandstone.
Drab clay.
Coal (No. 3).
Drab clay.
Sandstone, 25 feet.
Drab clay.
Coal (No. 2), 8 feet.
Drab clay.

Sandstone, about 2.") feet.
2 T p

18 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

3. Drab fire-clay, 4 feet.

2. Coal (No. 1), 11 to 14 feet.

1. Sandstone.

This section shows eleven beds of coal, some of which are worked, on a
thickness of six to fourteen feet; this in about four hundred and fifty feet
of measures. An analogous distribution is recorded by other sections in the
Boulder Valley and northward to ten miles south of Cheyenne.

Passing westward from Cheyenne, along the Union Pacific Railroad, the
Lignitic measures over the Laramie Plains are covered with more recent
deposits. The Cretaceous reappear in the valley of Rock Creek, and from
Medicine Bow to Carbon tlie Lignitic is exposed again. At this last locality
we have a section of the mines through one hundred feet of measures,
exposing three beds of good coal, which have been actively worked since the
construction of the railroad. The section at the shaft is, in descending, —

Feet.

Shale, clay, and sandstone at top 35

Ferruginous clay, with a profusion of dicotyledonous leaves 3

Clay shales and sandstone, with plants at top 18

Coal (main) 9

Fire-clay and shale, with dicotyledonous plants 20

Coal 4

Fire-clay and shale 8

Coal 4

From Carbon to Black Buttes, geological disturbances bring to the sur-
face older formations in the Rawlins's Basin, but the Tertiary soon reappears
ten miles farther west, in entering the so-called Bitter Creek series, near
Separation, where a bed of coal, reported eleven feet, has been exposed ; far-
ther, at Creston, where another coal-seam, four feet, has been passed by a bor-
ing eighty-three feet from the surface; then at Black Buttes Station, where
two beds of coal, one four and the other eight feet, are exposed and worked.
In following the railroad passing along the anticlinal ridge whose axis is near
Salt Wells, to Rock Springs, coal strata are still exposed at Hallville and Point
of Rocks. At Rock Springs, two beds of coal are worked, as at Black Buttes,
one four and one eight feet; and besides, as seen by the records of the borings
for water made at this locality, and copied in Annual Report, 1872, p. 335,
fourteen beds of coal were passed to the depth of seven hundred and twenty-

STRATIGRAPHY OF TOE LIGNITIC FORMATIONS. 19

eight feet. This gives sixteen beds of eoal above the great hard sandstone,
which was passed by the drill from seven hundred and eighty to eleven hun-
dred and eighty feet.

From Rock Springs to Evanston, tlie Lignitic is overhiid by the strata
of the Green River group, which is formed of beds of shale, some calcare-
ous, others sandy, with numerous strata of bituminous shale, but as yet no
lignite seams, until, reaching Evanston, we find still heavy deposits of
lignite coal, as recorded in the section (Animal Report, 1872, p. 838), from
top of the hill to base: —

Feet.

Conglomerate 40

Hard yellow, fine-grained, micaceous sandstone 32

Conglomerate, topped with coarse sandstone 37

Fine-grained and intermediate layers of coarse-grained sandstone 32

Conglomerate (lower banks) .' 27

Bituminous clay 10

Shale and clay banks, mostly covered 145

Sandstone in bank 11

Alternating beds of shale and shaly sandstone lOG

Shaly sandstone, very hard, sometimes in bank, with dicotyledonous

plants 11

Argillaceous shale, with ferruginous concretions and remains of plants. 96

Coal 5

Clay and shale 12

Coal * 7

Clay 3

Main bed of coal, with four bands of slate 26

Shale and clay 8

Coal 5

Clay and shale 15

Iron ore 3

Clay and shale 15

I could still mention the Lignitic beds worked at Coalville, not far from
Evanston; those of Sulphur Creek, near Beaver River, said to be seven and
a half feet; those also of Wasatch County, in Utah, which, according to

* The lower part of this section is from Dr. A. C. Peale, ADnual Report, 1871, p. 195.

20 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

Prof. Clayton, are of considerable importance, the prominent vein of the
mcat^ure being thirty-two feet in thickness and of excellent quality, But the
geological relation of these last lignite deposits is not as yet dctinitely ascer-
tained, some of them being apparently Cretaceous. What I have said is
more than sufficient to show the wide extent of the great Lignitic, its average
thickness, and also its capacity and importance for the production of coal.

To ascertain the exact value of the Lignitic coal, numerous chemical
analyses have been made and published. An analysis of coal of the Carbon
Mines is given already in Dr. Hayden's Annual Report, 1869, p. 197. Mr. J. P.
Carson, the chemist, finds in it 51.07 fixed carbon. In Report of 1870, tiie
bituminous coal from the old Placi^re mines of San Lazaro Mountaine, New
Mexico, is analyzed mostly by Mr. Persifer Frazer, jr., and shows on eight dif-
ferent analyses an average of 60 per cent, of fixed carbon; that of Evanston,
4'J. At page 321 of the same volume, there is a very interesting comparative
table of the result of chemical analyses of the coal of the more important
seams of the southern basin, by Mr. James F. Hodge. From it are
derived the following data: — Golden coal, fixed carbon, 45.57 to 47.58;
Murphey's Mine, 44.41 ; Marshall's, 49.72 ; Boulder County, 47.30 to
50.65; Carbon, 49.72; Rock Springs, 54.46; Evan.ston, 50; Coalville, 48.
From analyses recorded in my Report, 1872, the results are about the
same. The Carbon coal has of fixed carbon between 49.30 and 51.65;
Rock Springs, 52.45; from north of Trinidad a coal, of which I had choice
specimens, has of fixed carbon 57.60; Canon City coal, which is very
rich in fixed gas, has 54.70 of fixed carbon, and that of the Raton Mountains
55. Most of the analyses quoted here, with a large numl)er of others, are
presented in a table of the Annual Report of 1873, pp. 112-114. Though,
as everybody knows, the result of analyses present always marked differ-
ences, the compounds of each piece of coal taken from a mine l)eing more
or less varied, it is evident that the coal of the Western Territories is a lig-
nite of high value and of a quality at least equal, if not superior, to the com-
bustible mineral generally known and used in Europe under the appelhition
of lignite. Tlie highest average of fixed carbon in European lignite is 67 to
68, and this very rarely; it generally averages 45 to 46. This subject, how-
ever, need not be considered further. Its relation to the fossil flora is far
indeed, and it is sufficient to touch it in passing in order to omit nothing
which may afford some knowledge of the characters of the Lignitic.

THE AGE OF THE LIGNITIC FORMATION BY ITS FAUNA. 21

^ 3. — Tlie age of the Lignitic as indicated by its geological distribution and

its fauna.

The first explorers of the Great Lignitic seem to have recognized it as
Tertiary; for in their narration, Lewis and CUirkc mention that it overlies the
Cretaceous series. The definition of the age is, however, not positively ascer-
tained by the fossils which they collected from the upper part of the Creta-
ceous clay bed, where the coal scams commence, nor by those procured
later by Nicolet from the same locality, for they were determined by Dr. S.
G. Morton as Cretaceous.* Taylor asserts, however, that from specimens of
plants and animals from the vicinity of Fort Union, near the confluence of the
Yellowstone and Missouri Rivers, they derive incojitestable proofs of a fresh-
water formation. Taylor adds that the Upper Missouri Valley has yet to
receive examination from scientific geologists, and that there can be no
doubt but highly interesting results would follow from investigation in a field
so rich and extensive. On the report of Mr. Harris, the associate of Audu-
bon.t who ascended the Missouri to the mouth of the Yellowstone River, the
committee to whom this paper was referred close their reports with the re-
mark that the proofs thus afforded of a probably widely diffused fresh-water
formation in the region of the Upper Missouri, reposing upon the Cretaceous
strata, and imbedding remains of a manifestly Tertiary age, are just at this
time invested with considerable interest, from their according with the discov-
eries, recently made by Captain Fremont, of the presence of other and probably
extensive fresh-water Tertiary strata in the Oregon Territory.^ Taylor him-
self, considering the brown-coal formations of the Northwestern Territories,
calls them Tertiary.

Dr. Hayden rightly remarks, in the beginning of his report of 1874, that
prior to the time when he began his explorations in 1854, the observations
that had been made by various travelers in regard to the existence of coal
beds in different jiarts of the West were of so indefinite a character that they
cannot be used as evidence, though they may form a part of the early history
of discovery. That the conclusions to which he arrived from the first on the
Tertiary age of the Lignitic are not based upon superficial examination is
sufficiently known by the numerous memoirs published by him from 1857 to

* Proceedings of Academy of Natural Sciences of Philadelphia, October, 1841.
t Proceediiit;8 of tlio Academy of Natural Scieuces of Pbiladclplua, May, 184r).
t Taylor's Statistics, p. 177.

22 UNITED STATES GEOLOGICAL SUEVEY— TERTIARY FLORA.

1861, and afterward by his annual reports. Considering his opinion on the
age of the Lignitic, he remarks, in the last report:* — "It is well known that
I liave held with some tenacity the opinion that the coal formations of the
West are of Tertiary age, and I still regard the Lignitic group as transitional
or Lower Eocene until the evidence to the contrary is much stronger than
any which has been presented up to the present time."

It has been seen already that, in his explorations of 1854, Dr. Hayden
carefully surveyed the Lignitic beds along the Missouri River from their first
appearance near Fort Clarke to the mouth of the Yellowstone, and then up that
river to a point near the mouth of the Big Horn. "In all this distance, about
six hundred miles, following the windings of the river, the Cretaceous beds
appear but once, and then only along the bed of the river for a few miles,
while the entire country, with this exception, is occupied with the Lignitic
groups. It rests everywhere upon well-defined Cretaceous beds, No. 5, which
we have all along regarded as the highest known in the West, and have re-
ceived the name of Fox Hills group, from a locality on the Missouri River
called the Fox Hills or Fox Ridge, where this formation was first studied and
found full of MoUuscan life.f There is a gradual passage upward from the
black, plastic, shaly clays of No. 4, or the Fort Pierre group, to the yellow
calcareous clays of the Fox Hills group, and at the upper portion the sedi-
ments are arranged in thin layers, very arenaceous, indicative of their deposi-
tion in turbulent as well as shallow waters. In these arenaceous sediments,
the well-marked marine life ceases to exist, and soon after appear the brack-
ish-water species.'' From this kind of formation of the Fox Hills group, it is
not surprising that it is not of universal extent. It is the true transition
group, locally of a thickness of five hundred feet, but it is not constant. For
example, its presence is clearly marked from Rock Creek to Medicine Bow
along the Union Pacific Railroad; but I have not seen it anywhere under the
coal strata along the base of the Rocky Mountains, at least not with its
characteristic fossils. At the Raton Mountains, and all around Trinidud,
where the succession of the Lignitic to the Cretaceous is exposed at numy
places, the brackish beds overlying the Cretaceous No. 4 are already Lignitic
by their characters; for they do not contain any traces of Cretaceous remains,
but a profusion of fragments of dicotyledonous wood, evidently rolled with the

• Aunual Report, 1874, p. 20.

t See section of the Cretaceous of Nebraska aud Kansas in Cretaceous Flora, p. 14. It is copied
from Dr. Uayden'u AdduuI Keport, 1870, p. 87.

THE AGE OF THE LIGNITIC FORMATION BY ITS FAUNA. 23

sand by tlic waves. They lliere constitute (lie lower member of those heavy
beds of sandstone, which have been remarked already as forming the base of
the Lignitic, and which have in their remains, and also in their compounds,
the same characters as the lower flaggy or shaly sandstone overlying Creta-
ceous No. 4. No trace of aiiinial Cretaceous rcniains has been found in
connection with them, neither in the south nor in the north Lignitic field.
When Cretaceous No. 5 is not formed, the base of the Lignitic rests on No. 4.

In 1857, an account of the Tertiary of Nebraska was published by
Messrs. Meek and Hayden, and the same year a map of the region bordering
the Missouri River, together with sections and explanatory notes illustrating
the geological structure of the country, was prepared by Dr. Hayden. The
authors had then collected a great quantity of specimens of animal and veg-
etable remains from the base to the summit of the Lignitic group. Sjieaking
of the animal remains which were studied by Dr. Leidy, of the vegetable
fossils by Dr. Newberry and myself, and of the invertebrate by himself and
Dr. Meek, Dr. Hayden says,* ''None of us doubted even their Tertiarxj age";
an assertion forcibly resulting from the determination of the materials col-
lected in the exploration. For, considering merely the MoUusk, we see in llu;
volume of the Proceedings quoted above that of one hundred and fifty species
of Mollusks described from the Fort Union group, fifty-four are of the Ter-
tiary age, fifty strictly fresh-water species, and only four belong to genera
supposed to inhabit salt or brackish waters. The more prominent genera to
which these Mollusks are referred are Ostrea, Uiiio, Pisidium, Corbicula,
Potamomya, Melania, Melampus, Viviparity etc., all of Tertiary types.

The section of the Tertiary measures as recognized by Dr. Hayden finds
its place here as elucidation of the distribution of tlie essential groups of ani-
mal fossils which characterize the formation. The Tertiary divisions indi-
cated by the distribution of fossil plants may present some differences. Tiie
sulijcct has to be examined after the descriptive part of the vegetable remains.
The section is copied from Dr. F. V. Hayden, Annual Report, 1874, p. 23: —

•Proceedings of the Academy of Natural Sciences of Philadelphia, May, 1857.

24

UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

s
a

Subdivisions.

i

Localities.

to

■a

©
>

M
c
§

Fine loose sand, with some layers of
limestone; contains bones of Cam's,
Felis, Castor, Equus, Maetodon, Testu-
do, etc., some of which are scarcely
distinguishable from living species.

*5

o

o

o

On Loup Fork of Platte River, extend-
ing north to Niobrara River and
south to an unknown distance be-
yond the Platte.

6
§
1

p.

9
O

>■
M

White and light drab clays, wi*h some
beds of sandstone and local layers
of limestone. Fossils, Oreodon, Tita-
notherium, Chotropotamus, Ehinoctros,
Anchitherium, Hycenonodon, Machai-
rodtts, Trionyx, Teatudo, Helix, Pla-
norbis, Limnea, petrified wood, etc. ;
all extinct. No brackish-water or
marine remains.

6

o
B

o

■*^

o
o

Bad Lands of White River, under the
Loup River beds on Niobrara, and
across the country to the Platte.

a

(0

§

ii

00

1

s

a

Light gray and ash-colored sandstones,
with more or less argillaceous layers.
Fossils, fragments of Triongx, Tes-
ludo, with large Selix, Vivipara, pet-
rified wood, etc. No marine or brack-
ish-water types.

§

o

o
o

Wind River Valley ; also west of Wind
River Mountains.

w-

Fort TTnion group ; Lignite group.

Beds of clay and sand, with round fer-
ruginous concretions, and numerous
beds, seams, and local deposits of
lignite; great number of dicotyled-
onous leaves, stems, etc. ; Platanus,
Populus, etc., with very large leaves
of true Fan Palms. Also, Helix, Me-
lania, Vivipara, Corhieula, Unio, Oa-
trea, Potamomya, and scales of Lepi-
dolus, with bones of Trionyx, Emys,
Compsemya, Crocodilua, etc.

2
o

a
s

o

§

Occupies the whole country around
Fort Union, extending north into
the British Possessions to unknown
distances; also southward to Fort
Clarke; seen under the White River
group, on North Platte River, above
Fort Laramie; also on west side of
Wind River Mountains.

<6

p

It appears positive that the Fort Union group cannot be separated from
the Colorado Lignitic, or rather that they are both united under the Laramie
group into a continuous formation. Dr. Haydeii has remarl<ed it in the same
report when he says that the Lignitic group descends northward along the
east base of the Laramie Range, and reappears on its other side about ten
miles south of the Union Pacific Railroad. It is therefore continuous.

THE AGE OF THE LIGNITIC FORMATION BY ITS FAUNA. 25

When both areas, the north and south, are considered in regard to their
fossil faunas, they show, however, a marked diflerence, not in the characters
of the species of MoUusks, but in the nearly total absence of invertebrate
fossils south of the Laramie Range. I do not know of any locality where
fossil shells have been seen in the Southern Lignitic Basin except at Mar-
shall's, where a bed of clay iron ore, above No. 5 of the section, has a profu-
sion of fragments of Oyster shells {Ostreo subtrigoimlis?). The same species,
remarks Dr. Hayden, is abundantly found near Brown and O'Bryan's coal
mine, about twenty miles southeast of Cheyenne, also in the Colorado Basin.
Anyhow, no remains of invertebrates identifiable as of Cretaceous age have
been found in the true Lignitic Measures of the Colorado Basin from Chey-
enne to the Raton Mountains. In this whole area, therefore, of the Lignitic,
represented by what is called the Fort Union group in the nortli and the
Colorado Basin in the south, we have, from the distribution of the strata and
from the fossil faunas, evidence only of the Tertiary age of the formation.

There is some more difference in the Lignitic of the so-called Bitter
Creek series, and, as remarked by Prof E. D. Cope in his Report, 1873, p.
438, the authorities on this formation have presented views more or less at
variance with those entertained by him. The whole range from Black Butte
to Point of Rocks is the slope of an anticlinal whose axis is at Salt Wells;
and from the first locality to Point of Rocks or to Salt Wells, in a northeast
direction, the series of rocks is passed, which, in their superposition by the
southwest dip, has a thickness of three to four thousand feet, according to
the measurements of Messrs. Meek and Bannister. A huge Saurian discov-
ered by Prof. Meek in the overlying and burnt shale of the main coal of
Black Butte has been identified by Prof Cope as a Dinosaurian {Aga-
thaumas syloestris) of Cretaceous type. Lower in the series, below Black
Buttes, at Hallville, Prof Meek has found shells whose character is not
quite definite, but which he considers as Cretaceous, though the same locality
was admitted by him in his report of 1870 as Tertiary. But my lamented
friend, who has done so much for the paleontology of North America, has
so clearly discussed the question of the character of the fauna of the Bitter
series formations that I consider it a duty to quote some of the more pertinent
passages of the introduction to his list and description of fossils in the Annual
Report of Dr. F. V. Hayden, 1872. He says (p. 457):—

"Returning to the question of the age of Bitter Creek series, it may be

26 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

slated, ill Uic first i)lacc, ilial Emmons evidently regarded it as Cretaceous, as
may be seen from liis remarks in Jlr. King's report, ])id)!islied in 1870, wliile
Dr. llaydcn favored liie conclusion iliat it is a marine Tertiary group, or a
tnnisilion series between tlie Tertiary and the Cretaceous, in his reports of
thill and the following year."

"Tin; only fossils I had ever seen from this formation previous to visiting
the region during the jiast summer were two species of Ostrea and one of
Anomia from Point of Rocks, and two shells, one or possibly both rtilated to
Corbicula, from llallvillc. Tliose from Point of Rocks I referred to the Creta-
ceous, placing them in the Cretaceous list in Dr. Ilayden's report, 1871.
This 1 did mainly because there were among them no fresh-water or strictly
brackish-water types, while up to this lime we knew of no Tertiary of ex-
clusively marine origin in all this internal region of the continent. I was
also in pari influenced in making this reference by the similarity of one of
the Oysters to a Cretaceous species found in California, wiiile the Anomia
likewise closely resembled a IVxas Cretaceous shell described by Roemer
under the name of Ostrea anomio'formis, which certainly seems not to be a
true Oyster. The two shells from TTallville, however, I referred to the Eo-
cene, not only because they were closely allied to Eocene brackish-water
forms from the Paris Basin (peculiar depressed and elongated form of Cor-
bicula), but because I was not aware at the time that the llallville mines
occur in the same formation as the Point of Rocks beds, nor even within
fifty to seventy-five miles of the same locality.

"On visiting these localities, however, last summer, I was somewhat sur-
prised to find thai tlie llallville mines are only some seven or eight miles from
Point oi' Ivocks, and belong to liu^ same geological formation. A careful
examination, also, soon rendered it evident that all of the rocks for sixteen
hundred to eighteen hundred feet or more above the Hallville coal beds, up
to and including the stratum in which we found the large reptilian remains
at Black Butles, and for even a little greater thickness below the Hallville
horizon, certainly belong to the same group or series of strata, and that fresh-
and brackish- water types of fossils occur along with salt-water forms at all
horizons wherever we found any organic remains thronghout this whole
series."

I'^roni this f:icl. Prof ]\Icek was induced to modify his views, and to con-
sider the whole series Cretaceous, by some reason which 1 do not consider as

THE AGE OF TDE LIGNITIC FORMATION BY ITS FAUNA. 27

conclusive; for it seems that tlic author slioiild rather have admitted Point
of liocks, with Ihe two species of Ostrea, as Tertiary, tlian to liave considered
Hallvillc as Cretaceous, for the species of Molhisks are not more evidently
Cretaceous than are the Oyster shells of Point of Rocks. The Ostrca of Mar-
shall and I'ryant's coal mines are not less marine; species than those of
Point of Rocks. This seems the more surprising, that, considering further
the question of the age of the group, after discovering in the rocks of the
Bitter Creek series between three and four times as many species of fo.ssils
as had been known from the same, the celebrated prores.sor remarks, with that
adminible candor of mind which adorns all his work: —

"Although partly committed in favor of the opinion that this formation
belongs to the Cretaceous, and still provisionally viewing it as most prol)ably
such, I do not wish to disguise or conceal the fact that the evidence favoring
this conclusion (o be derived from the Molhisks alone, as now known, is by
no means strong or convincing. The genera are probaljly ail common both
to the Cretaceous and Tertiary as well as to the present epoch, unless Leptes-
thcs and Vcloritina, which have been separated subgenerically from Corhkula,
may be distinct genera; the European representatives of these being mainly,
if not entirely, Tertiary Ibrm.s, while they do not appear to include living species.
Cj oniohasis is also not known in cither Cretaceous or Tertiary rocks of the
Old World, but then it is an American type, greatly developed among our
existing Molhisca, as well as in the far Western Tertiary Rocks, and we can
scarcely doubt that it will be found in un<jiiestionable Cretaceous beds there,
even if some of the imperfect si)ecimens already known from the same are not
sucli. It should be remembered, however, that even the specimens I have
referred to this genus from Bitter Creek beds are not in a condition to show
the aperture beyond douljt to possess the characters of Ooniohasis.

''The entire absence among the fossils yet known from this formation of
BacuUtis, Scnphitcs, Ancyloceras^ Ptychoceras, Amjnoniles, Gyrodes, Anchurn,
Inoccramus, and all the other long list of genera characteristic of the Creta-
ceous, or in part also extending into older rocks, certainly leaves its Molluscan
fauna with a strong Tertiary facies. Nor can we quite satisfactorily explain
this away, on the ground thai the water in which this series of rocks was
deposited partook too much of the character of that of an estuary, to have
permitted the existence of any of these marine genera, because we do find
in it the genera Ostrea, Anomia, and Modiold, which prol)al)ly required water

28 UNITED STATES GEOLOGICAL SUEVEY— TERTIAKY FLORA.

salt enough to have permitted the existence of Inoceramvs, Anchura, and Gy-
rodes, it" not of some or all of the genera mentioned above. Indeed, at Coal-
ville we find Inoceramus associated with some brackish-water types, and the
additional Cretaceous genera Cypritnera, Anchura, Gyrodes, etc., in closely
associated beds."

"When we come to consider the invertebrate fossils yet known from this
formation in their specific relations, we find all, with possibly two or three
exceptions, new to science, and different from those yet found either at Bear
River, Coalville, or indeed elsewhere in any established horizon ; so that we
can scarcely more than conjecture from their specific affinities to known forms
as to the probable age of the rocks in which we find them. Considered in
this respect, their evidence, however, is conflicting. Two of the species of Cor-
bula for instance (C tropidophora and C. undiferd) are most similar to species
found in the brackish-water beds at the mouth of Judith River in the Upper
Missouri, that we liave always considered Lower Tertiary, though there are
some reasons for suspecting that they may be Upper Cretaceous. A Corhicula'
both from the Black Buttes and Point of Rocks localities is even so very
nearly like C. cylheriformis from the Judith River beds that I have referred it
doubtfully to that species."

"Again, the species Anomia gryphorhynchus, found so abundantly at
Point of Rocks in the same bed with the above-mentioned Corhicula and
Corbula tropuiophora, so closely resembles a Texas Cretaceous shell described
by Roemer under the name Ostrea anomiceformis that I am strongly inclined
to suspect they may be the same; though whether identical or not, at least
our shell is certainly not an Oyster, as it has its muscular and cartilage scars
precisely as in Anomia, while its beak is never marginal, and it has no liga-
ment-area. In all of these (and indeed in all other characters), the Texas
shell as illustrated by Roemer seems to agree precisely with ours, excepting
that he represents it as having only one central muscular scar instead of three.
In many of our specimens, however, the two smaller of these scars are very
obscure, and might be easily overlooked. It is true he figures a nearly flat
valve, without any byssal perforation, and a convex one as opposite valves, and
if they are such the shell would certainly not be an Anomia. Among a large
collection of our shells, including thousands of specimens, however, I have
not yet seen a single perforated valve, though they vary much in convexity,
some of the valves being nearly as depressed as the one Roemer figures as

THE AGE OF THE LIGNITIC FORMATION BY ITS FAUNA. 29

the upper valve, supposing it to be an Oyster. If these depressed specimens
in our collection are opposite valves to the convex ones, then the shell would
neither he an Ostrea nor an Anomia, but would almost certaiidy fall into
Morris and Lycett's genus Placunopsis, which, so far as known in Europe, is
a Jurassic group. Consequently, if our shell sliould fall into that genus, it
would, when viewed in connection with its associates and all tlie other known
facts, furnish a strong argument in favor of the formation being at least as old
as the Cretaceous. There are good reasons, however, for believing that these
depressed specimens, as well as the convex ones, are all upper valves of the
same shell, only modified in convexity by accidental circumstances of station,
as their slight obliquity, as seen, for instance, in a look at the interior of both,
is found to be in the .same direction instead of the reverse, as would be the
case if they were opposite valves of the same shell; while among thousands
of specimens no example of a depressed and a convex valve united has been
seen, nor have any been found that would come near fitting together.''

"On the other hand, the Corbiculas are decidedly Tertiary in their spe-
cific affinities, as well as in their subgeneric; C. fracta, for instance, and
C. crassa tell [for mis, from the Hallville mines, being very closely allied to
Paris Basin Tertiary forms, the first-mentioned species being the type of a
subgenus, .so far as known, peculiar to the Tertiary elsewhere. The same
may also be said of C. cytheriforinis, which also seems to belong to a group
{Velori./ina) peculiar to the Tertiary in Europe."

"But the most surprising fact to me, supposing this to be a Cretaceous
formation, is that we found directly associated with the reptilian remains at
Black Buttes a shell I cannot distinguish from Viviparus troch/formis, origin-
ally described from the Lignitic formation at Fort Clarke, on the Upper Mis-
souri, a formation that has always been regarded as Tertiary by all who have
studied its fossils, both animal and vegetable. The specimen mentioned
does not show the aperture nor all the body-volutions, but, as far as can be
seen, it agrees so exactly witri that very peculiar species in size, the form
and j)roportion of its volutions, the slopes of its spires, its surface-markings,
the nature of its suture, and, in fact, in every respect, so far as can be seen,
that I have scarcely any doubt of its identity with the same."

In resuming liis remarks, Prof Meek further states, p. 461, § 7 to 9: —

"That on the one hand, two or three of its species belong to sections or
subgenera {Leptesthes and Velorifina) apparently characteristic of t lie Eocene

30 UJJITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

Tertiary of Europe, and are even very closely allied to species of that age
found in the Paris Basin; while, on the other hand, one species seems to be
conspecific with, and two congeneric with (and closely related specifically to),
forms found in brackish-water beds on the Upper Missouri, containing ver-
tebrate remains most nearly allied to types hitherto deemed characteristic of
the Cretaceous.

"That one species of Anomia found in it is very similar to a Texas Creta-
ceous shell, and perhaps specifically identical with it; while a Viviparus,
found in one of the upper beds, is almost certaiidy identical with the
V. trochiformis of the fresh-water Lignite formation of the Upper Missouri,
a formation that has always, and by all authorities, been considered Tertiary.

" That the only vertebrate remains yet found in it are those of a large
reptilian (occurring in direct association with the Viviparus mentioned above},
which, according to Prof Cope, is a decidedly Cretaceous type, being, as he
states, a huge Dinosaurian."

"It thus becomes manifest that the paleontological evidence bearing on
the question of the age of this formation, so far as yet known, is of a very
conflicting nature, though, aside from the Dinosaurian, the organic remains
favor the conclusion that it is Tertiary."

Prof Cope has surveyed with the greatest care the whole series under
consideration, and found, from the lowest marine coal-bearing formation
to the Saurian bed of Black Buttes, an uninterrupted Cretaceous fauna, as
indicated by remains of vertebrate animals. He therefore, after considering
the facts exposed in favor of both opinions on the Cretaceous or the Tertiarj'
age of this Lignite of the Bitter Creek series, comes to the conclusion that
there is no alternative but to accept the results, that a Tertiary flora was
contemporaneous with a Cretaceous fauna, establishing an uninterrupted
succession of life across what is generally regarded as one of the greatest
breaks in the geological times.*

As no kind of Cretaceous animal remains have been discovered in the
Lignite of Colorado, none cither in that of the north, generally called the Fort
Union group, the question of age essentially bears upon that Bitter Creek
series. The line of demarkation between the Cretaceous and the Tertiary
is placed by Prof Cope above the Black Buttes Saurian bed. Prof King
fixed it in the middle of the series, near Point of Rocks; and, from my

• Annual Report, 1873, p. 44-2.

THE AGE OF TOE LIGNITIO FORMATION BY ITS FAUNA. 31

own observations when I explored the range, I found it near Salt Wells,
or below Point of Rocks. If, then, we find the fossil flora of this last
locality agreeing in characters with that of Black Butte, if this one bears the
same degree of relation to that of the Colorado Basin, and to that of the
North Lignitic or of the Fort Union group, we shall have to conclude that
these land formations of the great Lignilic are contemporaneous, and the
question of their age has to be decided, I think, hy the comparison of the
vegetable types represented in the whole Lignitic formation. This can be
done only after the description of the species and the exposition of their
characters.

PART II.

DESCRIPTION OF THE TERTIARY FOSSIL PLANTS.

CRYPTOGAMiE.
FUNGI.

We cannot, doubt the existence of vegetables of this order in former
geological epochs, when there was a profusion of woody plants whereupon
they (^ould thrive as parasites, as they do now upon organs of the same kind.
It is, however, clear that all the Fungi of soft, fleshy substance, like those which
we see in the spring and the fall, especially upon the wet ground, and which
are soon decomposed after their apparition, cannot have left any trace of their
existence in a petrified state. Their former life is revealed, however, by the
fossil remains of insects deriving their food from fleshy mushrooms only.
The Cryptogamous kinds which vegetate upon the bark and the leaves of
trees, and which are sometimes persistent, even when the wood or the leaves
are passing into a state of decomposition, are the only ones which may be
preserved by fossilization, and which we may expect to recognize in a few
instances. The characters of the Cryptogamous plants, however, are mostly
established by organs of fructification which are unperceivable to the eyes,
and, being in most cases enclosed into the substance of the plants, they can-
not be discovered in a fossil state ; therefore the determination of all the
Cryptogamous, even that of the AlgcB, is very unreliable. Spots of different
colors, small papillae, also similar to those which are imprinted or engraved
by living fungi upon stems and leaves of the present flora, are often remarked
upon petrified substances of the same kind, even in fossil remains of the Car-
boniferous. But they may be mere imitations produced by the deposit of
particles of stony matter or by the impregnation of foreign substances, espe-
cially of iron ; and, as their determination is therefore still more uncertain

33
3 T F

34 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

and of little importance, I have generally omitted describing them. The few
species of Cryptogamoiis described and figured in this memoir are the only
ones which seemed distinctly referable to plants, and they are published in
order, especially, to show the existence of this class of vegetables in the Lower
Tertiary measures of this continent.

SPHERL&, HaUer.

Spheria lapidca, Lesqz.

Plate I, Fig. 3 .

Spheria lapidea, Lesqx., AnDnal Report, 1872, p. 373."

Perithecia (receptacles) ronnd, highly convex, one to two millimetres broad, growing in lineal
series from nuder the bark, and pierciug it before opening; borders irregularly lacerated ; substance
membranaceous or coriaceous, wliitish.

This species is of subcortical growth, upon a petrified fragment of wood,
which has part of the bark preserved. The few perithecia under the bark
of the specimen in a are like small warts whose surface is rough and opaque.
Just at the borders of the bark, in h, the receptacles have perforated it in
irregular circles with lacerated borders; and still lower, in c, the intumescence
of the bark produced by the growth of the perithecium is seen prominent
and smooth, the epidermis being still entire, or not yet pierced through by
the plant.

Habitat. — Upon a fragment of wood, Raton Mountains, New Mexico.

Spheria m y r i c ce , Lesqz.

Plate I, Fig. 4.
Sphtria wii/riccB, Lesqx., Annual Report, 1872, p. 390.

Perithecia punctiform, minute, either 8pars&-or in circle, forming round spots.

The receptacles of this species are punctiform, very small, sometimes
irregularly scattered, more generally disposed in circles, forming rings one
millimeter in diameter, the centre of which is clear and of a light color. It
resembles Xylomites varius, Heer (Flor. Tert. Helv., plate i, fig. 9), but in
our plant the punctate form of the perithecia is distinctly recognized with
the glass.

Habitat. — Upon leaves of Myrica Torreyi found at Black Butte, Wyo-
ming, and of M. nigricans, from specimens of Green River Station {Dr. F.
V. Hay den).

' The numerous quotations of the Annual Reports of Dr. F. V. Ilayden's Geological Survey of the
Territories are hereafter indicated as above. The dates indicate the years of publication of the Reports.

DESCRIPTION OF SPECIES— CKYPTOGAMIA—LICIIENES. 35

S p li c ■- i :e r Ii y t i m 111 o I d c s , Lesqx.

Plato I, Fi^'s. 5, T) a.

Spheria rlii/linmaUJe^, Lescjx., Annual Keport, 1874, p. ;!08.

Purithocia ]mnctiforin, placed in a simple circular row, larger than in tlic former spocifs, with
borders irregular.

This species composes, like tlie former, circular spots, by the disposition
of its receptacles in a circle. These, however, are larger, five to six only in
a simple row, either separated or more generally connected. The spots vary
from one to two millimeters in diameter.

Habitat. — Upon the stem ot" Caulinites sparganioides at Black Butte.*

S c I e r o t i II III r II b e 1 1 II m , Lesqx.

Plate I, Figs. 2-2/.

Scleroiiuvi rubellmn, Lesqx., Annual Report, 1872, p. 375.

Perithecia oval or oUloug, obtuse, convex when young, then concave or channelled ; spores of a
red color.

The different phases of development of this fungus are seen in figs. 2
a to 2/. Before its maturity, the surface is convex, by an inflation of the
receptacle, as in fig. 2 o. I consider this as the first stage of its growth.
Later, the central part is depressed, and the borders appear somewhat elevated
around it. It then becomes either concave (fig. 2 b), or channeled (fig. 2 c),
or flat (fig. 2 d). When the perithecia are ripe, they appear opened, and the
inside is covered with a punctate red surface resembling a pulverulent matter
like spores. I was, however, unable to detach any of it, or to trace any form
of organism in this colored central part. The tbrm of these perithecia is
very variable, sometimes small and nearly round, more generally linear-
oblong, obtuse, two to four millimeters long, rarely broader than one milli-
meter, generally intermediate to the veins of Flabellaria, and like them
buried under the epidermis.

Habitat. — Upon fragments of leaves of Flahellaria Zinkeni?, Golden,
Colorado.

Sderotium pustuliferum?, Heer, mentioned (Annual Report, 1871, p.
yOO) as growing upon the leaves of Crjperus, is too indistinct for description,
and cannot be figured.

LICHENES.

The great scarcity of fossil remains of Lichens is perhaps more remark-
able than that of the Fungi. Professor Goppert, in his study of the vegeta-
ble fragments preserved in succin, or amber, has demonstrated the existence

* All the species without name of the discoverer are described from specimens fouud by the author.

36 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

of plants of this order in former geological times, and has been able to
refer most of them to species allied or even identical with those of our
epoch. The Lichens mostly inhabit the bark of trees or closely, adhere to the
surfaces of stones or rocks. Though hard they may appear, they are com-
posed of cellular tissue, easily destroyed by prolonged immersion, and it is
therefore extremely rare to find any species of this genus upon the bark of old
trees when their decomposition is advanced. They are also generally attached
to the epidermis, which is rarely preserved with the bark of decaying wood.
Though it may be„ excepting the species described from the succin, very few
other Lichens have been recognized as yet in a fossil state. Goppert remarks
the presence of a Verrucaria and two Graphides in the Lignitic of Germany as
an extremely rare occurrence; and Schimper says that he has been able to
find once only a fossil Lichen represented by a few specimens of Lecidea. It
was only after a careful study of the characters of the remains described
below that they were recognized as positively referable to Lichens.

OPEGRAPHA, Ach., Nyl.
Opcgraplia antiqua, Lesqz.
Plate I, Figs. 1-1 c.
Opegrapha antiqua, Lesqx., Annual Report, 1872, p. 390.

Peritbecia one to four millimeters long ; linear or sligLtly enlarged in the middle ; pointed or
obtuse ; more generally united two, three, or lour in oppobito directions ; sometimes flexuous or curved.

The form of the small plants, or of their peritbecia, is distinctly repre-
sented upon the specimen, a piece of clay wherein the nuclei have either
been left imbedded or have deej)ly stamped their outlines. They are com-
paratively small, scarcely half a millimeter broad, either in a line or curved
in various ways ; sometimes short, oblong, enlarged in the middle, joined two,
three, or more by the points, and diverging star-like ; sometimes single, linear,
falcate or hooked, and obtuse at both ends. From their impressions, they
appear to have been upraised above the surface to which they were attached,
slightly grooved in the middle, with round borders. The inside part of the
cavities has generally in the middle, vvhen the peritbecia or nuclei are out, a
streak of black carbonaceous-like matter, as seen on the right of the
branch (enlarged part, fig. 1 a). The print of the nuclei is marked, figs.
1 b, 1 c, still more enlarged. As resulting from an impression, the inside
part should be convex. The borders of these small plants seem to have been
detached, or arc too close to have allowed the penetration of clay between
them. The relation of this species is especially with Opegrapha astraa.

DESCRIPTION OP SrEClBS— CEYPTOGAMIA-ALGiE. 37

Tuck., of Texas. It is also more distantly comparable to Graphis elegans,
Nyl., of Oahu.

Habitat. — Black Butte ; upon a piece of clay shale, the counterpart or
impression of a stem of Caulinitcs sparganioides.

As for the Lichens, the exact determination of the fossil remains of
marine plants generally named Fucoids is not possible. Their forms, however,
are sometimes definable, and persistent also; and thus, if all their characters
are not positively determinable, and if even their generic references are mostly
uncertain, they may be at least compared, separated in groups under definite
names, and used for the identification of geological formations. Still for
this, they are not as reliable as land plants. Their types appear to be
preserved for long space of time, on account of the slow modifications of the
element wherein they live. Hence we find some species present in two
formations whose age is indicated as difierent by the remains of land plants.
The study of the fossil Algce is also rendered diflicult by their distribution
ill series of rocks, especially of sandstone, of soni<; thickness. Growing up
as last as the sand is heaped around and covers them, they send their l)ranches
in every direction, sometimes filling the rocks by their multiple subdivisions
in such a way that even large specimens do not represent the characters of
their general outlines. The most of the AlgcB are moreover of a soft sub-
stance, easily destroyed by decomposition, and then flattened and mixed
together in an amorphous mass. Their former existence is often recog-
nized only by flakes of carbonaceous matter, or even by the mere discoloration
of the rocks. Heavy beds of sandstone in New Mexico and Colorado,
especially, contain a profusion of fossil remains of Alga, and if it had been
possible either to cut them from the rocks or to study and to figure them in
place, a large number of very diversified forms could have been represented

and described here.

HALYMENITES, Sternb.

IIhI y iiicnitcs striatiis, Lesqz.

Plate I, Fig. G.

Hahjmeniteg striaius, Lesijx., Annn.al Report, 1872, p. 373.

Frond large, dichotomous, erect; branches short, obtuse, cylindrical, or more or less flattened by
compression ; snrface irregularly striate.

This species is related to the following by its ramification, which is
extremely variable. Sometinies the branches on a more or less obtuse angle

38 UNITED STATES GEOLOGICAL SDKVEY— TERTIARY FLORA.

of divergence come out from a tumescence of the principal axis, as repre-
sented partly in fig. 6 ; sometimes tliey are comparatively small and in
right angle, penetrating the sandstone in every direction and in such a way
that, as the remains are extremely abundant, it is very difficult to separate
them even in small fragments. The surface is more or less evidently striate ;
the young branches are nearly smooth, the larger stems deeply, somewhat
irregularly wrinkled or rather veined, as the striae appear here and there
parallel and distinct to the eye, but not permanent upon wide surfaces.

Habitat. — Sandstone underlying the coal beds at the base of the Raton
Mountains, New Mexico.

Halymenites major, Lesqz.

Plate I, Figs. 7, 8.

Halymenites major, Lesqx., Annual Report, 1872, pp. 373, 3U0.

Frond of the same size and mode of division as in the former species; snrface marked by round
contignous or separate tubercles.

The frond of this species appears to h.ave been still larger than that
of the former, cylindrical, rarely and slightly flattened ; pinnately, or dichoto-
mously, or abnormally divided in shorter, smaller branches, very variable in
length ; surface covered with round, elevated, or half-globular tubercles, two
to five millimeters broad, and as thick as broad. As in the former species,
the divisions from the main axis often proceed from an intumescence, or knot,
where from two to three, or even four branches come out diverging all around.
In other cases, the branches are merely alternately forking or dichotomous,
and regular in their divisions. The tubercles of the surface are either distant
or separate, as in fig. 8, or crowded and connected on the borders, as in fig. 7 ;
more or less irregular and variable in size, even upon the same fragment, but
always present. Even the specimens which appear to have been exposed for
a long time to atmospheric influence, or to abrasions by water, show distinctly
the scars of the impression of the tubercles by deep circular lines upon the
stony compound.

Habitat. — This species, extremely common in the sandstone of (he
Lignitic measures, is found at various stages ; at Black Butte, Golden, Carbon,
and especially at the Raton Mountains, with the former. Prof Meek found
it in the sandstone of Coalville, which he considers as of Cretaceous age.
It has been sent nearly with every lot of .specimens collected by Prof Hayden
and his assistants. Its highest station is at Carbon.

DESCIilPTlON OF SPECIES— C1{YPT<J«AMIA—ALG^. 39

II a I y 111 e n 1 1 c s m i ii o r 7 , F. O.

Plate I, Fig. 9.
Halymmites minor, Fiscb., Oost. D. Fobs. Fuc, pp. 56, 05, pi. xiii, xvi.— Lesqx., Annual Report, 1872, p. 37;t.
Branches small, half a centimeter hroad ; tuborcules small and flat.

The European species is known to me from description only, and IVoni
u small fragment figured in Urweli der Schiveitz by Ileer. I should liave
considered the fragment figured here as a branch of the former species, if the
tubercles were not much smaller and all flattened, apparently in a natural
way, and not by erosion. It is, however, admissible that, variable in its size,
its ramification, and other characters, as is the former species, it may be
represented also by fragments like the one doubtfully referred to H. minor,
a species which, according to Schimper, is already uncertain, and established
from insufficient materials.

Marine plants of the same type as these have been remarked in Europe
in the Jurassic and the Eocene formations. Count Saporta has represented,
in PI. Jurassiques, 1st suppL, pi. Ixviii, fig. 3, under the name of Fhymato-
(Irma Ccelatum, Sap., a species of Algce, similar to this H. minor, and Watelet
has, in liis Plantes Tertiaires du Basdn de Paris, an Eocene formation, a spe-
cies remarkably like H. major.

Habitat. — Sandstone beds of the Lower Lignitic; Raton Mount-
ains, etc.

DELESSEEIA, Lamx.

Dclesseria f ulva, Lesqz.

Plate I, Fig. 10.
Delesseria fulva, Lesqx., Annual Report, 1872, p. 376.

Frond memhranacoous, dichotomous, long, linear, with an irregularly inflated middle nerve;
divisions linear, distant, alternate, obtuse, enlarged and lobed at the point.

This fucoidal frond, or probably a mere branch of a frond, is remarkably
fine, its brown-reddish color contrasting with the white sandstone wherein it
is imbedded. The specimen was originally larger than the figured i)art, but
had to be cut from a rock in place, and, the stone being hard as flint, it was
impossible to break it out in its full shape. The part left out, however, did
represent only one more of the lower divisions, of which some traces are left,
indicating its length and its increasing width toward the top. The main
branch, measuring twenty centimeters long, before breaking it, is a little
more than one centimeter broad, dichotomous, the lower branches more
distant and longer than the upper ones, nearly all equally dilated toward the

40 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

top, where they are cut in two or three irregular, short, pointed or obtuse
lobes. The main stem and the divisions have in the middle an inflated large
nerve, more or less distinct, but traceable in the whole length and forking
with the branches, at a distance below the sinuses ; the substance of the
plant appears membranaceous or subcoriaceous, somewhat thick, as seen
upon the borders, which, here and there, are slightly raised above the stone in
their undulations.

The nearest relation of this species among the fossil plants is Delesse-
rites sphferococcoides, Ett., Eoc. Fl. des Monte Prom., p. 8, pi. i, fig. 1. The
affinity is still more distinctly marked with some species of the present marine
flora, D. alata, D. sinuosa, etc., but especially with the fine Botryoglossuin
platicarpum, Kutz, of California. Indeed, the coriaceous or cartilaginous sub-
stance of this species rather refers it to this last genus than to the delicate
membranaceous fronds of Delesseria.

Of the eight fossil species of this last genus described by European
authors, seven belong to the Eocene, or the Lower Tertiary, four being from
Mount Bolca and from Radoboy, and two from Mount Promina. The other
species described by Sternberg as an Haliserites, H. Reichii, and considered
by Schimper as a Delesseria, is from the Upper Cretaceous of Niedershoena in
Saxony.

Habitat. — On upraised white sandstone rocks under or between the coal-
beds of Golden, Colorado.

CAULERPITES, Schp.

Canlerpites incrassatus, Lesqz.

Plate I, Figs. 11, 12.

Delesseria incrassata, Leeqx., Annual Report, 1872, p. 374.

Frond siraple (?); lamiute either trifid from the base or simple, sessile, close, obovate, narrowed
downward ; surface rugose.

The specimen seems to represent the upper part of a frond by three
divisions, or laminae, apparently joined at their base to a common axis. These
segments, deeply impressed into the stone, are either rounded or slightly
cuspidate at the top, gradually narrowed toward the base, as seen in fig. 12.
From irregular, half-round scars, represented upon an indistinct continuation
of the axis, fig. 11, it seems as if the laminae had been sessile, half embracing ;
but the specimen does not show precisely any trace of divisions between the
segments, and it is therefore uncertain if they were separated or connected at
the base, opening altogether at the upper point of the axis as verticillatc, or

DESOEIPTION OF SPECIES— CRYPTOGAMIA—ALG^. 41

successively unfolding upon another as alternate. The substance of tlie
plants seems, from the deep impressions, to have been very thick and carti-
laginous or hard, the surface is rugose in the length, or rather deeply wrinkled,
as if the iaminai had been compressed on both sides; and the borders, irregu-
larly undulately lacerated, are either round or slightly cuspidate or even
emarginate at the top.

Though tlie relation of this species to the genus Caulerpa is not distinctly
marked, its affinity with Delesseria is still less positive. To this last {{cnus
it is related only by the outlines of its flat, thick segments, while it is refer-
able to Caulerpa by the mode of attachment of its divisions, comparable to that
of Caulerpa ericifolia, Ag., and by their form and the thick cartilaginous sub-
stance, of the same character as in C. proUfera, Lam., two species of Florida.

Some fragments of another species, described in Annual Report, 1872,
p. 374, under the name of Delesseria Ungulata, have not been figured, being
too incomplete and of a too uncertain relation. They appear sparsely and
separately strewn upon the sandstone, like detached fragments of some com-
pound fronds, or like the sacs of some Ulvacece, flattened by compression. The
more complete of these, two centimeters long, twelve millimeters broad, is
rounded at one end, broken at the other, slightly contracted in the middle,
and marked in its length by a costa. But for this last character I should
have considered these fragments as referable to the former species.

Habitat. — Raton Mountains, New Mexico, in sandstone ; both forms.

CHONDKITES, Schp.

Chondrites snbsimplcx, Lesqz.

Plate I, Fig. 13.

Chondrites subsimplex, Lesqx., Annnal Report, 1872, p. 373.

Froud cylindrical, more or less flattened by compression, with rare, dichotomons, long, flexuoiis
branches, mostly of the same size in their whole length.

This species is found generally flattened, with its expanded, long, flexuous
branches, covering large slabs or passing across layers of shaly sandstone.
Sometimes the filaments, or fronds, appear simple, linear, like those o{ Haly-
menites lumbricoides^ Heer, Urw., p. 244, pi. x, fig. 11, only longer, thicker, and
broader, not granulose; rarely they branch by a simple forking of the main
axis, preserving the same size; some of the divisions, however, gradually
decrease in size to a round point half as broad as the main stem. The sur-
face is irregularly I'oughened, especially slightly wrinkled across along the

42 UMTED STATES GEOLOGICAL SURVEY— TEKTIAKY FL011.V.

borders, and the middle is generally marked by a depression which seems to
indicate a fistulous character of the stem and its divisions.

Habitat. — Biise of the Lignitic tbrmations, Raton Mountains, New
Mexico.

Choiidrite<i bnlbosns, Lesqs.

Plate I, Fig. 14.

ChoHdritff buTbosMf. Lejsqs.. Annn.-U Report, l?T"i, p. 273.

Fiond flattened, irregularly, subpinnately divided iu opposite or alt<>ruate braucbes, dose to each
other, or distant, short, inflated, some of them like irregular tubercles.

Mere fragments of the branches of this species could be obtained for
illustration and description. The mode of division is very irregular; geuer-
iillv the main axis is slightlv larjier tbau the primarv branches, three to five
millimeters broad and irregularly dichotomous. The divisions, however, are
directed either in riijht angle or upward and downward in the same fragment,
and sometimes linear, sometimes inflated, taking the most diversified and
irregular appearances, or inordinately inflated and narrowed. This species
finds analogous representatives in Hdlymenites varius and other Algoids,
described by Sternberg under the same generic name, from the Jurassic tbr-
mations; the Tertiary species difteriug merely by its smooth surface and the
more distinctly inflated branches.

Habitat. — Raton Mountains, near Trinidad. New Mexico; base of the

Lignitic formations.

FITCTJS. Linn.

Fncns lignitn m. Lesqs.

Plate LXI, Figs. 24 and 24 a.

FttCKS Ugniiam, Le^s., Annual Beport, 1S74, p. 296.

Frond flattene«l. irregularly dichotomous; branches diverging obliquely ; branchlets short, terminal,
linear, divaricate, tufted, forking at the point.

The fragment figured here is the only part of the plant represented
upon the specimens. The lowest branches are four millimeters broad at
tlie base, but the size of the branchlets diminishing nearly one-half at each
dichotomous division, the terminal ones are very slender, scarcely half a milli-
meter broad. The upper divisions are like the first ones, short, split or forking
at the point and divaricate. The substance is membranaceous and yellowish,
We have here evidently mere fragments of apparently large compact fronds
like those of the living Fucus auialiculatus, Agh., common along the coasts of
the Baltic Sea, and which appears closely related, by its membranaceous con-_^
sistence, its mode of division, and the form of its branchlets, to the species

DESCPJPTION OF SPECIES— CnAIiACE^—MUSCl. 43

described here. This Fucus has been found in a fossil state in the Tertiary
formations of Spitzbergen, as recognized by Heer. Another species, Sphcero-
coccites crvipiformis, (Sternb.) Schloth., Petref, i, p. 35, pL iv, fig. 1, from
the Lignitic Tertiary of Bohemia, is also related to the American species.
Habitat. — Point of Rocks, Wyoming Territory (Dr. F. V. Hayden).

CHAPvACEJS.

We have to pass over i\\\s- order of plants without mentioning, as refer-
able to it, any fossil species from this continent. It is indeed peculiar that
remains of tliis kind have not been discovered as yet in our Tertiary measures,
for the plants of this genus are quite as frequent in the shallow lakes, the
ponds, and the calcareous springs of North America as they are in Europe.
The fossil Characece. are mostly known by their seeds, which, though small, could
be easily discovered in the soft black shale or clay of the Lignitic measures,
where they should be carefully searched for by the collectors of paleonto-
logical remains. The seeds vary in size from one-half to two millimeters,
and are easily recognized, though small they may be, by their round-oval
form, and their surface, upon which the valves are generally marked by dis-
tinct spiral lines. Of the thirty-seven species described by European paleon-
tologists, eight only are known by their stems and branches; all the others
are described from the size of the seeds and the disposition of their spiral
lines. Of the whole number, one species only is known from the Wealden
or Upper Jurassic; none from the Cretaceous. The other species have been
recognized from the different stages of the Tertiary.

MUSCI.

One species of Moss only has been discovered in a fossil state in the
Lower Lignitic of the West, while more than twenty-five have been described
by European paleontologists, all from the Tertiary measures. At the present
epoch, most of the Mosses, growing in the water, upon the ground, the rocks,
and the trees, are easily decomposed and destroyed, naturally destined as they
are to the formation of humus, and, when upon trees, to the absorption
of humidity, either as protection to the roots or as a more active agent of
decomposition of the wood. Most of the hard, woody species of mosses live
upon the peat-bogs and enter into their composition for the production of the
combustible matter. Their discovery in this case is out of question, for

44 UNITED STATES GEOLOGICAL SURVEY— TERTIARY I'LtJEA.

wo do not find in the lignite coal any identifiable remains of the plants. It is
only when these hard species of Mosses are casually deposited in imid or
clay, or buried into sand-banks, that their forms are preserved for an indefi-
nite length of time. The species described below appear to have been
imbedded in that way along the muddy borders of a shallow lake. Of the
European fossil species, all those referable to the division of the Acrocarpi
have been found in the Succin by Gocppert; the others, all Plnurocarjd,
especially referable to the genus Hypnwn, have been discovered in layers of
clay or in beds of sandy shale.

HYPNTJM, Linn.

Hyp II 11 III II ay den 11, Lesqz

Plate V, Figs. 14-14 b.

Hypnum, Haydenii, Lesqx., Annual Report, 1874, p. 309.

Stem rigid, sparingly divided in nearly opposite, snbalternate, short branches, slightly inflated
toward the top, or club-shaped ; leaves closely imbricated all around the stem and branches, ovate-
lanceolate, acuminate, concaye.

The specimen is figured as far as it is discernible. The fragment
resembles a branch of a coarse species of Hijpnum, like H. rugosum, H. Boscii,
especially, a spe^es which has its largest branches divided as in this fossil
Moss, and of equal size. The mode of division of this plant separates it from
the Lycopods, while the apparently thick leaves seem abnormal for a species
of Moss. It is well to remark, however, that the matrix wherein the fragment
is preserved is a hardened plastic clay, of very fine te.xture, where even deli-
cate small feathers, wings of insects, etc., are distinctly recognized, and that,
therefore, the form of the leaves of a hard species of Moss, even their convexity,
may have been easily impressed upon that kind of soft sulistance. No trace
of middle vein is visible, of course, for the species of Hypnum have rarely
the nerve prominent enough upon the back of the leaves to leave an impression
by compression and fossilization. The point is very acute and apparently
piliferous, but this last character is not positively ascertained.

Habitat. — South Park, Colorado, near Castello's Ranch {Dr. F. V.
Hay den).

LYCOPODIACEiE.

Species of this family represent an inijxirtaiit j)art of the vegetation of the
old coal-measures, not only by large trees, Lep'ulodendron, Ulodendron, and other
allied genera, but by tru(; species oi' Se/aginella and Lycopodium, known by tlieir

,i.^-<r^ , cuMi

\j>^. %Miry> Tq^^. . '"^-^

DESCiaPTION OF SPECIES— LYCOPODIACEiE. 45

branches, leaves, and fructifications. Three species of tlie first group and as
many of the second have been publislied by Goldcnberg from the UppcT
Carboniferous flora of Saarbruck in Alsatia, besides others, less distinct, known
as Li/cojwdites, also from the Carboniferous, even the Devonian. From the
Coal epoch to the present time, however, this family was until now repre-
sented by a single species, Lycopodites fu/catus, L. & 11., apparently a
Selagimlla, from the Oolite of England. Even among the thousands of
Tertiary species described by European authors, no plant of this kind is
mentioned. This is a remarkable fact, as the Lycopods especially thrive at
our time under the shade of Conifers, and as the Permian, the Jurassic,
especially, have in their flora, as far as it is known, a large preponderance
of species of this kind. The discovery of two or three species of Schginella
and a Lycopodium in our Lignitic flora is, therefore, truly remarkable;
and it is especially fortunate that Selaginella at least is represented by a
number of so well preserved specimens that the characters cannot be mis-
taken. The genus is mostly tropical or equatorial ; its species, especially
numerous in Brazil, cover the ground under the shade of Palm trees. An
association of this kind is remarked at Golden, where the remains of Palm
trunks and leaves are found in profusion. #

LYCOPODIUM, Linn., Spring.
Liycopodiiiin pro milieus, Lesqx.
Plate V, Figs. 13-13 h.
Lycopodium prominens, Lesqx., Annual Report, 1873, p. 409.

Stems or branches dichotomous ; divisions sliort, slender, erect or half open, distant; leaves alter-
nate or disposed in spiral order, cylindrical, inflated, and obtnsely pointed, apparently connate at the
narrowed base, half open, like the branches, slightly curved backward.

The fragment of frond representing this species is half imbedded into
the stone, or covered by a thin concretionary compound, the branches and
leaves being in relief upon the surface. Tlie slender divisions, of the same
size as the main axis, vary in length and in degree of divergence. The
leaves, three to four millimeters long, half a millimeter wide, are loosely
imbricated, open-erect, inflated toward the point or club-shaped, obtuse
or acute, some of them curved outside. As seen in fig. 13 h, no trace of a
middle nerve is recognizable upon them, even with a strong glass; but this
may be due to the thin crust of clayey matter which covers the whole plant.
There is upon the same specimen an obscure fragment, which appears to

46 UNITED STATES GEOLOGICAL SURVEY— TERTIAKY FLORA.

represent a cyliiulrical ear of Lycopodiu^n, which may be a fruiting branch
of this species. It is two centimeters long, three millimeters wide, and
seems to bear small round glomerules, like crushed sporanges of Lycopods.
These arc, however, too indefinite in form, and could not be satisfactorily
represented.

Habitat.— Near Elko Station, Utah {Prof. E. D. Cope).

SELAGINELLA, Beauv., Spring.

Selaginella Bertlioudi, Lesqz.

Plate V, Figs. 12, 12 a.
Selaginella Berthoudi, Lesqx., Annual Report, 1873, p. 395.

Frond dichotomous like the stems and branches, prostrate or creeping (f), slender ; divisions linear,
at an acute angle of divergence, short ; leaves four-ranked, lateral ones spreading, distichons, linear,
oblong or lanceolate, pointed ; middle leaves small, oval or nearly round, entire, closely appressed to the
base of the longer leaves and covering it.

This fine species, represented by very distinct specimens, seems to have
had, like many analogous species of the present flora, a large, difiuse frond,
either creeping or flattened upon the ground. All its divisions are dichoto-
mous and apparently on the same acute angleof divergence of 40° to 50°; the
ultimate ones are short, linear, obtuse or truncate, one to two centimeters long,
four millimeters^ide. The leaves are in two rows, the lateral ones open
distichous, imbricated by the lower side, longer, oblong, lanceolate-acute,
sessile, and distinctly nerved in the middle ; the intermediate ones, much
smaller, scarcely one millimeter long and half as broad, are of about the same
form, oval, obtusely pointed, and marked also by a middle nerve. All the
leaves are entire, alternate, and thickish, not pellucid, their surface being
generally covered by a thin pellicle of coaly matter. This plant, in its char-
ter, greatly resembles some of the present species of this genus inhabiting
sul)tropical regions, like S. stolonifera, S. Mertensii, etc.

Habitat. — Golden, Colorado {Capt. Ed. Berthoud), to whom this fine
species is justly dedicated.

Selaginella falcata, Lesqx.
Plate LXI, Figs. 12-15 ; Plate LXIV, Figs. 13, 13 o.
Selaginella falcata, Lesqx., Annual Report, 1874, p. 297.

Primary stem thick, round, dichotomous ; pinna) narrow, linear ; leaves close, two-ranked, distichous,
sessile, open, generally covering each other at the borders, entire, lanceolate-pointed, narrowed at their
point of attachment to a slender rachis, membranaceous, pellucid, without any middle nerve.

Though the stem, plate Ixiv, fig. 13, was not found connected with any
fragment of branches, the leaves which cover it all around, imbricating at

DESCKIPTION OF SPECIES— LYCOPODIACEiE. 47

the base, but generally directed toward both sides, are so exactly of the same
character as those of the branches of plate Ixi, figs. 12 and 14, that one can
but consider these fragments as representing the same kind of vegetable, its
stem and some of its divisions. The branches are slender, variable in length,
from one to three centimeters long, linear, five to ten millimeters broad, with
a very narrow rachis; the leaves distichous, nearly at right angles to the
rachis, are narrowed to the base, lanceolate-pointed, and slightly falcate,
generally close to each other, and often imbricating at the borders, without
trace of middle nerve; the substance is membranaceous and pellucid, of straw
color. A.S seen in fig. 15, some of the branches seem to come in fascicles out
of fragments which appear half decomposed, perhaps the bark of the stem
stem or a casual agglomeration.

The characters of this remarkable plant do not fully correspond to those
of Selaginella. Its divisions are not distinctly dichotomous, as seen in figs.
12 and 13 of plate Ixi, and the leaves are without trace of a middle nerve.
The form and position of the leaves might suggest its reference to some
peculiar species of Conifers, but the membranaceous pellucid leaves, without
middle nerve, as well as the very slender rachis, are against a reference of
this kind. It might perhaps represent some peculiar new type of floating
Fern. The fragments are very numerous, perfectly distinct; the characters
of the plant are exactly represented by the figures.

Habitat. — Point of Rocks, Wyoming {Dr. F. V. Hayden, Wm. Glehurn).

Selaginella laciiiiata, Lesqz.

Plate LXIV, Figs. 12, 12 o.

Selaginella laciniaia, Lesqx., Annual Report, 1874, p. 297.

Branches and subdivisioDS like those of the former species, leaves (?) deeply laciuiate ; lacina>i
linear, slightly inflated towaid the point, either simple from the base or dichotomous.

By their mode of vegetation and the form and division of the pinnae or
branchlets, these plants are exactly similar to those described above, and
indeed they seem to represent the same species under different forms. The
difference is in the laceration or thread-like lacinias of the leaflets. These
laciniae, distinct and in relief upon the stone, resemble the veins of Fern
leaves, when, by maceration and decomposition, the epidermis has been
destroyed, and the skeleton of the leaves only is left. In this case, the
thread-like branches are not equally and normally divided like veinlcts, but, as

48 UNITED STATES GEOLOGICAL SUEVEY— TEKTIAKY ELOKA.

seen in fig. 12 a, enlarged, they are either simple from the point of attachment
or forked once or twice, slightly tumescent at the point. Their distribution in
bundles along an axis is perhaps more analogous to that of rootlets of creep-
ing rhizomas, and the presence of thread-like filaments of this kind upon the
left branch of the specimen, plate Ixi, fig. 15, evidently part of Selagvmlla
falcata, might confirm this supposition; but there is no great analogy between
the characters of thread-like rootlets and those of this plant, for these are too
regular in their form, their divisions, and their curve in the same direction.
It may be that we have here, and in these two plants, the submerged and
the floating part of the same species ; that as it happens in some Phaenoga-
mous plants of the present flora, as, for example, in Nasturtium lacustre, the
immersed leaves are laciniate while the emerged ones are entire. I have,
however, vainly searched among species of Lycopodiaceoi of our time for any
traces of abnormal divisions like these ; even Lycopodium inundatmn has the
leaves of the immersed branches entire. On another side, some creeping
species of Selaginella have at the base of the stems, or upon the stolons, a
kind of tendrils, generally filiform, sometimes dichotoraous, divided toward the
ends, and the divisions slightly inflated. This character is marked in Sela-
ginella Mertensii, S. apus, but more distinctly in S. microphylla of Cuba, whose
tendrils are shorter and sometimes subdivided at the end in six or eight
branches. From this it appears that these two forms described above may
be parts of a species of creeping Selaginella, the stem, stolons, and tendrils
being represented on plate Ixiv, figs. 12, 13, the branches and leaves on plate
Ixi, figs. 12-15. The specimens figured upon this last plate are all from Dr.
Hayden, and, though very numerous, none of them has any trace of these
laciniate divisions, except the small branchlet, fig. 15, where the thread-like
filaments are simple, short, and apparently representing the base and remnants
of leaves destroyed by maceration, an appearance also less definitely marked
upon the lower branches of fig. 13: per contra, the specimens sent by Mr.
Cleburn from the same locality have only the part of stem, plate Ixiv, fig. 13,
and very numerous fragments of the form, fig. 12, but no branch with leaves
like those described as Selaginella falcata. For this reason, and that of doubtful
identity, they have been described and figured separately.
Habitat. — Point of Rocks, Wyoming ( Wm. Cleburn).

DESCRIPTION OF SPECIES— FILICES. 49

FILICES.

The Ferns are, after the Li/cojjodiacece, the oldest vegetables of the world.
From the Devonian, where already a number of species of peculiar types are
known, they become more and more numerous and predominant up to tlic
end of the Carboniferous; for, in this formation, they compose at least one-
half of the vegetation. After this, in the Permian, the Trias, the Jurassic,
and the Cretaceous, they are subordinate in number to the other groups of
plants, represented, however, in constantly modified characters appropriate
and particular to the different epochs. The Gkichenia, for example, appear
in the Jurassic, become predominant through the whole Cretaceous formation,
wherein a comparatively large number of species has been observed in the
Arctic zone, in Europe, and also in the Dakota group of Nebraska. In tlie
Tertiary, they disappear most entirely, and therefore it is not at all surprising
that among the numerous species of Ferns described from the Lignitic there
is no Gleickenia. This absence of a genus truly Cretaceous and generally
distributed through this formation, present also in the Dakota group, is,
among others, an important point of evidence of the Tertiary age of the
Lower Lignitic flora.

SPHENOPTERIS, Brgt.

Sphcnopicris L.akcsii, Lesqz.

Plate II, Figs. 1-1 a.

S^henopteris Eocenica, Eti., Eoc. Fl. d. M. Prom., p. 9, pi. ii, figs. 5-8.— Lesqx., Annual Report, 1872, p. 376.

Frond large, at least tripinnately divided ; secondary pinnaj long, linear-lanceolate, oblique to a

balf-round, n.arrow racbis ; piuunles obliquely turned upward, close, contiguous, and united below tbo

middle, acutely lobed; veins pinnate; divisions simple or forking once.

As remarked in the first description of this species, it differs evidently
from the one described under this last name by d'Etlingshausen {loc. cit.) by
the connection of the pinnules from below the middle, while they are separated
from the base in the European species; by the sharply pointed lobes of the
pinnules, described and figured as obtuse by the author, and also by the
nervation. In this species, the secondary veins are strong, flat, generally
simple, each of the divisions ascending to the point of a lobe. These differ-
ences are marked enough to force the separation of this species. From the
fragments seen of this fine Fern, it has the same characters in all the subdi-
visions of its fronds; at least the secondary pinnae are sessile, connected at

4 T p

50 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

tlu'ir base like the pinnules, and these arc cut in nioic or less deep lobes, also
of the same pointed form. The mode of division of the secondary pinnae is
seen in tig. 1 a. These are larger toward the base of the frond, and their
pinnules are pinnately, acutely lobcd; farther up, the pinnae are simply lobed,
like the pinnules. This fine Fern has also an evident relation to Aspknium
Wegmanni, Brgt., of the Sezanne Flora, by Saporta, p. 317, pi. 2, figs. 2 and 3;
a species described also by Watelet in the Flore Foss. du Bassin de Paris, an
Eocene formation like Se'zanne.

Habitat. — Golden, Colorado, where it is often found, but rarely in frag-
ments as large as the specimen figured and communicated by Rev. Arthur
Lakes, of the School of Mines of Colorado.

Splieiiopteris membranacca, LeBqz.

Plate II, Figs. 2, 2 o, 3, 3 a.
iSphenopleris memiranacea, Lesqx., Annual Report, 1873, p. 394.

Frond bi-tripinnate ; primary pinnce long, linear-lanceolate, rigid, erect or at an acute angle of
divergence from the main rachis ; pinnules narrow, linear-lanceolate, connected near the racbis by the
decurrent base, pinnately 5-6-lobed ; lobes short, slightly obtuse, distinct, singlt -nerved.

This form has nearly the same characters as the preceding, and may be
a variety of it. Its facies is, however, very diiferent, as seen from its mem-
branaceous shining substance, its rigid divisions, the narrower pinnules, all
separated to near the base, by which they are decurrent along the rachis, and
generally close to it, forming a narrow border. The slightly obtuse lobes and
the narrow pinnules give to this Fern a likeness still more marked to that of
the Mount Promina flora. It seems, however, to differ by its membranaceous
substance, and by the veinlets, always simple, while they are described by
d'Ettingshausen as forking in his species. The affinity is, however, evidently
very close. Fig. 2 a is from a fragment representing the same species, with
pinnules less deeply divided, and which seems intermediate between both
forms. The substance is also shining and membranaceous. A small frag-
ment, represented in figs. 3 and 3 a, has been merely mentioned in Rep.,
1869, p. 196, under the name of Lustrea arguta?, with the remark that it
might be a Pecopteris. It is evidently referable to this or the former species.

Habitat. — Golden, same locality as the former. The last fragment was
in the first lot of specimens sent to me from the collection oi Dr. J. L. Le Conte,
of Philadelphia. The other specimens have been obtained by Rev. A. Lakes.

DESCKIi'TlON OV SPECIES— FILICES. 51

S p iic II o ptcri s nigricans, Lesqz.

Plate II, Figs. 4-r)a.

S2)hcnopterin nigricans, Losqx., Annual Report, 1873, p. 394.

Frond polypinuate ; primary pinna) narrow, linear (as ninch as can be seen from tlio fraj^ment);
tertiary i(iiiii:e at a rijjlit an^le of diverfjenco from a sli^litly winded raeliis, short, sessile, linear,
abruptly rounded to a small, obtuse, terminal lobe; piunately diiiiply lobed ; pinnules in ri;;ht angle lo
the racbis, distinet to near their base, oblong, obtuse, d<^eply uiululate on the borders; middle vein
scarcely distinct, alternately piunately divide<l in four to six pairs of veiulets, curving in passing to the
borders, forking once, except the upper pair, which is simple.

The specimens are mere fragments ; the largest part (fig. 4) shows what
appears to be the middle of a pinna of second order, with alternate short divis-
ions two and a half centimeters long or a little more, one centimeter broad,
in right angle to the rachis, or very open, deeply undulate on the borders. Tiie
surface was apparently covered withacoatingof short hairs, as it is punctulate
and always colored black. 1 do not know any recent species of Ferns to which
this one is comparable ; its nervation is like that of some Ci/athea. Except
for its deeply undulate lobes, it is like Pteris hlechnoides, Heer, Flor. Tert.
Helvet., p. 40, pi. xii, fig. 8.

Habitat. — Golden and Black Butte ; always found in small fragments.

HYMENOPHYLLUM, Klf.

Ily lucnop ii y 1 1 uin coufiisuni, Lesqz.

Plate II, Figs. G-Ca.

Hymenophylliim confusum, Lesqs., Annual Report, 1873, p. 395.

Frond polypinnate; primary rachis thick, grooved ; ultimate pinna) lanceolate, deltoid in outline;
pinnules simple or bitid, cuneiform, enlarged upward from a decurring base, lobed ; lobes short, oblong,
obtuse, simple, bifid or emarginate; veins dichotomons; branches simple, each entering one of the divis-
ions of the pinnules.

The description is made from fragments, the species being represented
only by crushed parts of the frond mixed together and pressed upon another
in a confused mass. The largest portion discernible is the one figured. The
primary rachis is not represented, the fragments being too small, and irregu-
larly crushed ; the pinna; are apparently broadly lanceolate, divided to a dis-
tance from the rachis in pinnules, cuneate to their decurrent base, enlarged
and curved backward at the upper part, either simple and lobed or cut in two
divisions, as in the lower part of fig. 6. The lobes are either simple, oblong,
obtuse, or enlarged, and emarginate at the upper part into two equal, half-
round obtuse division's; the veins, simple and decurrent to the rachis, at the
base of the pinnules are dichotomous in ascending each of the ultimate divi-
sions, reaching the borders of one of the lobes, as seen in fig. Ga. When looked

52 UJJITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

at with a strong glass, the surface appears minutely punctulate, as if it had
been covered by a kind of soft villosity.

By the shape of its pinnules, decurrent upon the rachis, their divisions
and nervation, this Fern evidently resembles some of the living species of
Hymenophyllum. But it is as yet too indistinctly represented to offer posi-
tive points of comparison.

Habitat. — Golden, same place as the former species.

PTERIS, L.

Ptcris pseudopennseformis, Lesqz.

Plate IV, Figs. 3, 4.

PterU pennaformis, Heer, Fl. Tert. Helv., p. 38, pi. xii, fig. 1. — Lesqx., Annual Report, 1873, p. 392.

Pinnae linear-lanceolate, gradually narrowed upward, entire to above the middle, obtusely den-
tate at and near the point, subcoriaceous ; middle nerve thick, bi-grooved, and three-striated toward the
base ; veins in an acute angle of divergence, close, thin, mostly simple or forking once.

The fragments representing this species, and of which the best ones have
been figured, seem referable to the Miocene form described by Heer. The form
of the leaflets is the same ; they are also crenulate-dentate toward the point, and
the veins, at the same acute angle of divergence, are, as in the European spe-
cies, either simple or forking once. There is, however, a difference in the
more obtuse denticulations, which the author describes as somewhat sharp
{ziemlich scJiarf), and especially in the veins, which appear more numerous
and close in the American form.

A comparison of specimens only could determine either identity or a spe-
cific difference between the European and American forms.

Habitat. — Henry's Fork (Dr. F. V. Hayden).

Ptcris subsimplcx, Lesqz.

Plate IV, Figs. 5-7.

Pieris auisimplex, Lesqx., Annual Report, 1873, p. 392.

Fronds or pinnee coriaceous, large, entire or minutely crenulate, ovate-lanceolate, gradually nar-
rowed in a curve to the base ; middle nerve narrow, deep ; veins at an open angle of divergence, dis-
tinct and distant, simple, or merely forking once near their base or toward the middle.

The fragments of this Fern represent either simple fronds or some sepa-
rate pinnae of a large frond. They are ovate-lanceolate, tapering gradually
to a point (broken), and narrowed in a curve to the base. As seen in fig. 6,
the fragment is slightly uncquilateral at its base, and thus resembles a pinna
rather than a frond ; the part in fig. 5 also is different in shape and size, as

DESCEirXION OF SPECIES— FILICES. 53

would be a terminal pinna. It seems, therefore, that tliese fragments are
mere leaflets. They are in size from eight to ten centimeters long, and from
one and a half to three centimeters broad, vvilh a narrow midrib, and tliick,
distinct veins, all under the same angle of divergence, G0°, simple or forking
once, and reaching the borders straight, or witliout deflecting upward or down-
ward. The borders, by contraction at the point of connection of the veins,
are slightly crenulate. This character is unlike that of tlic nervation of
Osmundii leaves, to which this Fern might be compared, by the Ibrm of the
leaflets, resembling those of plate iv, fig. 1. Taken for simple fronds, they
have a relation to Acrostichum latifoHum, Sw., var. alismcefoUum of Cuba,
which has the same type of nervation. As leaflets of a compound frond, they
are comparable io Acrostichuin cervmum, Sw., or Lomariopsis Wrightii, Mett.,
both Cuban species also.

Habitat. — Golden, Colorado.

Ptcrls erosa, Lesqx.

Plate IV, Fig. 8.

Pteris erosa, Lesqx., Supplement to Annual Report, 1871, p. 12 ; Anntial Report, 1873, p. 392.

Pinna) large, broadly lanceolate, taper-pointed ; border obtusely dentate toward tbe point, irreg-
ularly crenulate-laceralo in the middle ; middle nerve thick ; veins at an open angle of divergence, fork-
iug once or twice, slightly turned up in reaching the borders.

The only specimen of this Fern figured here represents the point of a
large pinna, the fragment being five and a half centimeters long and three
centimeters broad toward the middle. The pinnae are near the point regu-
ularly obtusely dentate ; but lower the teeth become irregular in form and
size, and more or less deeply eroded. The veins are about at the same angle
of divergence as in the former species, to which this one has a ch)se relation ;
they differ, however, by more numerous divisions, and especially by tiieir dis-
position to slightly curve up in reaching the borders. By this character, the
species is allied to Fteris parschlugiana, Heer, Bornst. Fl., p. 7, j)!. 1, fig. 1,
which has the veins divided about in the same way.

Lately, after the printing of the plate and the preparation of the descrip-
tion, I have received from Kev. A. Lakes, and found at Golden, a splendid
specimen, with one leaflet preserved in its integrity. It is seventeen centi-
meters long, three and a half centimeters broad below the middle, linear-oblong,
rapidly narrowed near the point to an acumen, sharply, nearly equally dentate
from a little above the base, which is unequilateral, rounded on one side,
cuneate on the other, broken near the'point of attachment, where it is narrowed

54 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

to u short pedicel half a centimeter broad. This basilar conformation shows
it to be a leaflet of a very large pinna.

Habitat. — Raton Mountains, near Trinidad, New Mexico ; Golden, Col-
orado, where it is not rare, mostly found in fragments. The figured specimen
is from this last locality.

WOODWARDL&, Sm.

IVood w ardia latiloba, Lesqz, ,

Plate lU, Fig. 1,1a.

Tfoodwardia latiloba, Lesqx., Annual Report, 1873, p. 391.

Frond large, blpinuatifid ; pinnae opposite, decurriug along the round rachis, long, linear, gradu-
ally tapering to the point, equally pinnately lohed; lobes dinjointed to three-fourths of their length,
broadly lanceolate, obtusely pointed, scythe-shaped upward ; middle nerve thick, distinct to the point
of the lobes; secondary veins parallel to the rachis and the midrib, branching upward; areolation
formed by anastomoses of the divisions in one or two rows of large, irregular meshes ; ultimate veiulets
parallel, passing up obliquely to the borders.

From the fine specimen figured, the upper part of a large pinna, the
fronds appear to have been long, linear, very gradually diminishing upward;
secondary pinnae also comparatively long, twenty-one to twenty-four centi-
meters at least, linear, slightly broader in the middle, and gradually narrowed
to the point, three and one-half to four centimeters broad ; lobes equal,
disjoined to three-fourths of their length, entire or slightly crenulate, broadly
lanceolate, obtusely pointed, with narrow obtuse sinuses between them,
subfalcate, six to seven millimeters broad; upper pinnae more and more
obtusely and less deeply lobed, passing to mere equal undulations; middle
nerve thick, ascending to the point of the lobes; areolation discernible only
on the lower surface. As seen in fig. 1 a, the lower secondary veins follow
parallel to the rachis, branching upward, and forming one or two rows of
large rectangular areolae with brauchlets, either simple and parallel, or
forking near the point and entering the borders. The consistence of this
Fern is thick, coriaceous; the upper surface smooth, nearly polished. The
fructifications are as yet unknown.

Habitat. — Golden, Colorado; found in numerous large fragments, and
communicated by Jlev. A. Lakes.

Wood ward! a latiloba var. minor, Lesqz.

Plate IV, Figs. 9, 9«.
JVoodtvardia latiloba var. minor, Lcsqx., Aunual Report, 1873, p. 391.

PiniKu Hinall ; rachi» narrow ; lobes short, obtuHe ; secondary veins less divided.

This small fragment aj)pears to represent a mere variety of the species

DESCRIPTION OF SPECIES— FILICES. 55

described from Golden. As it represents only the end of ii pinna, the lobes
arc of course shorter and more obtuse; the subdivision of the secondary
veins is less multiplied, foniiiu<f a singli! row of basilar areas and the veins
generally merely i'orking. Tliis diilerence in tliat nervation is also probably
due to the position of the fragment relatively to the frond. This form is
related to Woodtcardiles arcticus, fleer, Flor. Arct., ii, p. 4G2, pi. .\I, iig. G.

Habitat. — ]51a(;k Butte, Wyoming. Days of researches at the same
locality failed to procure any other specimen than the fragment figured.

Diplaziuni ITIuellcri: Heer.

Plate IV, Figs. 10, 10 a.

Diplazium Muellm, Heer, Borust. Flor., p. 8, plate i, fig. 2 — Ltsqx., Annual Report, 1873, p. :W3.

Pinii.i; coriaceous, simple, nanowly lanceolate, gradually tapering to a long acumen; borders
margined, distantly, etinally, sharply serrate ; middle nerve broad, grooved ; veins at au acute angle of
divergence, very close, dichotomous, and a.iastomosing in right angle by cross-veinlets.

This Fern, as represented l)y the only fragment figured, is remarkable by
its thick sub.stance, truly coriaceous, and its inflated, cartilaginous, sharply
serrate margin. By the form of the pinna and its dentate borders, it is
evidently related to Heer's species. It is, however, difierent by the nerva-
tion, the veins being more divided and distinctly anastomosing. As the
coriaceous texture of this leaf does not allow to see clearly the disposition of
the lateral veins, and especially the anastomoses, which are visible only in some
parts where the epidermis is destroyed, it might be supposed that the Euro-
pean specimens, in a better state of preservation, did not expose all the
details of nervation as they are seen upon the fragment, fig. 10 a, a part of
fig. 10, w^ierethe epidermis is erased. Heer mentions the leathery texture
and peculiar dentate borders, which in his leaf are doubly dentate, and tiiere-
fore less regular than in this fragment. Tiiough it may be of the relation
of these forms, the nervation with its anastomoses seems to remove the
American species from the genus D/p/a'hn/i. It has some analogy with
species of Acrosiichum, A. alismafulium, for tlie thick border, and A. aureum,
for the close anastomosing veins; but this is a distant relation indeed. It
might be compared also to some species o[ Aneimia.

Habitat. — Henry's Fork {Dr. F. V. Ilayden).

56 UNITED STATES GEOLOGICAL SUEVEY— TERTIARY FLORA.

LASTREA, Presl.

Lastrca (Ooniopteris) Ooldiana, Lesqz.

Plato IV, Fig. 13.

Aajfidium Goldianum, Lesqx., Annnal Report, 1873, p. 393.

Frond bi-tripinnatifid ; primary piDOSB broadly deltoid in outline ; secondary pinnie linear,
alternate, parallel, at an obtuse angle of divergence, alterpately and equally pinnately lobed ; lobes cut
to two-tbirds or tbree-fourths of tbeir lengtb, oblong-lanceolate, obtusely pointed, inclined outside;
middle nerve distinct; lateral veins five to seven pqire, curving inward, simple, parallel; borders slightly
crennlate by the impressions of the veins.

The substance of this Fern is somewhat thick or subcoriaceous; the
secondary pinnae, apparently long and rapidly decreasing toward the point of"
the pinnae, are sessile, the inferior lower leaflet being free or not decurrent
along the rachis; the lobes are really entire as far as can be seen from the
recurved borders, and the crenulation is an appearance caused by the impres-
sion of the veins; the rachis, both of the primary and secondary pinnae, is
narrow, smooth. Its affinity is with a number of species of Cuba, like
Lastrea contermina, Desv., L. lonchodes, H. K., and especially L. scoloi)endro'ides,
Mett., var. pinnata. In the fossil species, its relation is with L. serrulala,
Heer, of the Boernstadt flora, differing widely, however, by the nearly entire
borders, and the lateral veins more numerous and at a more open angle of
divergence.

To this species are apparently referable a number of fragments repre-
senting mere parts of pinnae, with pinnules slightly longer than those of fig.
13, but of the same form and with the same nervation. One of the specimens
has a small part of a fructified pinna, the leaflets bearing small round sori
attached upon the middle of the simple veinlets at equal distance of the
borders and the midrib, as in species of Goniopteris.

Habitat. — Golden, Colorado. The fragments with fructifications have
been lately procured by Rev. A. Lakes.

liastrea (Goniopteris) intermedia, Lesqx.
Plate IV, Fig. 14.
Atpidium {Laatrea) palcliellum ? or A. Fiacheri f, Lesqx., Annual Report, 1870, p. 384.

Frond pinnate ; pinnae linear, oblique, alternate, connected by the decurrent base, pinnately
equally lobed; lobes divided to near the base, oblong, obtuse, oblique; lateral veins simple, curved
upward.

The specimen represents part of an apparently large pinna; the divisions
are short, linear, one to two centimeters broad, rapidly narrowing toward the
top of the frond, parallel, at an acute angle of divergence, oO°, and decurring

DESCRIPTION OF SPECIES— FILICES. 57

along the rachis by the expansion of the lower leaflets ; the lobes turned
upward are oblong, narrowed into an obtuse point, alternate, free to below
the middle; the middle nerve is strong, continuous to the point; the lateral
veins simple, five or six pairs, open, and curving upward in passing to the
borders. Comparing this species to both Lastrea pulchella and L. Fischeri,
Heer (Flor. Tert. Helvet., I, pp. 33 and 34, pi. ix, figs. 2 and 3), the close
relation of this fragment is easily remarked, the form of the lobes being
analogous to that of fig. 2, and the disposition of the veins, which are more
numerous, like that of fig. 3 of the Flor. Tert. There is, however, a difierence,
which prevents considering the American form as identical with these Euro-
pean species ; it is the evident connection of the pinnge by the lower and
inferior pinnules, broadly extending and decurrcnt by their base along the
main rachis, a character marked only, and indistinctly, upon L. imlchella, fig.
2 c of the same plate. It seems, therefore, advisable to consider this species
as separate. The specimens briefly described in Report, 1870, were mere scat-
tered fragments of pinnae, none of them showing the mode of attachment
to the rachis. This form is also comparable to AlsopUla Pomelii, Sap. (Sc'z.

FL, p. 40, pi. 3, fig. 2).

Habitat. — The specimens described in Annual Report, 1870, are from
Henry's Fork, a mixed lot ; the first one sent by Prof. F. V. Hai/den. That
figured here is from Golden, Colorado.

Lastrea ( Goniopteris) p o 1 y p odi oides? , Ett.
Plate IV, Figs. U, 12.
Goniopteris polypodMdes, Ett., Flor. of Prom., p. 10, pi. ii, figs. l-4.-Lesqx., Annual Report, 1873, p. 394.

Pinna) lar-^e, linear-lanceolate, pointed, undulately lobed, indistinctly denticul.ate ; primary veins
equidistant, parallel ; lateral veins at an acute angle of divergence, apparently alternate, simple, and
curved inward.

The fragments figured here, and described as probably representing
the species of d'Ettingshausen, are too obscure for positive identification.
The divisions, or lobes, are scarcely marked by undulations along the pinnic,
and also the distant, irregular, small teeth are not distinctly seen, the borders
appearing here and there like serrulate. The nervation is of the same type,
and quite as obscure as it appears to be on the European specimens. Judg-
ing from the figures of the Prom. FL, there is therefore a marked relation
between the species, but the identity is doubtful.

Habitat.— Sand Creek, Colorado ( W. II. Holmes).

58 UNITED STATES GEOLOGICAL SURVEY— TERTIAlll' FLORA.

GYMNOGKAMMA, Desv.

Oyin iiogra iiiiu a Gardncri, Lesqx.
Plate IV, Fig. 2.
Pieris Gardncri, Lesqx., Annual Report, 1873, p. 39.3.

Frond l.irgo, simply pinnate; pinuas large, linear, broader in the middle, in right angle to tbo
racLis, rounded to tbe base; borders deeiily undulate ; middle nerve broad, grooved in the middle, flat-
tened on the borders; veins in an obtuse angle of divergence, abruptly curving downward at tbo base,
or decurring to the racbis, forking once or twice, joined by anastomoses and forming by cross-branches
irregular, long areola?.

The large pinnse, three to three and a half centimeters broad in the
middle, where they are enlarged and broken, apparently lanceolate, with a
broad flat midrib, narrowly grooved in the middle, have entire but irregularly
undulate borders, and lateral veins at an open angle of divergence 70 to 80°,
thin, forking once near the base, sometimes twice, and here and there irreg-
ularly anastomosing by cross-veinlets. The substance is rather thin. I con-
sidered first this Fern as referable to Pteris^ as some species of this genus
have pinnjE of the same form, and the lateral veins sometimes anastomosing
as in Pteris grandifoUa Linn., which, however, shows this kind of division
of the veins only near the borders, and far more regular than in this species.
The genus Pteris has now two essential divisions: Heteroplilebium and
CamjJteria ; the first one has leaflets with lateral veins anastomosing toward
the margins, the second with veins connected at the base by arching vein-
lets. I supposed that the fragments described here might represent a section
intermediate between both, or with veins anastomosing in the middle of the
areas. But there is in Gymnogramma a species, G. Jajwnica, Desv., whose
characters are so evidently comparable to those of this fossil one that it seems
rational to admit it in the same genus. The Japan Fern has long, simple,
linear, oblong, acuminate pinnse, rounded at the base, the racbis flat on both
sides of the midrib, the veins close, forking once or twice, and joined by
cross-veinlets; all this as in the fossil species. The comparison of speci-
mens shows the relation to be very close.

Count Saporta informs me that quite recently a Fern has been discov-
ered in the Gypses of Aix, representing a species winch he considers as iden-
tical with this, or at least closely allied. He is disposed to refer it to the
genus Chnjsodlum, a ."-cction of I lie Acrosiickimi.

Habitat. — lioof ol' a coiil-niinc, Sand C!r('<'k, Colorado {A. (lardner).

DESCEIPTION OF SPECIES— FILIGES. 59

O y m n o g I- St in in a II a y d c n ■ i , Lesqz.

Plate V, Figs. 1-3.

Gymnogramma i7ay(J«nit, Lesqx., Annual Report, 1871, ji. 2'X>.

Frond bi-tripinnatiful ; iiinn;u long, lanccol.ate, obtusely pointed, pinnately divided to near the
r.-ichi8 in oblong-lanceolate, obtusely pointed lobes ; the lower ones distantly dentate, the upper ones
more or less distinctly creu.ate ; r.achis narrow ; middle nerve thin ; lateral veins close, at an acute angle
of divergence, dicbotomous.

The tliree specimens figured give a tolerably good representation of the
form and of the characters of a pinna of this fine species. These pinnae were
evidently very long, gradually tapering to the point, the lobes becoming
shorter and proportionally narrower, though distinct, to the base of the; ol)-
tusely pointed terminal pinnule. The largest lobes, oblong, linguiform, are
distantly dentate, while in ascending, the marginal divisions, gradually eflaced,
are merely crenate and the upper ones entire ; ihey are joined in acute sinuses
a little above the narrow rachis, and slightly decurrent. As the substance
of the Fern is thick, coriaceous, the veins, when the epidermis is not destroyed
by maceration, are not very distinct, the middle one being otlen effaced and the
division of the lateral veins here and there obsolete. These are thin, nearly
straight in passing to the borders at an angle of 30° to 40° from the midrib,
join the main rachis between the pinnules, preserve tlie same direction, and
are dichotonious, generally forking twice. The largest lobes, as seen in fig.
2, are six centimeters long, one and a half centimeters broad in the middle,
where they are somewhat enlarged. The division of the veins is seen in fig.
2 a enlarged. The surface of the pinnae is covered by particles of what
resembles a pulverulent matter hardened into coal. I do not know of any fossil
sjiecies to which this one might be compared. Spfienopieris (^Gymnogramma)
Blomstrandi, Heer (Flor. Arct, i, p. 155, pi. xxix, figs. 1-5), has only a distant
likeness by the outlines of its lobes, but is totally different by its nervation,
etc. This fossil Fern has a higher degree of affinity to G ymno gramma, tar-
tarea, Desv., a common species of the present flora of Tropical America,
especially related to some of the varieties obtained by cultivation.

Habitat. — Divide between the source of Snake River and the southern
shores of Yellowstone Lake {Dr. F. V. Haijdcn).

60 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

OSMUNDA, Linn.

Osniiiiida afriuis, Lesqz.

Plate IV, Fig. 1.

Pteris affinis, Lesqx., Annual Report, 1873, p. 392.

Frond simply pinnate; pinnaj subcoriaceous, short, oblong-l.anccolato, broader in the middle, rap-
idly decreasing to an obtuse jioint, gradually narrowed downward, and apparently rounded to the point
of attachment ; borders entire or irregularly undulate ; nervation thin but very distinct ; lateral veins
open, dichotomous, generally forking twice, distant.

The position of the pinnae (the middle one seemingly terminal), their
size and form, and also the characters of the nervation of this Fern seem to
indicate its generic relation with Osmunda rather than with Pteris. The
pinnae oblong, enlarged in the middle, tapering to a point, about five centi-
meters long and two centimeters broad, probably represent the three upper
leaflets of a frond; this arrangement may be, however, casual, for the upper
pinna, which appears to be terminal, though somewhat larger than the others,
and of a different shape, has the midrib exactly of the same size. The
lateral veins, at an open angle of 50° about, slightly curve from their point
of attachment in passing to the borders, generally forking twice, and are very
distinct and distant, as in species of Osmunda. The rachis is not preserved.

Recently, I have received from Rev. A. Lakes a splendid specimen, rep-
resenting part of a large pinna with four leaflets, a terminal one, the others
alternate, very oblique, joined to a broad rachis, which, in this specimen, as
in the one figured, is destroyed by maceration, its place being merely indi-
cated by the position of the leaflets. These are oblong-linear, four centi-
meters broad, at least sixteen centimeters long, judging from the preserved
part of one of the leaflets, which, broken apparently at the middle, measures
ten centimeters from the base. The nervation is of the same type as that of
the figure, the veins more generally forking quite near the base at the point
of attachment to the rachis. The substance is hard, rather coriaceous.

The fragment described in Report, 1872, p. 376, as Pteris anceps appar-
ently refers to this species.

Habitat. — Golden, Colorado ; not rare in fragments.

DESCIUPTION OF SPECIES— FILICES. 61

LYGODIUM, Sw.

Lyitodiuiii ncuroptcroidcs, Lesqz.
Plate V, Figs. 4-7 ; Plate VI, Fig. 1.
Lygodium neuropteroides, Lesqx., Anuual Report, 1870, j). /!84 ; 1871, p. 284.

Pinnules cordate, two to five palmately lobed, divisions oblong or obovate, lanceolate, obtuse;
middle nerve thin ; lateral veins close, numerous, dichotomous.

This fine species covers by its remains, mostly leaflets, with rhizomas
and their divisions, large and numerous specimens of soft-grained yellow clay.
Fragments of stems or rachis are rare, even, indeed, it is uncertain if the
thinly striated, flat fragments of woody tissue, varying from two to twelve
millimeters in width, represent stems or rachis, for they may be merely half-
decomposed branches of the rhizomas. The leaflets, narrowed downward to a
cordate base, are two to four palmately parted to the middle more generally,
sometimes lower; the divisions, oblong, enlarged upward, and very obtuse or
linear-lanceolate and obtusely pointed, are four to eight centimeters long, even
more, from the base of the leaflets ; the lateral ones generally shorter, all
obliquely diverging, with more or less obtuse sinuses, are entire or slightly
wavy on the borders. The nervation is simple for each division of the
leaflets, the middle nerve of each remaining distinct to the base, where it is
generally separated from the others by secondary flabellate veins; lateral
veins emerging in a very acute angle of divergence, dichotomous, curving in
passing to the borders. The lowest veins are forked three to four times, the
upper ones only twice, all very close, especially along the borders, where
thirty to thirty-three veinlets are marked in one centimeter. The division of
the lateral veins is marked in fig. 5 a, enlarged ; fig. 5 shows the longest of the
divisions of a leaflet, which appears only twice lobed. The rhizomas, two
and a half centimeters broad, the largest seen, are pinnately divided in
numerous alternate branches, six to eight millimeters broad, narrowly striate,
subdivided in branchlets about one millimeter wide, covered with very
numerous, close radicles, mostly in right angle, of various length, and of the
same thickness in their whole length. Whole large slabs are covered with
fragments of these rhizomas and of their branches, one of which is represented
in pi. vi, fig. 1. Though the connection to the leaflets, is not seen anywhere,
they evidently belong to the same species, as no remains of any other kind
are preserved upon the same specimens

By the division and the form of the pinnules, this species is related to
Lygodium Gaudini, Heer (Flor. Tert. Helv., i, p. 41, plate xiii, figs. 5-15),

62 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

clifTering greatly, however, by the broader divisions, and the nervation, which
is of a far different type. Indeed, I do not know any species of this genus,
either fossil or living, whose secondary veins are so close and as many times
forking, resembling, by this character, species of Neuropteris of the Carbon-
iferous age. The consistence of the leaves is somewhat thick.

Habitat. — Barrell's Springs, Washakie group {Dr. F. V. Hayden). I
have not visited the locality and I mark the geological reference of this spe-
cies as it is indicated upon the labels of the specimens sent to me already in
1870, Most of these specimens have leaflets and rhizomas only of the same
Lysodium. The others bear, besides this species, fragments of Palm rays
and of Equisetum Haydenii. The fragments of Palms have double, compara-
tively broad (four centimeters), distinctly veined rays, the primary veins at very
irregular distances, the secondary ones extremely thin, a kind of nervation
similar to that of Flahellarla Zinkeni or of Sabal communis, two species
described in the following pages. These fragments are undeterminable,
but merit to be mentioned on account of the geological references. As
yet, species of Palms have been found in Colorado and Wyoming, only
with the Lignitic No. 1. But though they are positively considered as
characteristic of the Lower Lignitic, or Lower Eocene, in its separation from
the Cretaceous, they may be found in the upper stages. A Sabal, for
example, is, in good specimens, with the plants of the Upper Tertiary, or
Pliocene, of California. In this case, however, and in the presence of a
Fern, type of a subtropical climate, we see a discrepancy in the climatic cir-
cumstances indicated by the other species considered as belonging to the
same geological horizon. No. 4, whose vegetable forms represent a more tem-
perate zone. It is therefore advisable to consider the reference of this
locality to the Washakie group as uncertain until we have more reliable docu-
ments in regard to its geological station. The fragments from the same locality,
referred to Cyperus Deucalionis, }ieer, in Rep., 1870, p. 384, though compara-
ble to the species by their nervation, are too obscure to merit consideration.

L. y g o d i 11 m M a r v i n c i , Lesqz.

Plato V, Fig. 8.

Lygodium Marvinei, Lesqx., Anuu.-vl Report, 1874, p. 309.

Leaflets tripartite?; divisions oblong, narrowed to a point, serrulate toward the point; middle
vein thin, very distinct ; secondary veins twice forked.

We have only the fragment figured here, too incomplete to give a good

DESCRIPTION OF Sl'EClEiS— FJLICES. 03

idea of the form aiul mode of division of the pinnules, but evidently repre-
senting a species of this genus. The divisions are short and comparatively
broad, oblong, pointed, and distantly serrulate near the point. The secondary
veins are on a more open angle than in the former species, diverging al)ont
40° from the middle nerve, more distant, and forking only twice, or even
once. This species has no relation to any fossil of the same genus, for no
fossil Lygodium has been described as yet with serrate or dentate lobes. In
the living Ferns, Lygodium venustum, Sw., may be compared to it.

Habitai. — Top of gypsum series, Grand Eagle Junction {A. R. Mar-
vine). This formation is referable to group No. 1.

Lygodium Dcntoiii, Lesqz.

Plate LXV, Figs. 12, l.'i.

Lygodium Dentoni, Lesqx., Annual Report, 1874, p. 309.

Leafleta small, tripartite, with short obtuse divisions and broad sinuses; primary veins three,
from the base, distinct; secondary veins also very distinct, forking once or twice, close along tho
borders.

The leaflets, round truncate at the base, rapidly widen upward, and,
near the top, are tripalmately divided in short, very obtuse, entire lobes,
separated by very broad sinuses; the middle nerves, somewhat thick, are
joined at or near the base ; the secondary veins, also comparatively thick and
very distinct, ascend at an acute angle of divergence, those between the lobes
nearly straight upward, the lateral ones more curving, all forking once or
twice, and becoming close to each other along the borders. This species
is intimately allied to Lt/godiu?n exguisilum, Sap. (lilt., iii, 2, p. 88, pi. i,
fig. 13), of the Gypses of Aix, a species diflering by the secondary ner-
vation, more simple, the veins more distant along the borders, and the
leaflets merely bifid. This last character is, however, of little importance, for
the only specimen figured by the French author bears, on both sides of the
leaflets, short obtuse lobes like those of a tripartite or (luadripartite pinnule
before its full development. In our species, the leaflets are somewhat larger,
two to three and one-half centimeters broad between the points of the lateral
lobes, which are one and a half to two and a half centimeters long, and as
broad as long.

Habitat. — Green River group, near the mouth of White into Green
River, Utah (Prof. William Denton).

64 UNITED STATES GEOLOGICAL SURVEY— TERTIAKY FLORA.

liygodiuni conipactum, Lesqz.

Plate V, Fig. 9.

Lygodium compactum, Lesqx., Annual Report, 1869, p. 19G.

Pinnnle or lobo linear-lanceolate, entire, slightly broader above the base ; lateral veins on a very
acute angle of divergence, all very close from the base, forking once or twice.

This fragment is scarcely sufficient to indicate the generic relations. By
its close lateral veins it might be compared to Lygodium neuropteroides, de-
scribed above. But the veins in this species, though as close to each other,
are more straight and less divided. The borders are lacerated on the right
side of the leaflet toward the base, and thus it appears to represent a mere
lobe of a palmately divided pinnule.

Habitat. — Marshal's Mine, Colorado {Dr. F. V. Hayden).

RHIZOOARP^.

This subclass is represented in a fossil state by Salvinia and Pinnularia
only. In the present flora, Salvinia has representatives in the boreal Asiatic
regions, and also in tropical America. Europe has one species only, S. natans,
which, according to Pursh, has been found also in a small lake of New York
State, but has not been seen until now by any other botanist. Five fossil
species have been described from the Miocene of Europe.

SALVINIA, Mich.
Salvinia cyclopliy Ila, Lesqx.

Plato V, Figs. 10, 10 o.

Salvinia cyclophyUa, Lesqx., Annual Report, 1873, p. 408.

Leaves nearly round, truncate or slightly cordate at the base, very entire ; middle nerve thin
lateral veins alternate, on a broad angle of divergence ; areolie polygonal.

The leaf, finely preserved, is twenty-one millimetres long, twenty-five
millimetres broad, therefore slightly reniform, with a half-round, narrow
middle nerve and lateral veins very thin, scarcely thicker than the nervilles,
on an open and variable angle of divergence, some of them in right angle to
the midrib, and apparently running to the borders; the areolation is irregu-
larly quadrate, polygonal, comparatively small, distinct.

By the form of its leaves, the species is related to Salvinia Reussii, Ett.
(Bil. Fl., p. 18, pi. ii, fig. 21), which, however, has a difl^crent areolation.
Our figure shows the secondary veins slightly more distinct than seen upon
the specimens, where they are scarcely distinguishable without a glass.

Habitat. — Michlle Park, Colorado {Dr. F. V. Hayden).

DESCKIPTION OF SPECIES— RIIIZOCAKP^. 65

Salvinia Allcni, Lesqz.

riiito V, Fi^. 11.

Ophioglossum AUeni, Lesqx., Aunual Keport, 1^72, p. 371.

Leaves oval, rounded in narrowing to the ba.se; lateral v<miis, none visible; areolic large, irregu-
larly sqnare or equilateral, inordinately distributed.

Leaf about three and a half centimeters long, twenty-two millimeters
broad, of a thia substance, with a thick middle nerve a>i<l irregularly quadrate
meshes, formed of very distinct black nervillcs, the primary oiuis more or less
in right angle to the middle nerve, with oblique, generally parallel veinlets
between them. The borders are black, slightly undulating, as if they
were formed by a vein coming out of the attenuated base and following
them. By its form, its areolation, its size, all its characters, indeed, it is
remarkably similar to Salvinia reticulata, Heer (Fl. Tert. Helv., iii, p. 156,
jjl. cxlv, fig. 16). It merely ditlers by the surface not smooth or polished,
indicating a very tliin, pellucid leaf; by the areolation not quite as large, all
the meshes in tlie same direction, rather tending upward than downward ; by
the gradually narrowing curve to the base, which form a regular oval leaf.
As in Heer's species, this one has its areas free, without trace of reticulation.
It may represent the same species.

Habitat. — South Park, near Castello's Ranch {Prof. J. A. Allen).

Salvinia atteniiatn, Lesqx.

Plate LXIV, J^igs. 14, 14 <i.
Salvinia attenuata, Lesqx., Aunual Report, 1874, p. 296.

Leaves small, opposite, joined by a narrowed, sbort-pcdiceled base, broadly oval or round, indis-
tinctly reticulated in vertical parallel rows of large quadrate cells, marked in the middle by black spots
composed of very small close cells or pores, without trace of a middle nerve.

The species is represented by the two small leaves figured here, one, the
largest, exactly round, measuring one centimeter both ways, the other, smaller,
ovate, only eight millimeters long and live millimeters broad. The leaves,
joined at the base, attenuated to a short pedicel, have no trace of a middle
nerve or of a separation in the middle, the surflice being composed of large
areolae exactly square, formed by veinlets ascending from the base and di-
verging, and from parallel veinlets crossing them in right angle from the
borders. These secondary veins are indistinct, and the surface of the leave*
appears, with the glass, like a small checker-board, with the squares marked
in the middle by an obscure spot apparently formed of round pores, or like a
very small wart. This species is closely related by its areolation and the
5 T F

06 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

size of the leaves to Sahnnia Mildeana, Gopp., as described by Heer (Bait.
Flor., p. 17, plate iii, figs. 1 and 2), diflTering by broader, less distinct, areolae,
the absence of a dividing middle nerve or free line, and the base narrowed
to a short pedicel. These differences may be due to the unripe state of the
plants, perhaps, when imbedded in the matrix where they are preserved.
This seems indicated by the smaller size of one of the leaves and the
indistinct areolation. Salv'mia natans of the present flora has its leaves
scarcely nerved in the middle before they attain their full development, or
rather the middle nerve becomes visible only upon full-grown leaves. The
more marked difference is in the form of the base of the leaves, which is cor-
date in the European species. This character is preserved also in the figures
and description of <S. Mildeana, Ung. (Sill., p. 5, pi. i, figs. 7, 8).
Habitat. — Point of Rocks ( Wm. Cleburn).

CALAMARI^.

Like the Ferns, this class of plants had its predominance in the Car-
boniferous period, when it was especially represented by the Cnlamites and
Equisefiles, already present in the Devonian ; plants of large size, reaching
the dimensions of trees, with trunks sometimes twenty centimeters in diam-
eter. The section of the Equisetacece has followed afler the disappearance
of the Calamiles, from the Keuper to the present time. In the old forma-
tions, these plants are still of large size, but from the Tertiary we find the
genus as it is now, with species whose stems are rarely more than two centi-
meters thick. They are about equally distributed from the Triassic upward.
This formation has ten species, the Jurassic has eight, the Wealden five ;
from the Cretaceous, two only are known, and fifteen from the Tertiary.
The Schizoneura, another genus of this order of plants, is Triassic mostly,
one species only being known from the Oolite. Phillotheka represents spe-
cies found in the Oolite of New Holland. Its place seems intermediate
between Calamites and Equisetum.

In the present vegetation of the world, the group of the Equiselacece
has about twenty-five species known until now. They are distributed in
'Europe, Asia, and the American continent, which has the largest number,
twenty-one species, of which nine belong exclusively to its flora.

DESCEIPTION OF SPECIES— EQUISETACE.E. 67

EQUISETACE^.

EQUISETUM, Linn.
E q u I s c t II in II a y d <■ u i i , Lesqx.

i'lato VI, Figs. ;>-.l.

EquUetum Haydenii, Lesqx., Annual Report, 1871, p. 284 ; 1872, p. 38,'').

Khizomas tbick, cylindrical, irn^gularly distinctly striate, articulate ; articulations distant, bear-
ing large oval or obovate tubercles narrowed to a round point of attachment, in whorls of eight to tcq
around the articulations.

The rliizomas, flatteneti by compression to about two centimetres in
witlth, are regularly cylindrical, slightly narrowed only at the articulations ;
these are distantly marked around by circular scars at the point of attach-
ment of oval or obovate tul)ercles, averaging two centimeters in length
and one and a half centimeters broad; the rhizomas are distinctly irregularly
striate, the tubercles obscurely ribbed in the lengtli, either simple and joined
to each other by their ends, like a string of beads, or double, two of them
being attached to the inflated end of a single one, as on the right side of tig. 3.
At a distance from the point of connection to the rhizomas, and as seen from
the numerous fragments of this species covering large slabs, the tubercles
become more and more elongated, and pass to mere cylindrical filaments, or
rootlets, which appear to divide into radicles. The point of union of the
tubercles, either to each other or to the rhizomas, is marked by comparatively
large scars, three to four millimeters wide, representing a double ring with
a central point. Fig. 4 represents one of these tubercles split lengthwise,
and exposing in the middle a central solid axis, one and a half millimeters
thick, while the parietes or intervals from the axis to the borders, four milli-
meters each side, appear formed of a spongy though compact medullar tissue,
becoming more compact or darker-colored near the borders. A vertical
cross-section of another tubercle from Carbon specimens, like the former,
shows it to be oval or somewhat flattened by compression, twelve millimeters
in one direction and only nine in the other. Still another specimen of the
same locality represents a linear rootlet, or rhizoma, whose main axis, four
millimeters wide and central, appears surrounded by a cylinder of cellular
tissue of equal thickness. This branch is marked by distant nodi and round
scars, the same as those of Barrell's Springs, from which the species was first
described. This species is comparable to Equisetum arcticum, Heer (Spitz.
Flor., p. 31, pi. 1, figs. 1-15), which has narrower rhizomas, with nearly as
large tubercles, more elongated and narrower at their point of attachment,
and also to Physngenia {Equisetum) Farlatorii, Ung. (Sillog., p. 4, plate 1,

68 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA,

figs. 5, 6), which bears much smaller, more regular tubercles, nearly exactly
oval. In both these species, the articulations are inflated*; in ours, tliey are
strangled, as seen in the figures.

Habitat.— Barren's Springs {Dr. F. V. Hmjden), Carbon Station, Wyom-
ing. The specimens of this locality are too fragmentary for positively
demonstrating specific identity.

Equisetuni Isevigatuni, Lesqz.
Plato VI, Figs. G, 7.
Equiaetumf Jceingatum, Lesqx., Aunual Report, 1873, p. 395.

Stem or rhizonia thick, distantly and obscurely articulated, irregularly wrinkled in the length;
articulations marked by round scars, distant from each other.

The only specimens obtained of this species are figured here. They
seem to represent remains of a large species of Equisetum by the indistinct
articulations and the scars perceivable around them. The large specimen
(fig. 7) is only part of a stem or rhizoma divided in its length; the articula-
tion is marked merely by two scars, round in the center, with a row of oval
impressions placed starlike around it, as it is sometimes the case in large fossil
species of Equisetum. It is narrowed above the articulations, and appears
enlarged at the nodi, but probably by a compression of the stem, more
marked there than it is above ; the surface of the specimen is smooth, merely
impressed by irregular wrinkles, not true equal striae. The other smaller
specimen (fig. 6) has its articulation marked on one side by a round scar, and
on the other by the outside of a scar of the same form, to which is attached
an articulated branch, flattened in the middle, then divided in two at its end,
either by normal separation or by splitting from mechanical compression.
This part resembles somewhat fig. 2 of the same plate, for the branch repre-
sents apparently less inflated tubercles, or the division of a rhizoma slightly
strangled at the articulations and disfigured by compression. I tlieretbre
consider these fragments as representing a species or perhaps two diffcront
species of Equisetum, the one with a much larger stem and the scars marked
by an outside row of oval impressions ; the other smaller, dilTering from
Equisetum Hmjdenii, especially by its smooth surface, the less distinct articu-
lations, with more distant round scars, and the irregular form of the tubercles.
Characters of this kind, and too indefinite, cannot be considered as specific,
perhaps. They are to be carefully considered, however, as the specimens
belong to a stage of the Tertiary different from those of the former species.

Habitat. — The large specimen (fig. 7) is from Sand Creek, Colorado,
eight feet above coal ( W. II. Holmes); the small one from Golden {A. Lakes)

DESCRirTIOX OF SPECIES— EQUISETACEJ3. 69

E q M I s e t u m W }* o in i ii d; c ii s c , Lesqx.

Plato VI, Figs. 8-11.

Equiseium Wi/omiiigfmc, Lpeqx., Auuual Report, 1873, p. 409.

Stems or rliizoinas equally distiuctly slriato, articulate; artieiilationH Hliort ; slieatlis acutely den-
tate; radicles in bundles from the articulations.

The specimens are large slabs of very hard white shale, covered with a
profusion of fragments of the same plants, rootlets, rliizomas, stems cruslied,
pressed together, and rarely separated distinctly enough to clearly show their
characters. The best ones are figured. They represent fragments of stems,
or of rhizomas, articulated, slightly narrowed at the articulations, which are
close, either naked or marked by round small scars, and bearing bundles of
simple rootlets, diverging starlike ; all the fragments of stems and rhizomas
are equally and regularly striate; tlie sheaths, as seen in figs. 8 and 9, equally
dentate on the borders, and short; fig. 11 is apparently an inflated articula-
tion of a rhizoma or tubercle, bearing at one end a compound branch or root-
let, with a whorl of rootlets in the middle; these rootlets are all of the same
size, less than one millimeter thick, linear or filiform, simple or forking, as
seen on the lateral branch of fig. 8. The stems and rhizomas, also of equal
size, average half a centimeter in width.

Habitat. — Three miles east of Green River Station, Wyoming Territory.

E4iui»>etuiii liiuosuiii!, Linn.

Plate VI, Fig. 5.

Squiselum Untosum, Linn., Annual Report, 1871, p. 299.

Stem simple, deeply striate, articulate ; articulations short ; sheaths short, acutely dentate.

Tiiough it is not possible to prove identity from a mere fragment of
Equisetum representing only part of a stem and its sheaths, it is also impossi-
ble to find a point of difference in comparing this stem with that of the living
Equiseium limosum. As in this last species, the stem is ccjually ten-ribbed
or deeply striate, tlic sheaths of the same length, blackened also in the upper
border, with narrow, acutely pointed, equal teeth, appressed upon the stem.
It is evidently different from the former species by the more numerous and
narrower teeth of the sheath.

Habitat. — Near Yellowstone Lake, among basaltic rocks. The geo-
logical age of the formation is not indicated with the specimens; it may be
some recent deposit by hot springs or volcanic agency {Dr. F. V. Hayden).

70 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

PIUENOGAM^.

GYMNOSPERMtE.
CYCADINEiE.

The two essential groups of this family, the Zamiea and tlie Cycadeee,
have at our time a comparatively small number of species distributed in the
tropical and subtropical regions of the world. The genus Zamia is mostly
American, its species inhabiting Mexico, the West Indian Islands, and the
mountains of the Pacific coast; one species is found in Florida. Of Cycas,
no species is indigenous upon the American continent.

In the old geological formation, the Cycadinece were very numerous,
especially most abundant in the Trias and the Oolite, which, taken both
together, have more than one hundred and sixty species of this order, out of
about two hundred, constituting the whole flora of this formation, as known
by authors. A number of vegetable remains from the Devonian and the Car-
boniferous, leaves known under the generic name of Flabellaria, Cordaites,
Nocggerathia; and fruits, Trigonocarims, Rhahdocarpus, Cardiocarpus, etc., are
until now of uncertain relation, some of their characters referring these plants
to this family, others to the Conifers. Very few CycadinecR are known in forma-
tions above the Jurassic. One, Pteroplyllum, is described from the Cretaceous
of Kansas, and another, closely allied, from that of Germany. This formation
has also two Strobiles described as Zamiostrobus; a third is known from the
Miocene.

ZAMIE^.

ZAMIOSTROBUS, Endl.
Zamiostrobus! mirabilis, Lesqz.

Plate LXIII, Figs. 1-1 d.

ZamiostroiHS mirabilis, Lesqx., Annual Report, 1874, p. 309.

Fragment of a large silicified cone, checkered upon its onter surface by the rhomboidal obtuse top
of quadrangular long seeds, black-colored, and apparently of a hard substance, embedded into a white
celluloso-vascular matrix, separated from the common cylindrical axis by a zone of the same matter, or
fixed upon it by their base.

The specimen represents a fragment only of the cross-section of a large
cone, measuring about fourteen centimeters in diameter, and perfectly silici-
fied. As seen in fig. 1, it is composed of a cylindrical axis, c, made up of frag-
ments of a dark opaque matter, agglutinated and amorphous, to which are

DESGKIPTION OF SPECIES— CONIFERiE. 71

attached fruits or seeds apparently of hard consistence, represented as tl)cy
are by a black compact silex, and pierced in the length by pores, or ducts,
continuous from the top to the base. These seeds, three to three and a half
centimeters long, six to eight millimeters broad, are quadrangular, somewhat
flattened, and therefore transversely rhomboidal, as seen from their cross-
section in tig. 1. From the outer surface, where the angles are rounded or
more obtuse, they pass down, through a white celluloso-vascular substance
wherein they are imbedded, to the axis c. In fig. 1 h, which represents part
of the reverse of the same specimen, the seeds appear shorter and separated
from the axis c by a zone, b, of the same whitish substance as that which
surrounds them. The intervals are as wide as the space filled by the seeds.
This white matter appears, when seen with a strong glass, composed from
very thin, parallel, linear filaments descending from the surface to the axis.

Though I do not know any fruit presenting an evident likeness to this
remarkable fragment, I believe it referable to some species of Zaviiett. By
the form and disposition of its surface-scars, it is comparable to species of
An droit robus, a genus established by Schimper for the description of cylindrical
Cycadeous male cones formed of imbricated scales, bearing sessile anthers upon
their lower surface. By the position of the seeds, and also by their form and
size, it has a distant relation to Zamiostrobus gibbus, Reuss, represented by
a cone, which shows, in its section, oblong, obtuse seeds placed at right angles
to a cylindrical axis, with the tops appearing at the outside surface. Both
these cones are figured in the Atlas of Schimper's Pal. Vcget., pi. Ixxii,
figs. 1, 2, 14, 15. There is, however, a great difference in the large size and
in the characters of this silicified Strobile from the species mentioned above.

Habitat. — Near Golden, Colorado, found by Dr. F. V. Hayden on the
surface soil, without connection to any stratum of rock. This, however,
does not, positively at least, force its reference to an antecedent formation, as
silicified trunks, especially of Palms, and rounded fragments of the same
kind, are abundantly found around Golden and in Colorado, together with
petrified and silicified trunks of trees still standing and in place. The
fragment of this cone has been apparently detached and drifted from the same
formation.

CONIFERiE.

The Conifers appear already in the Devonian, where remains of large
trunks have been found and referred from the characters of their internal

72 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

structure to species of this family. Fossilized trunks of Conifers have
been found also in the Subcarboniferous of England. In the true North
American Carboniferous, we have as yet scarcely any trace of Conifers ; at
least, the remains referred to them, those of the genus Arfisia, for example,
are still of uncertain relation. If the Cordaites of the coal have apparently
characters which relate them to Cycadece, the species of Noeggeratkia, of
the same formation, might as legitimately be referred to Conifers. This ques-
tion is still in a state of uncertainty. Immediately above the Coal-Measures,
even in tlieir higher strata, and in the Lower Permian, the Conifers appear
in a fair proportion by two new genera, Walchia and Volzia, which persist
into the Trias, with a few Cypressinece. The Jurassic has genera of the
Araucarice, the Ahietinea:, and the TaxodicE essentially predominant, as in the
Cretaceous also. But the larges.t number of Conifers is known in the Ter-
tiary, wherefrom two bundled and twenty-five species or more are described
until now. This does not positively indicate what may be the distribution in
the other formations, whose flora, comparatively to that of the Tertiary, is
scarcely known, except that of the Carboniferous, however, from which four-
teen species of Conifers are described in Europe from fossilized wood. Of the
genera still present in the flora of our epoch, Arnucaria appears in the Trias,
is more abundant in the Jurassic, and absent from the flora of the Miocene,
now ^inhabits the countries south of the equator, especially New Holland.
Sequoia, Cretaceous and Miocene, is left in the present flora by the two well-
known species of California, S. semiyervii-ens and S. gigantea. Finns, with a
few representatives in the Cretaceous, becomes extremely predominant in the
Miocene, from which more than one hundred species are described ; while
Larix, Cedrus, Abies, and Taxodium, which appear at first in the Tertiary,
continue to the present time.

CUPRESSINE^.

WIDDRINGTONIA, EndL
Widdrins^^tonia! coniplanata, Lesqz.

Plate LXII, Figs. 13, 14.

Widdringtonia complanata, Leaqx., Aunaal Report, 1874, p. 299.

Stem thick, disticho-pinnato; braucblets short, thick, obtuse, alternate, obliiiue; leaveB small,
apparently in spiral order, closely imbricate and appressed, oblong, lingulate-pointed upon the primary
branches, ovate-pointed or rhomboidal and short upon the brauchlcts.

■ This species appears to have been of a soft though thickish substance.
All the specimens are flattened, even the branches, and, in that way, the form

DESClilPTlON OF SPECIES— CONIFERJE—CUPEESSINE^. 73

and the relative position of iho leaves are not satisflictorily distinguishable.
The primary branehes, more or less tlexiious, two to three millimeters thiek,
are long, diverging in a more or less acute angle from the main axis, with
their leaves distant, linear or Ungulate, abruptly pointed; the first divisions,
distichous, turned upward, are variable in length, some of them regularly
pinnately sulxlivided in simple, short, obtuse branchlets, otiiers elongated and
simple from tlie middle. The leaves of the divisions are siiort and small, one
to two millimeters long and al)<)ut lialt" as broad, closely iml)ricatc(i and ap-
pressed upon each other, the upper outlines only being discernible, ajiparently
placed in spiral order, a disposition, however, remarked upon a very small
specimen (fig. 13), the only one where the thin coat of coaly matter is pre-
served, but not distinctly seen upon the large specimen (fig. 14), which is
apparently a counterpart or impression, whereupon the leaves are marked only
by black points, or short lines whose relative position is irregular. The spe-
cies is related to Widdringtonia antiqua. Sap. (Et., 2, 1, p. GJ, ])I. 1, fig. 4),
by the form of the leaves, which, however, arc more closely appressed in our
species, and more distinctly in spiral order arouixl the branchlets. The
celebrated French author considers this American Conifer as very interesting
and remarkable, representing either a new type or an Artlirutaxis, or even a
Dacridium. Its generic relation is therefore indefinite.

Habitat. — Point of Rocks, Wyoming Territory {Dr. F. V. Hayden,
Will. Cle.hurn); found in numerous but small specimens. Tlie brancii in
fig. 14 is the largest; the others have only secondary divisions and branchlets,

as in fig. 13.

TAXODIUM, Rich.

Taxodium distich iini niioceniciim, Heer.

Plate VI, Figs. 12-14 a.

Taxodium duhium, Heer, Lesqx., Anunal Report, 1S72, p. 389 ; 1673, p. 409.

Bniuclilets caducous, liliform ; leaves distaut or soinotiuies two close together, altoruate, distich-
ous, uarrowed to the very short-petioled base, liDcar, obtuse, or obtusely pointed, flat, with a distinct mid-
dle nerve.

This species, described by Ileer in the Arctic Flora as T. duhium, and
separated in the Baltic Flora under the present name, is represented in our
Tertiary flora by numerous fragments, branchlets only. The cones described
and figured I>y the author as subglobose, with large scales marked in the
middle by a transverse costa, umbonate at the center and verruco.se, have not
yet been found in our Tertiary measures. The leaves greatly vary in size,

74 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

sometimes being very small and narrow, as in fig. 12, sometimes compara-
tively broad, as in fig. 13, with a distinct middle nerve. Fig. 14 represents a
branch with narrow leaves, and could be perhaps referred to T. angusti-
foUum, Heer (Arct. FL, p. 156, pi. xxx, fig. 1). But the lower part of fig.
13 is evidently of the same character, and therefore seems to be a form of
this variable species. The American specimens, like those described and
figured in the Arctic Flora, have generally linear, slightly obtuse leaves,
while those of the Baltic Flora have them slightly lanceolate. Fig. 14 a is
an enlarged representation of fig. 14. The leaves vary in length from four to
eleven millimeters, and in width from one-half to more than one and one-
half millimeters.

Habitat. — Elko Station, Utah {Prof. E. D. Cope), in numerous very
fragmentary specimens. I found it also abundantly at Carbon, and it was
collected in profusion at Fort Fetterman by Lieut. Vogdes. We have one

specimen from Evanston.

GLYPTOSTROBUS, Endl.

Olyptostrobus Europteus, Heer.
Plate VII, Figs. 1, 2.

Glyptostrobus Europceus, Heer, FI. Tert. Helv., i, p. 51, pi. ix, xx, fij;. 1 ; Fl. Foss. Arct., p. 90, pi. iii, fige.

2-5, pi. xlv, figs. 20-22; Mioc. Bait. Fl., p. 20, pi. iii, figs. 8, 9, etc.— Lesqx., Annual Eeport,

1873, p. 409.
Glyptostrobus Ungeri, Heer, Fl. Tert. Helv., i, p 52, pi. xviii.— Ludw., Palseont., viii, p. 71, pi. xii, fig. 6.
Glypiostrobites (Enhigmsis, Al. Br. Stizen. Verz., p. 73.

Branches irregularly pinnately divided ; branchlets variable in length, obtuse ; leaves squamiform,
imbricated, decorreut at base; male catkins ovate, single, sessile upon lateral branches; fertile catkins
ovate, comparatively large; scales imbricated, semicircular, obtusely dentate on the borders, costate in
the length ; seeds ovate, with alate borders.

The fertile catkins of this species have not been found as yet in the
American specimens communicated to me, but only fragments of branches
and branchlets, like those figured here, some of them bearing male catkins,
The description of the fertile cone is made from European authors. The
shale bearing our specimens is, however, covered with remains of this species,
branchlets especially, together with, deep, oval impressions, which seem to
represent crushed cones, whose scales are deformed and undeterminable.

Habitat. — On thin soft shale, near Castello's Eanch {Dr. F. V. Hayden);
near Florissant {Prof. E. D. Cope).

DESCRIPTION OF SrECJES— CONIFERS— ABIETINE^. 75

ABIETINEiE.

SEGUOIA, Torr.

Sequoia ariiuis, Lesqz.

Plate VII, Figs. 3-5; Plate LXV, Figs. 1-4.

Sequoia affinis, Lesqx., Annual Report, 1874, p. 310.

Branches long, slender, pinnately irregularly divided ; leaves short, oblong, or Ungulate, obtusely
pointed, imbricated, erect, or appresstd; br.inclilets bearing fertile cones, open ; strobiles small, oval,
obtuse; scales large, rhomboidal, with entire borders, a central oval mamilla and wrinkles passing from
it to the borders all around ; branches bearing male calkins erect, with slightly more acute and longer
leaves; catkins small, broadly oval, obtuse, covered to the top with imbricated, liugulate, pointed scales.

We have of this species numerous and well-preserved specimens, some
of which are figured in pi. Ixv, figs. 1-4. The divisions are generally pinnate,
sometimes fastigiato-pinnate ; the leaves linguiform, more or less abruptly
narrowed to a point, decurrent, without apparent trace of a middle nerve.
In the fragments represented in pi. vii, figs. 3-5, the leaves are more obtuse,
as also the scales of the male catkins (fig. 5). These fragments may repre-
sent a different species, but they are too imperfect for exact specification.
They may be even referable as variety to Glyptostrobus Europceus, being
upon the same pieces of soft shale as those of figs. 1 and 2 of the same plate
As represented in pi. Ixv, Sequoia affinis has a remarkable affinity to <S.
Coutsicp, Heer (Bovey-Tracey, Mioc. Fl., p. 10.51, pi. lix, Ix, Ixi); still more
to S. Tournalii, Brgt., as figured in Sap., fit., ii, p. 2, pi. ii, fig. 1, and may
be considered as an American form of this type. It merely diflbrs from both
species by the more obtuse point of the fertile, scale-like leaves, which are
also longer and more gradually pointed upon the sterile branches ; by the
slender, longer branchlets, bearing the cones at their end only; by the more
distinctly oval form of the slightly smaller strobiles, and by the absence of a
middle nerve on the back of the convex or inflated leaves. The seeds, if
the one figured in pi. Ixv (figs. 4 and 4a enlarged) belongs to this species,
are smaller, distinctly cordate, and obtusely pointed, bearing near tlie point
a mere trace of a line, whicli divides, diverging on both sides toward the
borders. In S. Coutsice, the seeds are oval-cordate, with a small, inflated,
lanceolate-pointed nucleus, surrounded by a narrow wing. We have seen,
however, only one of these seeds, and, though upon the same specimen as
the branches of this species, its position docs not positively indicate a relation
to it. Moreover, those small bodies present varied forms, according to their
position in I lie matrix wherein they are imbedded. One scale, enlarged,
shows the wrinkles of the surface (fig. la); its borders are more inflated and

76 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

rugose than represented in those of the European species; fig. 3a is an
enlarged fragment of a branch, with male catkins and leaves.

Habitat. — Castello's Ranch {Dr. F. V. Ilai/den), specimens figured in
pi. I.w ; Elko Station {Prof. E. D. Cope)., as represented in pi. vi, figs. 3-5.
These last fragments were doubtfully referred to Sequoia Langsdorffii, Brgt.,
in Report, 1874, p. 410.

The two cones (figs. 5 and 5a), which, flattened by compression, merely
show displaced apophyses somewliat similar to those of this species, are not
positively determined. The scales are distant, -seemingly scattered, more
enlarged on both sides, bearing a central mamilla, from which numerous wrin-
kles, or strise, are diverging around to the borders. They resemble the scales
of Geinitzia formosn, Heer, of the Cretaceous flora of Quedlinburg ; a distant
likeness, however. The relation of these cones is still uncertain. They come
from the same group, the upper division of No. 4.

Sequoia L.ang:sdorffii, Brgt.

Leaves close together, rigid, coriaceous, liuear, obtusely pointed, flat, open, distichous, narrowed
and decurrent at base ; middle nerve strong ; cones broadly oval or subglobose ; scales numerous, peltate,
mucronate in the middle.

Though I have seen, upon specimens from different localities, especially
from Black Butle, a number of detached leaves, apparently referable to this
species, I have been unable as yet to find a fragment distinct enough for
positive identification. It has been described by Dr. Newberry from numer-
ous specimens obtained by Dr. Hayden from the Yellowstone Lignitic.
Heer has it also described from very fine specimens, from the roof of the
coal mines of Nanaimo, Vancouver's Island, as well as from Greenland and
from Alaska. It is represented in Prof J. D. Whitney's collection of Mio-
cene plants of Oregon by specimens from John Day Valley, bearing branches
and cones, in connection w'xih. Betula macrophylla, Giipp., Quercus pseudo-nlnus,
Ett., Q. furcinervis, Heer, Platanus nobilis, Newby., Cinnamoinum lanceolatum,
Ung., Acer, etc.; therefore its presence in the Colorado and Wyoming
Lignitic may be admitted as more than probable. Even, as it has been
suggested by critical remarks, the two forms described here below, under
diflTerent specific names, may be mere varieties of it. In Europe, Sequoia
Langadorjfii is one of the more generally distributed plants of the Miocene,
and has been described, from numerous localities and under different names,
by most of the paleo-piiytologists : first J^axites Longsdorffii, Brgt., 0. Webb.;
it is T. pidegetonteus, T. Kosthorni, and Cupressiles taxijormis, Ung., Cupres-

DESCRIPTION OF SrECIES— CONIFERJ^— ABIETINE^. 77

sites Hardlii, Goei)p., Chamcenjparites, Eiull., Sequoia ? Senogalliensis, Mass.,
Steinhauera minuta, Sternb.; even Taxodium duUum and T. laxum, Ett. Its
affinity to plants of the present lloia is witli Sequoia sempervire7}s, Endl., llie
Redwood of California.

Habitat. — Black Buttes ; Haley coal mines, with S. hngij'olia; Floris-
sant {Dr. F. V. Harjden).

Sequoia aiigusitifolia, Lesqx.
Plate VII, Figs. G-10.
Sequoia angmiifoUa, Lesqx., Annual Keport, 1872, p. 372 ; 1873, p. 409.

Brancblets short, slender ; leaves at unequal distances, sometimes very close, two to tbreo together
or very distant, often dimorphous, linear-lanceolate, taper-pointed, open or curved backward, dccurreiit ;
middle nerve indistinct.

This form, though very variable, preserves its peculiar characters : the
narrow, lanceolate, acute leaves, decurrent but not narrowed at base, with a
thin, scarcely distinguishable, middle nerve. These leaves arc not evidently
distichous, but sometimes placed all around the branches. The cone is as yet
unknown, but I refer to it the seed in tig. 10, which is oval, obtuse, and nar-
rowly margined or winged. From tlie former species, to which it is evi-
dently closely related, it difiers by the consistence of the leaves, which are
not as thick and not coriaceous; by their shape, narrower and more acutely
pointed, not narrowed at the decurrent base ; by the scarcely visible middle
nerve. The similarity of this form to that described and figured by d'Ettings-
hausen in Bil. Fi., p. 34, pi. xii, figs. 1-13, as Taxodium diibium and T. laxum, is
indeed striking, and as all these fragments have been identified with Sequoia
Langsdorffii, our species should be, therefore, and has been, considered as a
variety of this. But even here there is a marked difference in the form and
position of the leaves, which are represented by the European author as all
narrowed at the base and generally distichous, and another still in the seeds,
which, though of the same form, are twice as large in the American species.

Habitat.— Elko Station, Utah Territory {Profs. S. W. Garman, E. D.

Cope).

Sequoia Ileerii, Lesqx.

Plate VII, Figs. 11-13.

Seqitoia Eeerii, Lesqx., Annual Report, 1871, p. 290.

Brancblets slender ; leaves short aud narrow, distant, linear, obtuse or obtusely pointed, narrowed
at the base and decurrent ; middle nerve thin ; strobile borne upon a naked pedicel, round, flattened at
the top ; scales peltate, rhomboidal, obtuse in the upper and lower side.

Except that the leaves are generally obtuse and narrowed to the i)oint
of attachment, and more distant, this species is similar- to the former. The

78 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

leaves vary from three to seven millimeters long, averaging a little more than
one millimeter in width, generally linear or slightly enlarged toward the
point, of the same consistence as those of the former species, neither rigid
nor coriaceous, with a thin middle nerve. As seen in fig. 11a enlarged, they
are evidently decurring upon slender branchlets, none as thick as any of those
described and figured of Sequoia Langsdorffii. The cone, broadly globular
and flattened at the top, is larger than that of this last species, figured in Fl.
Arct., pi. ii, fig. 2, and about of the same form, and with few scales, but
smaller than that of pi. xlv, fig. 14, of the same work. Therefore the
difference in size is of no specific value, and also the total destruction of the
scales or leaflets upon the pedicel may be considered as resulting from
maceration. We have therefore merely, for specific character of this, the form,
the size of the leaves, and their distant, irregular position upon the slender
branches. Sequoia Langsdorffii, as figured in Flor. Alask. (pi. i, fig. 10),
and S. NordenskiUldi, Heer (Flor. Spitz., pi. iv, figs. 4, 34), have the
leaves linear-obtuse, but they are rigid, coriaceous, close to each other, longer
in the last species, twice as large in the former, with a thick middle nerve
in both. Numerous specimens, all from the same locality, represent frag-
ments or branchlets and cones with the same characters as those described
here.

Habitat. — Sage Creek, Montana {Dr. F. V. Hay den), in connection
with undeterminable fragments of Ferns and some leaves of Ilex dissi?nilis,

Lesqx,

Sequoia brevifolia, Heer.

Plate LXl, Figs. 25-27.

Sequoia hrmifolia, Heer, Fl. Arct., p. 93, pi. ii, fig. 23; Mioc. Bait. Fl., p. 21, pi. iii,fig. 10; ix, fig. 5 c—
Lesqx., Annual Eeport, 1874, p. 298.

Branches somewhat thick, flexaoas, pinnately divided; branchlets opposite, more rarely alternate,
open toward the base, and diverging, curving upward and erect from the middle; leaves biform, either
small, short, scaliform. Ungulate, pointed, at the base of the branchlets or covering them when young,
or longer, linear-oblong, enlarged in the middle, abruptly narrowed to a point, rarely obtuse, distichous,
close to each other, obliqne, gradually decreasing in length toward the top and the base of the branch-
lets; middle nerve distinct.

Besides the branch figured here, a large number of more fragmentary
specimens, mostly branchlets, like fig. 26, have been sent from the same
locality. This fine species is described by Heer (loc. cit.) from Greenland
and from Spitzbergen specimens, and in the Baltic Miocene Flora from Kran-
tepcllen. North Germany. It is distinctly characterized by the form of its
generally short, distichous leaves, abruptly pointed, sometimes slightly obtuse,

DESCRIPTION OF SPECIES— CONIFERS— ABIETINE^. 79

and generally enlarged above the middle and gradually narrowed toward
the decurring base. We have, however, among the specimens a number
of them bearing branches with narrower, nearly linear, closely approached,
somewhat longer leaves, as seen in fig. 27, which show a notable deviation
of the normal form, the leaves being on a more acute angle of divergence,
parallel, and not turned backward. Between this and the normal form, the
specimens show intermediate varieties, which recall them all lo the more
generally represented type, that of fig. 25. The leaves are seen enlarged
in figs. 25 a and 27 a. The cone of this species is not yet known. One
of the specimens from Point of Rocks, covered with scattered branchlets
and leaves of this Sequoia, has a crushed cone (pi. Ixi, fig. 30), which appears
to be a flattened cross-section, or perhaps the flattened base of a strobile
turned upward, the broken pedicel marking the central point around which
the scales are imbricated from the center to the borders. The scales, as far
as they can be discerned, are oblong, cuneate, narrow, emarginate, or irregu-
larly crenate along the borders. They rather represent scales of a cone of
Glyptostrohus than those of a Sequoia. The relation of this cone is therefore

uncertain.

Habitat.— Point of Rocks, Wyoming Territory {Dr. F. V. Hayden,
Wm. Cleburn). Nearly one-half of the specimens of the collection of this
last contributor represent this species in its various forms. The fine spe-
cimen in fig. 25 was kindly communicated by Mr. E. H. Clarke, agent of the
Point of Rocks station.

Sequoin I on g ■ f o lia, Lesqz.

Plate VII, Figs. 14, 14 a; Plate LXI, Figs. 28,29.

Sequoia longifolia, Lesqx., Annual Report, 1874, p. 298.

Branches and branchlets thick ; leaves close, open, slightly recurved or erect, long, linear-lanceo-
late, acuminate, slightly narrowed to the decurring base, thick ; scars deep, Ungulate, obtusely pointed,
marked by a deep groove in the middle.

As figured in pi. vii, fig. 14, the leaves arc lanceolate, slightly narrowed
to the base, less, however, than represented in the enlarged figure (14 a),
gradually acuminate, with a broad, though indistinct middle nerve, and sur-
face regularly thinly striate. The specimens in pi. Ixi, figs. 28, 29, have
the leaves much longer and narrower, linear, and also gradually narrowed
to an acumen; their consistence is still thicker; they are more closely
appressed and more numerous, forming by compression a coating of coaly

80 UNITED STATES GEOLOGICAL SURV EY— TERTIARY FLOKA.

matter where they become mixed together and somewhat indistinct. The
surface is less evidently and less regularly striate than in the other form, and
they do not appear narrowed to their point of attachment. It may be, there-
fore, that these specimens, the first from Haley coal mine, the others from
Point of Rocks, represent two different species. The leaves of figs. 28 and
29 are not distinctly seen to their base or to the point of attachment to the
stems. The indistinctness of the specimens may account for the differences
remarked above.

Habitat. — Haley coal mine, ten miles northeast of Greeley, Colorado
{A. a Peak); Point of Rocks {Dr. F. V. Hayden).

Sequoia acuminata, sp. uov.
Plato VII, Figs. 15-10 a.

Branches thick, narrowly striate ; leaves thick, rigid, with a thick, distinct, middle nerve, linear-
lanceolate, gradually acaminate, narrowed to the decurrent base ; surface smooth.

This species differs from the former by the proportionally narrower
leaves, with a very distinct middle nerve, and smooth surface; also, by the
stem, which is striate when decorticated, as in fig. 15, or naked and without
scars, fig. 16. These differences may not be considered important enough
to authorize specific distinction. The general appearance of the fragments
representing both forms is, however, different; for the leaves of those ascribed
to this species have a smooth and polished surface, appearing more rigid than
those of the former, though their substance is not as thick. The average
size of the leaves is about the same in both forms, the leaves varying from
three to six centimeters long, and from two to four millimeters broad. These
two species have a remarkable affinity to Torreya Californica, Tor., by the
form, the consistence, and the disposition of their leaves. They, however,
differ from Torreya by the evidently decurring base slightly narrowed, while
in this genus they are rounded at the base to a short petiole.

Habitat. — Black Butte, in the black shale of a lower coal than the main
coal, opened at a short distance north of the station. No other vegetable re-
mains were found in connection with these.

Sequoia bi form is, Lesqz.

Plato LXII, Figs. 15-18 a.
Sequoia biformia, Lesqx., Annual Beport, 1874, p. 298.

Stems or branches thick, irregularly pinnatcly divided; branehlets short, obliquely diverging;
leaves of two kinds, either longer, linear, obtnsely pointed, or shorter and broader, lanceolate, taper-
pointed, and slightly, gradually narrowed to the decurrent base, generally incurved falcate, either dis-
tant and irregularly placed or crowded and imbricated ; stems distinctly marked by triangular or
lingulate-pointed scars.

This species bears two kinds of leaves, even upon the same fragments

DESCRIPTION OF SPECIES— CONIFEE^—ABIETINE^. 81

of branches ; either long, two centimeters, very narrow, linear, less than one
millimeter broad, or shorter, broader, generally more crowded, lanceolate,
taper-pointed, somewhat eidarged in the middle, and gradnally slightly nar-
rowed to the decurrent base, more than one millimeter broad and only eight
to ten millimeters long; the middle nerve is clearly and dee|)ly marked upon
both kinds of leaves. In fig. 15, the leaves are short, much crowded, imbri-
cate and falcate; fig 17 has very narrow leaves of various lengths, straight,
or llexuous, or slightly falcate; figs. 17 and 18 have both kinds of leaves;
some very short and narrow, some broader and longer. Besides the specimens
with leaves variable in size, as pointed out in the figures, then; is a large
number of others where variations of torra may be clearly seen, and, there-
fore, all evidently represent the same species. By the falcate form of the
shorter leaves, it has some likeness to Sequoia Reiclienhachi, Heer, of the
Cretaceous.

Habitat. — Point of Rocks {Dr F. V. Hayden) ; found in numerous speci-
mens. It appears, however, locally distributed, as the collection of Mr. Cle-
burn made at the same locality has not any specimens representing it.

ABIETITES, Gocpp.
Abietites dubius, Lesqx.
Plate VII, Figs. 19-24.
Abkiites dubius, Lesqx., Annaal Report, 1869, p. 196 ; 1872, p. 374.

Branches thick ; leaves open (those of the branchlets erect ), loosely imbricated around the branches,
lanceolate, gradnally narrowing to a sharp point ; broadest at the base, where they are abruptly rounded
to the point of attachment.

This species, found first at the Raton Mountains by Dr. J. Le Conte and
later by myself at the same locality in numerous specimens, has been also
sent from different places, all the specimens presenting the same charac-
ters. The branch leaves, loosely imbricated, as seen in fig. 19, are generally
n.ore open, slightly narrower; those of the branchlets more erect and close;
all exactly lanceolate, gradually narrowed to an acute point, abruptly rounded
to the point of attachment, flat, concave inside, and marked by a thin,
somewhat indistinct middle nerve ; the stems bear distinct leaf-scars, pre-
senting, as seen in figs. 19, 22, 23, 24, various forms, according to their age,
their size, and tlie different stages of maceration and compression. The
length of the leaves varies from seven to thirteen millimeters, the width from
a little less to a little more than one millimeter; the substance is not coria-
ceous, and they appear easily destroyed by maceration, the branches and
6 T F

82 UNITED STATES GEOLOGICAL SURVEY— TERTIAEY FLORA.

branchlets found being very often divested of them, or with mere fragments,
as seen in fig. 22. The cones of this species are as yet unknown.

Habitat. — Raton Mountains, New Mexico, where its remains are very
abundant {Dr. J. Le Conte). I found there mixed with them large pieces of
bark, covered with large oval-obtuse tubercles, placed in rows; they seem
referable to this species. Specimens in a good state of preservation were
sent also from Fort Ellis by T. Savage, with mostly undeterminable fragments
of dicotyledonous leaves. Frof. B. F. Meek obtained also, from chocolate clay
shale underlying a bed of coal near Fort Steele, fragments of stems bearing
scars similar in form and disposition to those of the branches of the Raton,
These specimens, however, were without any leaves, the only other discerni-
ble remains, figured in pi. Ix, fig. 37, being a fruit or scale of a cone, appar-
ently referable to Nordenskioldia borealis?, Heer.

Abietites setfger, Lesqz.

Plate VII, Figs. 17, 18.

Abietites setiger, Lesqx., Auuual Report, 1872, p. 404.

Leaves distant, very narrow, needle-form, placed in right angle to and around the branches, or
curving backward and reflexed from near the point of attachment.

None of the fragments representing this species are better than those
which we have figured. The leaves, eighteen millimeters long, less than one
millimeter broad, are exactly linear, filiform, grooved and nerved in the mid-
dle, abruptly pointed, and slightly enlarged at the point of attachment, distant,
and, as seen from the scars upon the branches and placed all around them,
either diverging in right angle or curved downward from near the base.
Both this and the former species are without relation to any species of fossil
Conifers known as yet from this country. But Count Saporta writes me that
he is surprised to find that two forms apparently specifically identical with this
and the former species are found in the Upper Cretaceous of France, that is, in
the Lignite formations of Saint Paulet, Gard. He says: — "I have a specimen
from Saint Paulet, which is like your figs. 17 and 18, and I have another,
obtained from Brongniart, which is undoubtedly identical with your figs. 19 to
24." He adds that the horizon of the first species, Abietites duhius, ought to be
the same as that of -4. setiger; that Sequoia hiformis is intermediate between
them, and seems to unite both these so different forms, as possibly representing
the same species. To this last supposition, I remark, that each of these three
species, es^ec\&\\y Sequoia hiformis and Abietites duhius, is represented by a large

DESCRIPTION OF SrECIES— CONIFERiE— ABIETINEiE 83

number of specimens, all with the same characters; tliat Ahietites duhius can-
not be referable to Sequoia, on account of the rounded and not decurrent base
of the leaves, and that also AbietUes setiger, with its puncti(()rm scars of
leaves, is, like the other species, forcibly separated from Sequoia.

Habitat. — Six miles above Spring Canon, near Fort Ellis (Dr. A. C.
Peak). The specimens from Fort Ellis, obtained by Mr. Savage, have no
fragments of Ihis species, but have some of Ahietites duhius. l\'r contra,
those collected by Dr. Peale have none of A. dubius, but have also dicoty-
ledonous leaves indiiferently preserved, among which Quercus Pealei, Lesqx.,
Rhamnus acuminata, Web., and Gymnogramma Haydenii, are recognizable.

PINUS, L.

Pin us palseostrobus!, Ett

Plate VII, Figs. 25,31.

Pinites paloeostrohus, Ett., Foss. Fl. v. Hier., p. S.'i, pi. vi, figs. 22-33.— Ung., Icon., pi. xlii, figs. 16, 17.
rinua i>ala;ostrohm, Heer, Fl. Tert. Htlv., i, p. 50, pi. xxi, fig. 6.— Sap., fit., ii, p. 70, pi. iii, fig. 1,
iv, fig. 3 a.— Gaud., Cont., ii, p. 34, pi. i,tig. 8.— Heer, Mioc. Bait. Fl., p. 56, pi. xiii, figs. 1,2.
Pinus polaris, Lesqx., Auuual Report, 1873, p. 410

Leaves by five, long, liuear-filiform. abruptly pointed; middle nerve thick; lateral veins thin
bnt distinct.'

The first specimens obtained from this species were mere fragments
of leaves, like those of the enlarged figs. 26 to 30, and were, by their
narrow, filiform shape and their nervation, compared and referred to Finus
polaris, Heer, whose leaves are by two only, or of the Pinaster section of
the Pines. The specimen in fig. 25 shows these leaves to be by five, and
therefore of the section Strobus. The leaves are comparatively long, seven to
eight centimeters, about one millimeter broad, flat or canaliculate, abruptly
pointed, with a comparatively thick midrib, and two or three thin lateral veins
on each side. The support is not clearly defined, the leaves appearing sessile
upon a basilar receptacle, rather than surrounded by a sheath. Two of the
leaves seem larger and shorter, as seen in fig. 25. They are apparently flat-
tened fragments, do not show any trace of middle nerve and lateral veins, and
may have been crushed after maceration; this would indicate for the leaves of
this species a soft and at the same time a somewhat thick consistence. The
fragment of branch (fig. 31), with its rhomboidal scars, seen enlarged in figs.
31 « and 31 h, appears referable to this species, though fragments like this
may represent far different kinds of Conifers. One, for example, similar to
this, is figured as Glijptostrobus Europeus in Heer (Bait. Fl., pi. xiv, fig. 13). A

84 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

longer and more slender branch, with scars of the same form, is, however,
referred to Pimis palceostrohus by Saporta (fit., 2, pi. iii, fig. 1 (/). The species is
described by d'Ettingshausen {loc. cit.) for leaves of the same form as ours,
apparently flexible or soft, many being broken or curved. Saporta gives, besides
the leaves and Ijranches, a cone referable to it. As we have leaves only, the
identification of our specimens to the European species is not positive. I have
been unable also to see these leaves either square or triquetrous, as shown by a
cross-section in Flor. Tert. Helvet., pi. xxi, fig. 6 e, and Flor. Bait, pi. xiii, fig. 1 b.
They seem merely flat or canaliculate, and therefore keeled on the back. Tlie
specimens bearing fragments of this Pine have numerous seeds of Conifers, one
of which, represented in fig. 33, does not bear any likeness to that referred to
this species by d'Ettingshausen in Hiir. Fl., pi. vi, fig. 22, but rather resembles
those of Pinus polaris, Heer (Spitzb. Fl., pi. v, figs. 9 and 10). The same
shales have also large scales of cones in fragments, like that in fig. 32, with
broad rhomboidal apophyses, rough and wrinkled lengthwise on the borders,
and a large, deep, central cavity of the same form. Though these fragments
may be compared to some species of Pines, they are undeterminable, and
have been figured here as points of comparison for future researches.

Habitat. — Near Castello's Ranch {Dr. F. V. Hayden) ; near Florissant
{Prof. E. D. Cope).

TAXINB^.

SALISBUEIA.

Salisbui'ia polymorph a, Lesqz.

Plate LX, Figa. 40, 41.

Saliaburia polymorpha, Lesqx., Am. Jonrn. Sci. and Arts, 2d ser., No. 81, May, 1859, p. 3C2; Auuual
Report, 1872, p. 404.

Leaves cnneiforiu, gradually enlarging from the base upward, irregul.arly more or less deeply cut in
obtuse lobes; middle nerve distinct to half the length; veins very thin and close, slightly curving iu
ascending, many times dichotomons.

The leaves of this species, first described from Vancouver's specimens,
are extremely variable. Cuneiform to the base, or enlarging upward, their
borders are variously divided in more or less deep, generally obtuse lobes,
sometimes split to the middle, sometimes merely wavy around the top, and
marked downward by folds along the veinlets, which thus inflated are like
multiple primary nerves, as seen iu fig. 40, which is, however, made from an
obscure specimen. In one of the Vancouver's specimens, the leaf of about

DESCKIPTION OF SPECIES— MONOCOTYLEDONES. 85

the sanity lonn as in tig. 41, but much hirger, has six of these iuflated veins,
passing up IVuiii Ww base, parallel to the secondary veins, and the midrib is
there indistinguishable. As tar as tliey are known, the leaves vary in size from
four to sevi^i centimeters long, and from two to three centimeters wide across
their upper divisions. The secondary veins are very thin, scarcely distin-
guishable without a glass, close, nearly straight up or slightly curved to the
borders, forking in ascending. As remarked above, the specimen in fig. 40 is
obscure, the veins indistinct, and the surface variously folded in the direction
of the veins. Its reference to this species is not jnisitively ascertainable.

Habitat. — Six miles above Spring Canon, near Fort Ellis {Jos. Savage).

MONOCOTYLEDONES.

GLUMACEiE.

Graminetje and Ci/peracea are as yet poorly represented in tlie North
American Tertiary flora; not so much on account of the deficiency of speci-
mens as from the impossibility of determination of fragments of leaves or
blades whose reference, even generic, is always problematic. I have, there-
fore, abandoned a number of species which I had formerly described as
Cyperitct: Cxjperites angustior?, Al. Br (Annual Report, 1872, p. 403); C.
Braunianus ? , Heer (Annual Report, 1871, p. 285), which is characterized
especially by its tubercles, while our specimens represent merely fragments
of stems without them; C. Deucalionis ? , Heer (Annual Report, 1871, p.
285; Poacites l(Bvis, Heer (Annual Report, 1870, p. 385), etc., a fragment of
vvhicli traverses fig. 1 of pi. xliii. As the determination of these species is
still uncertain from far better specimens than those which we have in our
possession, and as none better of the same kind have been discovered since
1870, it is advisable to leave them as undeterminable until others are found,
which may afibrd some more light by the possibility of comparison. As seen
in the description of species ol' Arundo, which are represented with positive
characters, those of the seeds with glumes and pallets, and also of a Carex,
we may expect from further researches impc^rtant discoveries, and, therefore,
the opportunity of more evident references for the fragments which are until
now of uncertain affinity.

86 UNITED STATES GEOLOGICAL SURVEY— TEKTIAEY FLOKA.

GRAMINE^.

ARUNDO, L.

Arundo Ooepperti!, Munst.

Plate VIII, Figs. 3-5.

Arundo Goepperii, Heer, Fl. Tert. Helv., p. 62, pi. xxii, fig. 3, pi. xxiii. — Lndw., Palasont., viii, p. 80, pi.

ivii, figs. 1-6. — Ett., Fobs. Fl. v. Bil., p. 19, pi. iv, figs. 1-4. — Lesqx., Sapplenient to Annual

Report, 1871, p. 5.
CttlmiUn Goepperii, Miinst., Beitr., v., p. 113, pi. iii, figs. 1-3.

Stems large, irregularly narrowly striate, marked with rouud knots; leaves large, flat, with thin
veins, equal in distance, all of the same size.

The large fragment of stem (fig. 3) appeals, by compari.son with the
description of this species by Heer, referable to this species. The lines cover-
ing it lengthwise, as also the round tubercles, irregularly distributed, are of
the same character as in fig. 11, pi. .xxii. The longitudinal larger striae,
mere irregular folds or splits of the epidermis, are not continuous. The
specimen of ours, of which a part only is represented, is evidently of a very
large stem, which, however, has been split and flattened; it does not bear
any trace of articulation, and therefore may represent another kind of vege-
table. The same may be said of the two fragments of branches or leaves
in figs. 4 and 5, which have the same kind of nervation as the large leaves of
A. Goepperii, but have no analogy whatever by their size. The surface, as
seen in figs. 5 a and 5 b, much enlarged, is narrowly grooved or striate, but tlie
veins are at equal distance, fig. 5 b representing them as seen enlarged twice,
5 a as seen enlarged eight times. The articulation and small tubercle in fig.
4 are characters of a branch, and not of a leaf; these fragments are found upon
the same specimens with those of fig. 3, and from the same locality as those
of the following species, to which they may be referable; they are somewhat
inflated above the articulations.

Habitat. — On fine-grained, buff-colored shale, cut off along the railroad
west of Green River, above fish-beds {Dr. F. V. Hayden). The specimens
from this locality are now very rare; those of Dr. Hayden, which represent
many fine and remarkable species, were obtained while the work of tearing
out the rocks for the construction of the railroad was in progress. My
own researches at the same locality, and long time after, did not afford any
discovery of importance.

DESCKIPTION OF SPECIES— GRAMINEiE. 87

Arundo repcrta, Lesqz.

Plato VIII, Figs. 6-8.

Jrundo reperta, Lesqx., Annual Report, 1874, p. 311.

Stem thick, diat.antly articulate ; surface striate, marked with round, obtuse knots, either placed at
the articulation or here and there scattered upon the stem; fruiting ;)aDicIe crushed, oval-oblong,
bearing ovate-lancoolate seeds and pallets mixed with a coating of hairs.

The specimen figured as marked above is very interesting, and proves
indeed, by the characters of the preserved organs, the reference of this i)lant
to the genus Arundo. The thick stem, two and a half centimeters broad,
somewhat flattened, very closely nerved or striate, with veins twice as close
as they are in the former species, is distinctly and distantly articulate, with an
indistinct knot under the convex narrow ring of" the joint, and two larger
convex tubercles at a distance above, in the middle of tlie stem. The same
specimen bears a crushed ear, where glumes or pallets and seeds are mixed
with a coating of short filaments, apparently hairs. The pallets in fig. 7 a are
ovate-lanceolate, acuminate, rounded at the base, evidently veined in the lower
part, slightly turned to one side at the point or straight. The seeds in fig. 7 b
are shorter, but as broad at the truncate or subcordale base, ovate-lanceolate,
pointed, striated on the borders around the convex central nucleus, measur-
ing six millimeters from the base to the point and two and a half millimeters
across below the middle. The same specimens bear numerous fragments of
stems and rhizomas, or roots, like the one in fig. 8, which are all flattened, exactly
linear, irregularly striate, and marked without order, or here and there with
oval concave impressions, in the form of rings around central points, evidently
scars of rootlets. As remarked above, these fragments may possibly repre-
sent the same species as the former. The stem, however, is more closely stri-
ate than the large one in fig. 3 of the same plate. It is comparable to that of
Fhrtigmites CEningensis, Al. Br., as figured by Heer (Spitzb. Fl., pi. vi, fig. 16).
A i)allet referable to the same species is also represented in fig. 15 of the same
work; it is oval-lanceolatc-obtuse, narrower tlian that of our species.

Habitat. — Cut-off above Green River Station, "Wyoming Territory
(Dr. F. V. Hayden).

Arundo? obtiisa, Lesq^.
Plato VIll, Figs. 9-9 c.

Arundo obtusa, Lesqx., Annual Report, 1874, p. lill. •

Stem doubly veined, obscurely articulate, slender ; ])rimary nerves somewhat thick, with four or
five intervening thinner secondary veins ; pallets broadly ovate-lanceolate, acuminate or truncate; seeds
large, obtuse, truncate at base.

The different organs preserved all together ui)on tlic, same .specimen in

88 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

fig. 9 may belong to a species of the same kind as the former. The charac-
ters, liovvever, are somewhat different. The small stem, which, flattened,
measures scarcely one and a half centimeters in width, is indistinctly articu-
late, and bears, just above the articulation, a round scar, about like that of
the largQ stem in fig. 6, described above; but the nervation of the epidermis
is double and distinct, the primary veins, two millimeters distant and com-
paratively thick, being separated by four or five secondary thin veins, as seen in
the enlarged fig. 9 c, about as in Phragmites CEningensis. The other organs
which I refer to the same species, and seen enlarged in figs. 9 a and 9 h, are
two pallets and one seed. Of the first, one is broadly truncate at the base,
rapidly narrowed to a truncate or bicuspidate point; the other is narrower,
ovate-lanceolate, acuminate, rounded at base; both veined lengthwise. The
seed {b) is truncate at the base, short, oblong, or lingulate, very obtuse, smooth,
a little shorter than the pallets, but about of the same width, four to five
millimeters long, a little more than three millimeters broad. The seed has
not the form of those of Arundo; and, as these vegetable remains were found
in connection with Palms', they probably represent some Gramen of a trojiical
or warm climate, like the Bambusice. I have, however, been unable to find
any specimen with seeds for comparison. The stems and leaves of Ba?n-
husia arundinacea have the same nervation as that of our stem in fig. 9.

Habitat. — Golden, Colorado; very rare, and found only in small frag-
ments.

PHRAGMITES, Trin.

Phragmites <Euingensis, AI. Br.

Plate VIII, Figs. 1, 2.

Phragmites fEningensis, Heer, Fl. Tort. Helv., i, p. 64, pi. xxii, fig. 5; xxiv, ixvii, fig. 86; xxijE, fig. 3«. —
Ludw., Palseont., viii, p. 80, pi. xvi, fig. 1 ; xviii, fig. 2; xxlv, fig. 7. — Ett., Foss. Fl. v. Bil.,
p. 21, pi. iv, figs. 6-10. — Lesqx., Annual Report, 1870, p. 364; 1871, p. 289; Supplement to
Annual Report, 1871, p. 10 ; 1872, pp. 374, :576, 391, 399.

Phragmiteaf CEningensis, Al. Br , Stizcnb. Verz., p. 75.

Phragmites Zaniionii, Mass., Syn. Fl. Foss. Senogall., p. 8.

Culmites arundinaceus, Ung., Ett., Fl. v. Vien., p. 9, pi. 1, fig. 1.

Bambusium sepuUum, Andrie, Fl. Siebenb., pi. il, figs. 1, 3.

Rljizomas large, creeping, articulate ; roots linear, with rootlets in right angle, placed in alternate
rows or indistinctly along the divisions; stems long; leaves large, distinctly veined, like the stems,
• without middle nerve.

Fragments referable to this species, more common still in the European
T(;rtiary than in ours, have been found in most of the localities where Ter-
tiary fossil plants were discovered. The essential characters which serve to
idcntiiy them are the creeping, articulate rhizomas, bearing at or from the

DESCKIPTION OF SrECIES— GKAMINB^. 89

arlicukvtions flexuous branches nearly at right angle, two to three millimeters
thick, linear, with two or three rows of radicles, more or less regularly placed,
sometimes in lines, sometimes distributed without order, filiform ; the stems
articulated also, bearing scars of branches at the articulations, and striate in
the lengtli, like the leaves, with primary veins distinct, about two millimeters
distant, and three to six thin veiulets between. As the plants of this kind
were very large, easily crushed on account of their hollow stem, they are
merely found in fragments; at least, we have never seen, in our Lignilic
formations, large specimens of them ; and these fragments, though referable
to the typical forms, present a great diversity of shape. It is, tlierefore,
probable that some of the references are uncertain, and that, when the species
is known by better specimens and more distinct characters, some of the
vegetable organs considered as pertaining to this species will have to be
distributed with other kinds of vegetables. Thus the roots and rootlets first
mentioned in Annual Report, 1870, p. 384, as rather comparnhle to those
of PhragmUes CEningensis, have been recognized and described (pi. vi,
fig. 1) as those of a Fern, Lygodium neuropteroides. It must be remarked,
however, that when these specimens were examined, they were the first ones
seen from the Western Tertiary Measures, when we had no point of com-
parison whatever. This deficiency was already supplied, in a certain degree
at least, for the Report of 1871, and here we have (p. 286) the species
mentioned in numerous fragments of leaves, stems, and rhizomas from Elko
Station. Then (p. 289) a fine stem, with articulations, scars, and l)ranches,
is identified w-ith this species, from Medicine Bow's coal beds, whose station
is referable to the Washakie group. In the supplement (Report, 1871, p. 10),
a stem from Evanston is described, about half an inch broad, with primary
veins deeply marked, strong, separated by thin intermediate veinlets, articu-
late, bearing at the articulation the round scar of a branch. It is more
deeply striated than in most of the specimens figured of this species, agree-
ing, however, with a branch described by Sismondi (Paleont. du Piemont,
p. 410, pi. vi, figs. 3-5). A stem of the same kind, and with the same char-
acter of nervation, is also mentioned in the same supplement (p. 13) as found
with fragments ol' Abie/ ites duhius at the Raton Mountains. In Report, 1872,
p. 376, a fine specimen, with an articulation and scar, is described from the
white sandstone under the Lignite beds of Golden. It is the specimen of our
pi. viii, fig. 1. At the same locality. Prof B. F. Meek discovered a number

90 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

of specimens with rootlets and their capillary filaments. From Marshall's
coal mines, I obtained a number of" fragments, especially of deeply striate
stems without articulations, and of leaves witli a less distinct nervation.
From Black Butte, we have also roots and rootlets like those represented by
Heer (FI. Tert. Helv., pi. x.\ii, figs, be, be), and in the red baked shale
of the same locality, specimens of the same kind. The species is especially
abundant at the Canon City coal beds, where a hard sandstone, at the base of
the highest bluff, and already at a distance above the main coal bed, is filled
by fragments of this species, and also in the hard white sandstone of Grolden.
The specimen represented by fig. 2 of pi. vii is from that locality, found by
Mr. Wm. Cleburn. It is remarkably similar to the rootlets described by
Ludwig, in Palseont., vol. viii, p. 80, pi. xviii, fig. 2c, as those of Fhragmites
CEningensis. In this sandstone, also, Mr. Cleburn found a cylindrical, some-
what conical, specimen, with articulations close to each other, the size of
the stem or rhizoma diminishing with each articulation, and nerved like stems
of this species

As remarked by Heer, this Phragmites is closely allied to P. commums,
which is very common through Europe and North Asia. The fossil species
appears to have been larger, with broader leaves, without middle nerve. It
is remarkable that, though recorded by most of the paleontologists who have
had opportunity of studying vegetable remains of the Tertiary, its racemes
and fructifications have never been seen, a single pallet only being described
by Heer from the Miocene flora of Spitzbergen as referable to it. This
celebrated author has described the species from a profusion of fragments of
stems and leaves, in the clay shale of La Rochette, near Lausanne, some of
them beautifully represented in pi. xxiv of his Fl. Tert. Helv.

Habitat. — As remarked above, in most of the localities where Tertiary
fossil plants have been found, except in the upper group 4 in Wyoming and
Colorado Territories. It is especially abundant at Golden and Cation City
(Dr. F. V. Hmjden, Prof. B. F. Meek, Wm. Cleburn, etc.).

Plira|;niit4>s Alaskana, Heer.

I'l.ate VIII, Figs. 10-12.
Phragmites Alaskana, Hocr, Fl. Alask., ji. '24, pi. 1, fig. 12. — Lesqx., Annual Report, 1871, p. 29G.

Leaves narrow, nerved in the length ; primary nerves distinct, lots distant than in the former spe-
cies; veiulots obsolete, discernible only under the epidermis, three in each interspace.

The fragments represcniing this species indicate leaves uiucli narrower

DESCEIPTION OF SPECIES— GRAMINBiB. 91

than those of the former, and therefore a smaller plant. These leaves, from
one to two centimeters wide, linear, obtuse or obtusely mucronate, arc nerved
in the length, with primary veins one millimeter distant, separated by three
thin, obsolete, secondary ones. The substance of these leaves is hard and
somewhat thick, the epidermis thin but corneous-like, covering the veins and
veinlets, and rendering these indiscernible. This epidermis is, however, some-
times separable from the surface, and then the veinlets are distinctly visible.
The specimen of our fig. 12 has the same size and appearance as that of lleer
{loc. cit., fig. 12). But in fig. 12 b of the Swiss author, the intermediate second-
ary veins are marked more numerous, or by six, though the primary ones are
at the same distance, of one millimeter, as in our specimens. As the author
remarks that the veinlets are obsolete, and as in the corticated specimens of
ours the veinlets appear more numerous, on account of their indistinctness
when seen through the epidermis, it is possible that the number of these
secondary veins has not been distinctly seen, or that, as it is the case with
Phrag?ni(es CEnifigensis, to which, according to the observations of Prof
Heer, this new species is closely allied, the veinlets are variable in number.
Though I consider our species as identical with that of Alaska, I do not assert
that it represents a Phragmites. If the leaf in fig. 10 is rightly placed, and
is obtusely mucronate, this same character, though somewhat less marked,
is seen in leaves of Phragmites CEningensis, as figured by Heer (Fl.Tert. Helv.,
pi. cxlvi, fiig. 22); if, per contra, the specimen is overturned, and if that mucro-
nate part represent the base narrowed to the point of attachment, this would
force the separation of these leaves from the genus Phragmites^ and indicate
their reference perhaps to Bambusia. But it is possible that we have here two
species, as that in fig. 12, which more positively agrees with Heer's description
and figures of P. Alaskana, is from a specimen of a different locality from those
of figs. 10 and 11, the only ones where the secondary nervation could be dis-
cerned by abrasion of the epidermis. These, therefore, might be referable
to Bambusia, and that in fig. 12 identical with the species from Alaska. The
relation of the specimens in figs. 10 and 11 is remarkably close to Phragmites
Cretaceus, Lesqx., as described in Cret. Fl., p. r)5, pi. xxix, fig. 7.

Habitat.— Green River group, specimen of fig. 12, in connection with
numerous leaves of Ficus; the others six miles above Spring Canon {Dr. F.
V. HaTjdeu).

^

92 UNITED «TATE8 GEOLOGICAL SURVEY— TERTIARY FLORA.

CYPERACEiE.

CYPERUS, Linn.

Cyperus Cliavaiieiisis, Heer.

Plate IX, Figs. 1,2.

Cyperus ChavanenMs, Heer, Fl. Tert. Holv., i, p. 72, pi. xxii, fig. 7, xxviii, fig. 1, cxlvi, fig. 22. — Sism.,
Mater., p. 23, pi. vii, figs. 5, (i. — Ett., Foss. Fl. v. Bil., p. 26, jil. vi, fig. 3. — Lesqx., Annual
Report, 1871, p. 300.
Leaves large, obscurely carinate in the middle, nerved ; primary nerves variable in distance, inter-
mediate veins three or more, crossed in right angle by transverse veinlets.

The fragment (fig. 1) represents ajjparently part of a large leaf of this
species, like fig. 1 c, pi. xxviii, of Heer, loc. cit. It has, however, no trace of
a middle nerve or of a carina, and the nerves are somewhat unequal in dis-
tance. It is the same in fig. 2, which has the primary nerves twice as dis-
tant, and seems to represent part of a stem of the same species. In both
these specimens, the cross-veinlets are discernible by a strong glass. This
species is apparently very variable. Sismoiidi {loc. cit.) represents a leaf
narrower than our fig. 1, but indistinctly keeled, and the distance between
the primary veins is the same as in our specimen ; in d'Ettingshausen (Bil. Fl.),
the same character of nervation is remarked ; our specimen (fig. 2) has the pri-
mary nerves at a greater distance, double than that indicated in fig. 1 ; the
spaces of different shades of color resemble the fragment figured by Heer
(pi. xxii, fig. 7). These analogies are uncertain and obscure indeed, and I
should perhaps have done better in leaving undescribed these fragments, as
I have done for others formerly ascribed to this genus. They may be used
for comparison with specimens of other localities. One from Evanston,
for example, represents a flattened stem, one centimeter broad, without
any articulation, obscurely striate, with primary veins distinct and variable in
distance, and veinlets thin and joined by cross-branches. It has been also
described as referable to the same Cyperus.

Habitat. — Evanston, below the coal {Dr. A. C. Peale); Green River

group {Dr. F. V. Hayden).

CAREX, Mich.

C a r c X B c r t li o II d i , Lesqz.

Plate IX, Figs. 3, 4.

Carex Berttiaudi, Lesqx., Annual Report, 1872, p. 377.

Leaves fiat, marked by an obscure middle nerve and bordered by more distinct ones; intermediate
veins thin, close, sometimes obsolete ; seeds ovate or broadly oval, flattened, with oval more or less
pointed and broader perigynia.

The leaves are all in small fragments, Ironi the nature of the matrix

DESCRIPTION OF SPECIES— SMILACEJS. 9^

where the retnains are preserved, a soft white chiy cut in stiuill cuboiiiul
pieces by cleavage They are narrow, two to four millimeters l)roa(I, not can-
aliculate, but with a middle nerve sometimes indistinct, and narrow interme-
diate veinlets very close and thin, the ones nearest the borders somewhat
thicker; the characters are seen in the enlarged figure 4^/. The seeds are
numerous, more or less imbedded into the clay, and therefore of dilferent
aspect according to the plane of their position into the matrix; small, aver-
aging two millimeters in length and only half as broad, ovatcvroundcd at one
end, obtusely pointed at the other, with perigynia of the same form but
slightly broader, forming rings around the aehenia. Their dilferent forms are
figured enlarged double in b and foipr times larger in 3«. Fig. d represcMits
apparently a seed separated from its envelope, and fig. c a very small one, with
the achenium attached to the base and not in the middle of tlie perigynium.
All these different appearances result merely from tlic angle and degree
of compression of these small organs. The fragments of leaves are reflated
to those described by Heer in Spitzb. Fl., p. 48, pi. vi, fig. 45, as represent-
ing a fragment of the culm of Cijperus arctkus; but this has no trace of a
middle nerve. The seeds are comparable to those of Carex antiqua, Ileer
(Bait. Fl., p. 28, pi. iii, figs. 18-20), but these do not have larger perigynia,
forming borders as in ours.

Habitat. — Golden, South Mountain, in white soft clay, with remains of
Flabellaria Zinkeni {Capt. E. Bcrthoud).

CORONARI.E.

SMI LAC P]^.

The species of this order of plants inhabit at the present time the trop-
ical and temperate regions of both hemispheres. We have in the North
American flora fourteen species; Europe has only three or four left in the
Mediterranean regions. This small number is remarkai)le indeed compared
with the great predominance of species of this order in the Tertiary flora of
the same country ,for no less than forty-four are described by European palaeon-
tologists from the Miocene of the south of France and Italy. A number
of fragments of leaves of Smilax have been observed in the Lower Lignitic
measures at Golden and Black Buttes; but they are mostly specifically inde-
terminable. The best specimens were obtained at Carbon. None have been

94 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

seen until now in the Green River group, and none in the Pliocene of Cali-
fornia.

This seems a distribution in a contrary direction to that observed for
Europe. It is, however, very probable that, as the leaves of Smilax have
been found at two different stages of the Lignitic, and as even the type seems
to be represented already in the Cretaceous of Nebraska by vegetable remains
referable to the DioscorecB, the researches in the still unknown field of the
North American flora will fill the gaps by new discoveries.

SMILAX, Toum.
Smilax grandiToIia, Ung.

Plate rX, Fig. 5.

5»nt7ai grandifoUa, Heer, Fl. Tert. Helv., p. 82, pi. xxx, fig. 8.— Ett., Fl. v. Bil., p. 28, pi. vi, figs. 15,
16.— Ung., Syll. PI. Foss., p. 7, pi. ii,fig8. 5-8.— Lesqs., Annual Reports, 1872, p. 385; 1873, p.
395.

Smiladtes grandifoUa, Ung., Cblor. Protog., p. 129, pi. xl, fig. 3.

Leaves brciilly ovate, cordate at the base, gradually acuminate, seven-nerved, the two outside
nerves ascending to the middle, the inside ones passing uj) in a curve to near the point; veinlets distant,
oblique.

This leaf, about seven centimeters long (point broken), five and a half
centimeters broad toward the base, where it is enlarged, cordate at base,
curving and narrowing upward and apparently acuminate, is in all its charac-
ters of form and nervation identical with that in fig. 6 of Unger {loc. cit). It
merely differs by its smaller size, a character of no moment in leaves of this
kind. The secondary veins are slightly more oblique to the middle nerve.
This species is not rare in the Miocene of Europe; it has been described
after Unger, by Heer, d'Ettingshausen, Weber, and represented by leaves of
a shape and nervation more different from the typical form than ours, for the
species varies much in the size and form of its leaves, some being as broad as
ten centimeters, some merely rounded, not cordate, to the base. To this last
variety may be referable a leaf described in Annual Report, 1873, p. 395, whose
lower half only is preserved, round, not cordate, abnormally five-nerved by
the division near the base of one of the lateral veins on one side, and on the
other by a marginal veinlet from the top of the petiole; the nervilles are less
oblique to the middle nerve than in the leaf which we have figured. The
essential difference of this fragment, too incomplete for representation, is the
round base of the leaf and its small size, of the same size as that of fig. 8 of
Unger (loc. cit).

DESCRIPTION OF SPECIES— ZINGIBERACE^. 95

Habitat. — C.irbon Station, in shale from below the main coal witli Equi-
setum Haydenii, Topulus arctica, and other Miocene species. The other leaf
not figured has been communicated by Dr. A. C. Peak, from the white sand-
stone of Canon City Lignitic Measures. I have found it also in fragments
about three miles east of Colorado City, near the Gehrung coal, with Sabal
leaves, and seen it also in fragments at Black Buttes. These fragments, too
incomplete for determination, may belong to a different species.

SCITAMINEtE.
ZINGIBERAOE^.

ZINGIBERITES, Heer.
Ziugiberites dubiiis, Lesqx.
Plate XVI, Fig. 1.
Zingiberitesf undulatus, Lesqx.," Annual Report, 1873, p. 3'JG.

Fragments of a largo leaf, outlines not preserved ; surface equally undulate, multinerved, in right
angle to the folds; primary veins distinct, parallel ; intermediate veinlots six to seven, very thin.

The relation of these fragments is uncertain. They represent parts of
a subcoriaceous leaf, with surface undulations formed by deep furrows, which,
scarcely marked in some places, do not break the connection of thu primary
veins. These are somewhat thick, separated by six or seven thin veinlets,
indistinctly seen through the epidermis. Of the three species of this genus
published by Heer, the only one to which ours may be compared is 2^. mul-
tinervis (Fl. Tert. Helv., iii, p. 172, pi. cxlviii, figs. 13-15). It represents
some traces of the outlines of a large leaf, with a thick midrib, its surface
distinctly nerved with somewhat undulate primary veins, which, as seen in
fig. 15 b, have five thin intermediate veinlets. The character of the nerva-
tion is therefore the same as in the fragments which I refer to this genus,
the primary veins being at the same distance, of about three millimeters, as
in fig. 14, and in some places scarcely two millimeters, as in figs. 13 and 15
of the same author.

Habitat. — Golden, Colorado Territory; very rare.

* The uame of Z. undulatus, sp. nov., was employed by mistake in the short description of this
species, the same having been used by Heer in Bait. Fl., p. 64, a work then unknown to me.

96 UNITED STATES GEOLOGICAL SURVEY— TEKTIAEY FLOKA.

MOSACE^.

MUSOPHYLLUM, Goepp.

miisopliylluiii complicatum, Lesqz.

Plate XV, Figs. 1-6.

ilusopliylhim eompUcatum, Lesqx., Annual Keport, 1873, p. 418.

Stipe thick, wrinkled-striate in the length ; leaves large, with a thick, irregularly veined midrib;
veins distinct, but very thin, simple or dichotomons toward the borders, more or less oblique to the
midrib.

Stipe thick, wrinkled or striate, with parallel lines much larger and more
distant and more irregular than the veins of the leaves; leaves very large
with undulate entire borders, apparently parallel to the irregularly veined
midrib, closely narrowly veined; veins thin, generally simple, scarcely one-
third of a millimeter distant, sometimes forking near the borders, joining the
midrib in an acute angle of divergence, then open and in right angle to it.
The nervation is exposed in its details in fig. 3 a, representing the point of a
small leaf, which seems attached by its whole base to the stipe.

The exact characters of these leaves, especially their form, their size,
and their relation to the main stem, or stipe, are very obscure. I have found
a bed of shale nearly one foot thick filled entirely with fragments of this
species and detached leaves of a Sapindus, and have worked a whole day with
a miner, trying, without avail, to get specimens more definite than those
which are figured here. Large pieces of shale are covered with fragments
of leaves, folded in various ways, where no trace of any middle nerve may
be discovered. This proves the large size of these leaves, which, however,
may be sometimes narrow and linear, as seen in fig. 1, where the borders are
clearly defined on both sides of the large, flat, striate midrib, by a black,
slightly inflated, undulate line, like that of the borders of fig. 2. The frag-
ment in fig. 3 represents, it seems, a stipe with part of a descending rhizoma,
which has the same kind of surface wrinkles, and above it a pedicel obliquely
placed upon the stipe, and who.se top is covered and obliterated by folded
portions of leaves. The midrib, irregularly striate, is more distinctly seen
in fig 1, and less so in fig. 2, where it is covered at the base by fragments,
perhaps referable to leaves, but whose nervation is coarser and more distant
than in any of the numerous portions of leaves of my specimens; the undu-
late entire borders arc not laceration of a broad linear leaf, but true bor-
ders. Fig. 4 has the veins very close ; it may represent the top of a young

DEiCKlPTION OF SPECIES— ENSAT^. 97

leaf, the veins apparently converge toward the niuldle, as it happens some-
times at the top of a middle effacing nerve. The lower fragment of fig. 2
pertains to the same leaf split in the middle; it has its borders still more
distinctly defined. The specimens bear these crushed leaves generally mixed
with fragments of rhizomas and radicles (fig. 6), whose ultimate divisions are
short, capillary, and so immerous that they cover sometimes whole specimens
as by a thick felt, so that every kind of form and every trace of the stone are
obliterated. The rootlets are linear, flattened, resembling those of Phragmites
CEningensis, while the rhizomas, or perhaps their ])rimary divisions, measuring
seventeen millimeters in thickness, are straight, narrowly striate, like the
diverging branches of fig. 3, exactly linear, with branches half as thick, passing
down in an acute angle of divergence, bearing rows of somewhat di.stant scars,
formed by double, deep, circular lines and a central vascular point. These
roots and rootlets evidently pertain to this species, for the leaves of Sapindus
ohtusifolius are rarely mixed with those of this Musopkyllum, and are found
separately and in profusion at a little distance of the same bed of shale. Mr.
Cleburii, who visited the same place, collected only leaves of Sajnndus, with
a single one of Alnus Kefersteinii and no fragments of MusophyUum. The
surface of the leaves of our species is sometimes covered with parts of the
epidermis, which, seen with a strong glass, appears marked crosswise with
thin veinlets, ag in M. bllinicuin, Ett. (Bil. Fl., p. 28, pi. vi, fig. 11). But
in this last species, the veins are not in an acute angle of divergence to the
midrib, and then open and in right angle to it, a character remarked in the
American form, and also in M. hohe.micum^ Ung. (Sillog., p. 8, pi. 1, fig. 13),
which, however, differs by thinner, still closer veins, without cross-veinlets.
Another species, M. speciosum, Sap. (fit., i, p. 77, pi. v, fig. 2), is repre-
sented by a fragment without middle nerve, the veins being thin and crossed
by veinlets, and therefore of the same character as in the American species,
differing, however, by the position of the veins, very oblique to the borders.
This last species is Upper Eocene, or from the Gypses of Aix; the others
Miocene.

Habitat. — Shale over a thin bed of coal, eight miles southeast of Green
River station; referable to the Washakie group.

ENSATil.

This class of plants, and that of the Fluviales, so abundantly represented
in the flora of the present time, have left very few traces of their life during

7 T F

98 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

the geological epochs. Scarcely a dozen species have been described by
European authors as referable to the first division; the second has about
sixty, half of which belong to the genus Potamogeton.

HYDEOCHAllIDE^.

OTTELIA, Pers.
Ottelia Americaua, Lesqz.

Plate LXI, Fig. 8.

OUdia Americana, Lesqx., Annual Eeport, 1874, p. 300.

Spathe oval, narrowed to a round, striate pedicel, surrounded by an undulate, wrinkled fringe,
truncate or emarginato at the top.

The figured specimen is the only one seen of this species. It represents
an oval, tumescent, convex nucleus, the center of a spathe, wrinkled, striate
in the length, three and a half centimeters long, seventeen millimeters broad
in the middle, narrovs^ed to a thinly striated or veined, half-round pedicel,
twelve millimeters long and a little more than two millimeters broad. In the
middle of the spathe, the longitudinal striae are thick and deep, irregular,
gradually passing toward the base to narrow, parallel lines, which enter the
pedicel. This spathe is surrounded by an undulate wing, wrinkled across,
eight millimeters broad in the middle, gradually narrowing to the base of the
spathe, where it joins the enlarging pedicel. The cross-wrinkles of this wing
correspond to the sinuses of the undulations.

The determination of this fine and remarkable species is due to Count
Saporta, to whom I owe the communication of a good figure of Ottelia alis-
moides Pers., a living species of Ceylon. The only difference is that, in the
living plant, a triphylle calix is persistent at the top of the spathe, while in
ours its place is occupied by part of the wing, which possibly represents a
flattened calix. No fragments of any other organs of this species have been
discovered. Leaves of one species of this genus have been described by
Saporta from the Calcaire grassier (Middle Eocene) of Paris.

Habitat. — Point of Rocks (J)r. F. V. Hayden).

POTAME^.
NAJADEJE.

CAULINITES, Brgt., Ung.

Stems branching, striate, marked by sublunar and round scars; leaves flat, doubly striate;
rbizomas piano-articulate, with double, irregularly round scars of branches, or warts and dots, marking
points of attachment of radicular lilaments.

This genus is slightly modified, according to the characters of the frag-

DESCRIPTION OF SPECIES— POTAME^. 99

ments which I refer to it. The relations of the vegetable organs which have
been described under this generic name are uncertain; the genus itself is thus
temporarily limited.

Caulinitcs sparganioides, Lesqx.

Plate XIV, Figs. 4-11.
CauUnilea sparganioides, Lesqx., Aonaal Report, 1872, p. 391.

Stems or rhizoiuas flattened by compression, cylindrical in the natural state, irregularly striate in
the length, horizontally wrinkled, distantly articulate or marked acroHs by semilunar scars, like im-
pressions of half-embracing leaves, and irregularly round, large warts; branches alternate, distant;
leaves flat, linear, veined.

It is indeed uncertain if all the fragments figured on the plate represent
stems only, or if some of them are referable to rhizomas. The largest frag-
ment of a stem of this species is three centimeters across. All the speci-
mens which we have seen, and they are very numerous, are flattened by
compression. They are, in the natural state, cylindrical, more or less inflated
at the nodi, semi-articulate, or cut by a deep double line surrounding the base
of the branches, embracing half the stems, as in figs. 4 and 10 ; they seem,
however, to bear traces of real articulations, indicated by scars of rootlets
surrounding them, as in figs. 5 and 6. If it is the case, these fragments
might be considered as rhizomas and their branches, while the others (figs.
4, 7, 8, 10) would represent stems and branches, with sheathing-leaves under
them. The presence of sheaths is marked under the branch-scars by the
deep grooves, half embracing, in figs. 4 and 10; but one is seen also at the
lower end of fig. 6, which, besides, has in the middle two distinct branches
without appearance of sheath, and between them traces of an articulation.
For this reason, the relative position of all the fragments is not clear. Fig.
8, for example, has in its lower part a branch similar to an unopened bud,
with a round wart at its base, while the upper end shows like a bud of
the same form, half sheathing or embracing, with apparently a fragment
of a leaf behind it. The diflerent representation of these two buds, or
branches, may result from their position, the one at the top being seen
flattened vertically, the other laterally. Fig. 7 shows the scar of a small
branch, underlaid by a wart, a kind of scars considered and described by
autliors under this name, but which may be remains of abortive or adven-
titious rootlets. They seem to be very rare upon the parts which I
consider as the branch-bearing leaves of the plants, as in figs. 4 and
10. These stems, when in a good state of preservation, arc covered by

100 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

a comparatively thick jjellicle of coal, which, cleaved in right angle, appears
to have formed by compression the cross-lines visible upon the fragments
in figs. 4 and 10. They are not remarked upon the others, which bear
more or less distinct, irregular points, seemingly the scars of detached
filaments or scales, as in fig. 9. I consider fig. 11 as a fragment of a leaf
of this species. It is flat, exactly linear, twelve millimeters broad, doubly
nerved; the primary veins are equidistant, one and a half millimeters, the space
between being filled by eleven or twelve thin veinlets. The scars in fig 10 pre-
sent the peculiar appearance of a row of rays surrounding it, as seen sometimes
around the branch-scars of Equisetu?n. From the appearance and difference
of characters of the fragments figured, it may be supposed that they belong
to different kinds of plants. I believe, however, that they are all of the
same species; I found them, and studied them in place, all at the same
locality, where, from a number of specimens which have not been figured, their
different appearances and characters were seen evidently united by interme-
diate forms.

I was at first disposed to consider this species as identical with Caulin'Ues
horealis, Heer (Fl. Arct., i, p. 145, pi. xxiii, fig. 13), which the author sup-
poses referable to his C. duhius (Fl. Tert. Helv., iii, p. 170, pi. cxlviii, figs. 1
and 2). But the characters do not agree, except perhaps in the round form
of the warts, and even these have not inflated borders in our species. Heer
figures and describes the stems as very narrowly and equally lined, while, in
the American form, they are distantly and irregularly striate ; the position of
the warts is not the same, nor are tlie apparent articulations remarked upon
the Miocene specimens from Iceland. In the same shale with these frag-
ments, I have seen long, ilexuous, ribbon-like rhizomas, fifteen millimeters
broad, their surface wrinkled lengthwise, bearing long, linear, flexuous, simple
rootlets, three millimeters broad, diverging in right angle. These rootlets
come out single and opposite, or in whorls of two to four, from inflations at
the point of attachment, and thus produce an appearance of articulation upon
some of the primary roots or rhizomas. This agrees well enough with what
is seen upon the specimens figured. The fragments in figs. 16 and 17 are of
doul)tful reference. The nearly regular position of the scars in rows, and the
opposite branches, seem to separate thetn from this species. Saporta considers
them as branches of Salishuria.

Habitat. — Abundant at Black Buttes in the sandstone above the main

DESCRIPTION OF SPECIES— POTAME^. 101

coal; rare at Golden. I have also referred to this species, with doubt, how-
ever, a few fragments from six miles above Spring Canon, in the collection

of Dr. Hiujden.

Caulinitcs fecundus, Lesqz.

Plate XIV, Figs. 1-3.

CatiUnitea fecundus, Lesqx., Aunual Report, 1872, p. 384.

Brancbcs of racemes two millimeters wide, smooth, inflated along the borders, divided in opposite,
erect branchlets, half as thick, bearing on each side, and on short pedicels, simple round capsules inflated
as by a central nucleus.

It is very questionable if these fine fruiting branches may be referred to
this genus. They have this in common only: a monospermous (?) nucula with
a cellulose envelope. As I have been unable to find either in the fossil species
described until now, or in the collection of living plants which I was able to
overlook, anything to which they had an apparent relation, I have left them
in this as yet vaguely defined genus. The disposition of the nutlets along a
narrow branch like a common racliis has some likeness to that of Potamoge-
ton. The capsules are mostly opposite, close to each other along the branch -
lets, twelve or more on each side of an elongated common pedicel ; one and a
half millimeters broad, flattened, but with a convex center marked by a slightly
smaller nucleus represented by a vesicle of coaly matter easily separated from
its envelope (figs, la and 1 6, enlarged). The fragment in fig. 2 represents the
lower part of the raceme; it is narrowly, distinctly lineate in the length; but
the divisions are smooth, with slightly inflated borders, as in the large middle
branch of fig. 1. Fig. 3 represents a flattened linear fragment of a rootlet
apparently detached from a rhizoma, as seen from the half-rcund scars which
it bears near the upper end. These may be scars of smaller brandies diverg-
ing around and from the axis, as in the former species. This fragment
seems positively referable to the one described here, as no other kind of fluvial
remains of plants were found in connection with it. It is lineate in the
length, the lines regular and equal, crossed in right angle by narrow wrinkles,
and thus has a facies similar to that of the stem (fig. 4) of the same plate.
These racemes might represent the fruiting part of the former species (?).
As remains of Palms were found in connection with them, they may be
also the undeveloped flowers of some Palm. In the fourth volume of
the Arctic Flora, Prof. Heer describes and figures, from the Jurassic of East
Siberia, as a Fern, Thyrsopis Maakiana (p. ol, pi. i, figs. 1—3), which has,
by its fruiting pedicels, a remarkable likeness to this species of the Lig-

102 UNITED STATES GEOLOGICAL SURVEY— TEETIARY FLORA.

nitic. But in (liis the nutlets arc hard, compact, all of the same form, the
young and the old ones, and tlius unlike sporanges of Ferns. The analogy
seems rather to be with the species described and figured as Leptomeria gra-
cilis, Ett. (Foss. Fl. V. Htir., pi. xiii, fig. 5).

Habitat — Erie, Colorado, sandy shale above main coal.

FLUYIALES.

LEMNACE^.

licmna scutata, Daws.

Plate LXI, Fig8. 2, 5.

Levma ecutata, Daws., Eep. on the Geol. of the 9th Parallel, AppendixA, p. 329, pi. xvi, figs. 5, 6. — Lesqx.,
Annual Report, 1874, p. 300.

Frond round, entire, slightly undulate on tbe edges, single or grouped; roots numerous, iiliform,
proceeding from a round spot near the notch of the frond.

The fronds of this species, as represented in our figure, are eight to
twenty-five millimeters broad, exactly round in outline, either and more
generally naked, without radicles, or bearing a fascicle of filiform very slender
rootlets from a narrowed base resembling a short pedicel. Those without
radicles (as in fig. 2) show the basilar (I) part in the center of the frond,
and thus resemble a flattened vesicular plant. In both figures, distinct
veins are seen passing up from the short pedicel, or, as in fig. 2, diverging
around from the center. Comparing them with those of the author's, the
similarity of shape of the fronds is striking, but the fronds figured from
Canada have scarcely any trace of veinlets, a few only being indistinctly
marked in fig. 5 {loc. cit.), and the fascicles of radicles are attached, it seems,
to the borders without any pedicel. From the observation of Prof. G. M.
Dawson, who collected the specimens, this species is found upon shale break-
ing very easily, and no sufficient representation could be obtained of the spe-
cies, though its remains were plentiful. As remarked by the author, it "/.v asso-
ciated tvith great quantities of roots and rootlets of filiform, subaquatic leaves",
and our specimens are in the same way intermixed to a mass of radicles, so
thickly interwoven that it is not possible to precisely see their points of con-
nection to the numerous fronds mixed with or deposited upon the tissue.
Each frond, however, when considered separately upon detached fragments,
looks as if it was completely surrounded by rootlets connected to or depend-

DESCRIPTION OF SPECIES— AEACEiE. 103

ing from it. From tliis it seems that the identity of these plants of Point
of Rocks and of those described by Prof. Dawson is not positively ascer-
tained. In my opinion, they represent the same kind of vegetables, and are
referable to the following species, with wliich they arc mixed.
Habitat.— Point of Rocks {Dr. F. V. Hayden).

SPADICIFLOR^.

ARACEiE.

PISTIA, Linn.

Pistia corrngata, Lesqx.

Plate LXI, Figs. 1,3,4,6,7,9-11.

Pistto corrugata, Lesqx., Annual Report, 1874, p. 299.

Leaves broadly obovate, incrassateU from tbe miilillo toward tho base, bordered upward by a
wavy mar-in, gradually narrowed into a abort pedicel witli bundles of radicles at its base; veins
going out from the pedicel in two or three compact fascicles, dividing in passing up from tUobaseof the
leaves, and forming, by cross-branchlets, large irregular polygonal meshes.

The leaves, round when young, arc, when fully developed, broadly
obovate or round at the upper border, gradually narrowed from the middle
downward to a short pedicel, varying in size from two to six centimeters
long and from two to four centimeters broad; nervation distinct, formed by
the°subdivision of veins, inflated into the pedicel, and dividing irregularly, in
more or less dichotomous branches, in ascending to the borders, forming, by
nerviUes, oblique or in right angle, distinct quadrangular areola?, which
become smaller and quadrate along the borders. The lower part of the
leaves appears inflated or thickened, and is generally surrounded by a
deep line, the inside of which is slightly convex, and passes around and
under a flattened border whose areolation is generally more distinct and
.smaller. This line is more or less discernible upon most of the specimens,
which arc very numerous; but sometimes it is marked near the base only,
as in figs. 1, 4, 6, 7, and, when passing up, disappears into the meshes of the
areolation along the inside line of the flat borders. Sometimes, as in fig. 3, it
is more deeply marked up^vard, and disappears on the sides, leaving the
lower part inflated as far down as the pedicel. In small, apparently young
leaves, as in fig. 1, the circular line is less distinct, and its internal part does not
seem inllated; even in very small leaves the Ijorders are not separately traced,
and the nervation is not disconnected from the base to the circumference.

104 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

This is remarked especially in the two specimens (figs. 2 and 5) which I have
described as representing Lemna scutata, Daws., but which seem positively
referable to this species; the first by its nervation from the center, appearing
as if a leaf, like that of fig. 1, had been inflated and compressed, and its
pedicel flattened to the central ])art, wherefrom the veins are diverging around,
and in fig. 5, which, with the outlines of the larger leaves, has its primary
veins ascending from the pedicel to the borders, scarcely divided as yet, on
account of its incipient development. As the specimens are very numerous,
distinct, and the leaves compressed and imbedded into the stone in various
directions, it is not surprising to see this diversity of forms, which, however,
is merely casual, and which seem explainable in comparing the fossil
leaves to those of Pistia spathulafa, Michx., of the swamps of Louisiana.
These leaves have the same obovate shape, and the same type of nervation, by
inflated primary veins diverging from the point of union of the pedicel with
the lamina, and dividing upward in an irregular dichotomy, forming by cross-
nervilles an areolation similar to that of the fossil species. Moreover, most
of the leaves of the living plant, especially the old ones, bear on the under
surface an inflated spongious coating, which covers them from the base to
above the middle, especially along the primary veins, and which is exactly
similar to that observable on the lower surface of the fossil leaves. It is true
that the black lines encircling the intumescence are not remarked in leaves of
Pistia spathulata. But they may be traced by folds caused by compression,
the folding following of course the border of the part inflated by the peculiar
deposit of the under side, which seems formed by an agglomeration of radicles
and of their detritus by decomposition. In some of the fossil leaves, as in
fig. l,for example, the disconnection of the nervation along the lower rim of
the flat border is scarcely noticeable, and, though more distinct in fig. 3, a
slight folding along the rim would sufficiently account for it. It is not so
easy to explain tlie central appearance of the pedicel or base of the leaf of
fig. 2, just in the central part of an exactly round outline, if this specimen
represents a leaf of the same kind. This could be done only by supposing
that the lower part of the leaf with its pedicel has been folded up, com-
pressed, and eflTaced by maceration, leaving only the space marked in the upper
part of the leaf as trace of its existence. The lower part has not any veins,
while the other half has them corresponding in size and mode of ramification

DESCRirTION OF SPECIES— AKOIDE^. 105

to those of the other leaves. As for the small specimen (fig. 5), it is scarcely
possible to doubt its identity with this species; it is evidently a young leaf,
the nervation is, as also its shape, of the same character. All these leaves
are membranaceous except the middle inflated part, and in all, the veins are
distinct, as if tiie substance of the leaves was transparent. The radicles,
coming out in bundles from linear rootlets, confirm the reference of this spe-
cies to Pistia, for P. si^atliulata has long flexuous rootlets of the same kind,
with capillary radicles, often forming a coating on the surface of the water,
and seemingly supporting the plants. Comparing these plants in any of Iheir
forms, none of them can be considered as representing species of Lemna,
not otdy on account of their size, which, even in the smaller specimen (fig. 5),
is greater than in any species of Lemna known at our time, but especially
on account of the position of the radicles, which, in Lemna, are neither
pediceled nor attached to the borders. This observation is applicable
equally well to the plants considered as Lem^na by Prof Dawson.

No species of Pistia has been published to this time from fossil speci-
mens. Count Saporta has recently found, in the Upper Cretaceous of Fuveau,
France, leaves of this kind {Pistia Mazelii, Sap., ined.), a species which, as
seen from the figures kindly communicated, has not any relation to ours.

Habitat. — Point of Rocks, often covering large pieces of shale by
numerous leaves and radicular filaments. Both Dr. F. V. Hayden's and Mr.
W?ii. ClthurrCs collections have a large number of specimens representing
this species only.

AROIDE^.

ACORUS, Linn.
Acorns brachystaciiys, Heer.

Plato XIV, Figs. 12-15.

Aconite brachyslachys, Hecr, S[)itzl). Mioc. Flor., p. 51, pi. viii, figs. 7,8. — Lesqx., Annual Report, 1871, p.
a88 ; 1872, p. 385.

Scapo rouud, narrow, striato leugthwisf, distantly articnlate; flowering ears oblique, small,
oblong; flowers numerous, in spiral around the axis.

The scapes, four to seven millimeters thick and more or less distinctly
lineatc lengthwise, distantly articulate, bear small flowering racemes, citlier
oblique or drooping, as in fig. 12, short, about one centimeter long and tiiree to
five millimeters thick. Our fragments, as represented especially in figs. 12
anil 13, are so exactly similar to those described and figured by Heer {loc. cit.)

106 UNITED STATES GEOLOGICAL SURVEY— TEKTIARY FLOEA.

that it is impossible to doubt their identity with the European species; and
though fig. 15 has the ears somewhat thick, this difference of size is marked
also in both the specimens from Spitzbergen. Possibly figs. 16 and 17 of
our plate arc referable to the same species. They have been described in
Annual Report, 1873, p. 410, as Acorus affinis, spec, nov.? Fig. 14 repre-
sents a young scape; fig. 15 a is the enlarged ear of fig. 15.

Habitat.— Creston, Washakie group {Dr. F. V. Haijden, fig. 12); Car-
bon (figs. 13 and 14). Fig. 16 is from a specimen from Castello's Ranch,
communicated by Prof. Cope, and as this locality is Upper Miocene, its iden-
tity with fig. 17, which is from Black Buttes and Lower Eocene, is therefore
rendered doubtful.

Monocotylcdoncs incertck sedis.

ERIOCAULON, Gronov.

Eriocaiilon! porosiim, Lesqz.

Plate XVI, Figs. 2, 2 a.

Eriocaulonf porosum, Lesqx., Annual Eeport, 1873, p. 396.

Leaves basilar, rosulate, spreading, entire, linear-lanceolate, broader in the middle, gradually
tapering upward to a sligblly obtuse point and downward to the sessile (f ) base (not seen) ; substance
thick, spongions.

By the thick, apparently porous and spongious consistence, by the
rosulate superposition, and by the form, these leaves are referable to this
genus. They, however, differ by their larger size and the appearance of a
middle nerve. As seen in fig. 2 a, enlarged, the middle nerve is traced by a
broad, flat depression, along which the veins are parallel, as in some species
of this genus; Pcppalanthus melaleucus and Eriocaulum mode.stum of Brazil,
for example. The leaves of the fossil species, four to five centimeters long,
seven millimeters across in the middle, are broader and longer, and have also
the surface narrowly wrinkled across or in an oblique direction to the middle
(fig. 2 a), these wrinkles tending downward and passing down along the
borders, sometimes like anastomoses of the veins. The base of these leaves
is either covered by superposition of others or destroyed; it is therefore
impossible to further extend the comparison. Aholhoda poarchon, Sieb., of
Brazil, a species of the same group of tlie XiridecB, also offers a likeness by
its leaves to those of this fossil plant.

Habitat. — Sand Creek {Mr. W. II. Holmes), w ith leaves of NelumUum
and other species found also at Golden, and therefore of Lower Eocene type.

DESCKIPTION OF SPECIES— PALM^. 107

Phyllitns improbatus, Iiesqx^

riate XIV, Fig. 18.

Ehizocaulon gracile, Lesqx., Annual Report, 1873, p. 39G.

Branches slouder, irregularly forking ; leaves (?) oblong, recurved or oblique, narrowed to a very
sbort pedicel; nervation obsolete.

Comparing originally some fragments of this plant, all still more incom-
plete than the one figured, with Rhizocaulon iwhjstachiuin. Sap., as figured in
Schimper's Pal. Vdg(it., pi. Ixxx, fig. 8, I found a kind of likeness in the form
of the spikelets, which, when crushed, as are some of Ihc upper part of the
figures, seem to represent a surface covered, like our plant, with a carbona-
ceous layer, marked in the middle by an indistinct depression like a midrib.
I had not then obtained the admirable work of Saporta, iStudes, where the
genus Rhizocaulon is not only described in detail, but where many fine spe-
cies are illustrated. From it I had to see the double error of my former
nomenclature and description, Rhizocaulon gracile being one of the species
described by Saporta, and this fragment of ours being without relation wliat-
ever to species of this genus. It is still uncertain if the branch figured here
bears leaves or spikelets rendered obsolete by compression. Some of the
so-called leaves have no trace of a midrib, and seem mere flakes of carbona-
ceous matter of an oval, oblong, obtuse shape, seemingly narrowed to a very
short pedicel, or sessile. The fragments should have been omitted, as of a
character too uncertain for description, and are mentioned here merely to
correct a double error of determination.

Habitat. — Black Buttes, burned shale, above main coal.

PRINCIPES.

PALM^.

Specimens of Palm leaves and fruits are very abundant in the Lower
Lignitic Measures of this continent, especially at Golden, the Raton Mount-
ains, and in Mississippi. The number of species which they represent is
large; but their characters, when taken from fragments of leaves, or from
the rays only, are rarely definite enough to authorize specific or even generic
separation. I have therefore described and figured only the types more posi-
tively characterized, either by their leaves or by their fruits.

The Eocene species of Palms, as represented by specimens of the Lower
Lignitic Ibrniations, relate, as far as we know them until now, to lliree gen-

108 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

cral subdivisions of this family, and are accordingly distributed to the following
genera : —

1. Flabellaria, Sternb., modified in the generic characters, and limited to
Palm leaves or fronds whose rays are all attached to the top (either rounded,
or truncate, or obtusely angular) of the rachis. This generic name has been
employed, when referred to Palms only, for all the fossil species of this
family whose relation is not well ascertained. But it seems an anomaly to
describe in the same generic division species of Palms with flabellate leaves
and acutely carinate rays, all attached to the top of an obtuse rachis, and
others with either flabellate or pinnate fronds whose divisions are fixed along
a generally narrow and very long rachis.

2. Sabalites. Fronds with rachis broad, often enlarged at the top,
gradually narrowed up to a long acumen, bearing deeply cariuato-costate rays
attached along it, and flabellate, like Sahal.

3. Geonomites. Fronds with a long and comparatively narrow rachis,
simple at first, but soon pinnately divided, or laciniate, with rays carinate
er half-cylindrical toward the base, joining the rachis by their whole base,
sometimes half-sheathing.

In describing Flahellaria longirachis, Ung., Schimper remarks, in Pal.
V^g^t., ii, p. 492, that it evidently constitutes the type of a peculiar genus,
which, by its characters, the length of the rachis, etc., seems intermediate
between the Palms with flabellate and those witli jjinnate fronds. Count
Saporta writes the same in regard to the species here referred to this genus,
and believes, according to his remarks in S(jzanne Flora, p. 339, that they con-
stitute a distinct group, having some analogy with the Geo7ioma oi^ the present
time. This genus, according to Willdenow, who established it, is distributed
by a number of species in tropical America, between 20° latitude north and
10° latitude south. Its characters have some analogy with those of (he fossil
species of this division by the fronds, at least, which at first, simple and
flabellate, soon divide in irregular pinnae, and become laciniate, the rays
sometimes half-sheathing, etc.

With the fragments of Palm leaves, and in the same strata, numerous
hard fruits have been found in the Lower Lignitic. They are especially
common at Golden. As they bear (he. characters of fruits of Palm, it seems
advisable to describe them separately under a more appropriate name than that
of Carpolithes. The name of Pahnocarpon is therefore used here for the clas-

DESCRIPTION OF SPECIES— PALM^. 109

sification of all llie so-called Carpolithes, whose relation to Palms seems
evident. The fruits attached to a Palm leaf, and apparently referable to the
same species, are described under the same specific name as the frond.

The Palms, those ''noble children of the earth and of the sun", as Martins
names them, mostly iidiabit the intertro[)ical regions of the globe. They live
in the humid bottoms of the equatorial rivers, of the Amazon especially, on
the shores of the oceans, sometimes upon the slopes of high mountains, either
in dense forests or solitary, or, perhaps, grouped a few together, in vast plains
deprived of any other kind of arborescent vegetation. In the North American
continent, they do not pass above the 34° of north latitude, following the same
distribution upon the Atlantic and the Pacific slopes. In Europe, they reach
the 43° ; in the southern hemisphere, the 36°. The northern species, Chu-
tncBrops, or Sabal, are of small size, and, though elegant in their form, scarcely
give an idea of the splendid, graceful shape, and of the enormous develop-
ment, which impart to the vegetation of the tropics a character of magnificence
and grandeur of which no description, no representation, may give a just idea.
Trunks of Palms of less than oi.e foot in diameter, cylindrical, simple, or clear
of any branches, bear, one hundred feet and more above ground, their crowns
of leaves, sometimes resembling fans, of such a size that one of them is large
enough to cover and wall in the habitation of a whole family. The shape
of these leaves, though most diversified, is always strikingly beautiful.

In the geological times, the Palms appear in the Cretaceous, wherefrom
one or two species have been described in Europe. They become more pre-
dominant in the Tertiary, being already abundant in the Eocene period, where
European paleontologists have discovered twenty-one species ; and still more
predominant in the Miocene, from which forty-two species are described,
mostly from its lower divisions and from the South of Europe. No remains
of Palms have l)een until now recognized in geological formations of Europe
above 52° north. Heer has described none from the Baltic Miocene flora
and none from the Arctic. In Nortli America, there is an indistinct trace
of the presence of Palms in the Cretaceous of Nebraska, by small fragments
of striated leaves, described as Fmhellaria? minima. In the Lower Lignitic
Eocene, immediately at the top of the Cretaceous Measures, the Palms are
already extremely abundant at Point of Rocks, at Black Buttes, and still more
at Golden, where the Eocene facies of the flora is marked, as in Europe, by
a profusion of remains of trunks, mostly silicified, and thus distinctly pre-

110 UNITED STxVTES GEOLOGICAL SURVEY— TERTIARY FLORA.

senting to the naked eye the well-known characters of the wood of the Palms.
The collections of vegetable remains from the Raton Mountains are composed,
for more than one-half, of fragments of Palm leaves. In the Eocene flora of
the Mississippi, the proportion of Palms is quite as large, if not more, indi-
cating, it seems, a small but gradual increase in the degree of temperature
toward the south. We have, however, in the United States, specimens of
large fronds of Palms from Fort Union, near the southern limits of British
America, at about 50° of latitude north, and still higher, from Vancouver's
Island, at 52°, the same latitude where, as seen above, the limits of Palms
have been recognized in Europe at the Miocene epoch. From the Eocene
times, the Palms seem to gradually lose in preponderance in the subsequent
formation of this continent. No specimens of this kind of plants have been
seen at Evanston, Carbon, or the Washakie (Laramie) groups. One species
is represented in the Miocene of Oregon and one in the Pliocene of the
chalk bluffs of Nevada County, California, a formation from which we know
only fifty species of plants, whose characters indicate a climate analogous to
that of the Gulf shores, or of the American Southern Atlantic States, at our
time. Hence, a gradual diminution of atmospheric heat seems to have been
continued from the Eocene to the Glacial epoch, at least, in considering
the distribution of the Palms.

FLABELLAEIA, Schp.
Flabellaria Ziiikenit, Heer.

Plate IX, Figs. 6, 8.
Flalellaria ZinTcenii, Heer, Boernst. Fl., p. 11, pi. ii, figs. 3, 4.— Lesqx., Annual Report, 1872, p. 377.

Eays linear, flat or obscurely carinate ; primary veins distinct, with four to six intermediate tbin
veinlets.

I refer with some doubt to this species fragments of Palm rays found
altogether in great number, but in such small specimens that the characters
of the leaves are lefl indefinite. These rays, rarely conjoined, or generally
separated like blades of grass, varying from five to seventeen millimeters
in width, are flat, somcliraes convex, as in b, fig. 7, or obscurely carinate, with
thick primary nerves, slightly convex upon their surface, one to one and a
half millimeters apart, separated by four to six thin intermediate veins, accord-
ing to the distance of the nerves, and distinctly seen with the glass, as marked
in the same fig. 7 c, enlarged four times. The best of our fragments have
been figured here. Tiiough very similar to those which have been figured

DESCRIPTION OF SPECIES— PALM^. HI

and described by the author, I cannot consider this species as positively iden-
tified with that of Boernstaedt, whose rays are Hghtly {leviter) carinate with
intermediate veinlets sometimes three to eleven, though more generally five
to seven, according to Ileer's remark. Our specimens scarcely show any
trace of carina, the primary nerves being sometimes convex, but they have
all the same appearance upon larger or narrow rays; and the intermediate
veinlets are four to six in number, rarely seven. In some fragments, as in
fig. 8, the nerves are effaced, and the intervals, much larger, are apparently
filled by numerous indistinct veinlets. The fragment a to Z), fig. 7, apparently
represents the same part as that of fig. 5 of Ileer, loc. ciL, which the author
considers as a floral involucre. The veins are all equal, and close to each
other. The presence of these fragments among both the European and the
American specimens, and their similarity of characters, seem to prove identity
of species.

Habitat. — Golden, Soutli Table Mountain, in a stratum of white hard-
ened clay, with Carez Ikrthoudi. Fragments apparently referable to this
species are mixed with Lygodkm neuropteroides from Barrell's Springs.

Flabellaria Eoceiiica, liesqz.

Plate XIII, Figs. 1-3.

FlaleUaria Eocevica, Lcsqx., Annual Report, 1872, p. 391.
Sahal communis, Lesqx., Annual Report, lb74, p. 311.

Frond large, rays convex, semi-cylindrical toward the base, flattened in the upper part, diverging
from the top of a broad rachis, distinctly nerved; primary nerves distant; intermediate veins thin,
close, averaging ten in number ; rachis truncate at its upper face, rapidly narrowed to a point on its lower

The two figures (1 and li) are counterparts, and therefore show the
two sides of the fragment of a frond and of its petiole. The rachis is flat or
about four to five miUimeters thick in the middle, cut in a broad angle on the
upper side, more elongated on the lower, the prolongation measuring about
two and a half centimeters, with the tip abruptly and obtusely pointed ; sur-
face very narrowly and somewhat irregularly veined lengthwise. Rays all
attached to the top of the rachis, round truncate at base, comparatively few,
about thirty, diverging on both sides at right angle, deeply carinate, without
costse at the upper, half-round surface, rapidly increasing in width, and flat
in the upper part of the frond ; carina? broadly costate ; primary veins more
or less distant and thick, generally black when the epidermis is removed,
one to two miUimeters apart ; intermediate veins thin and numerous, ten to
twelve in the intervals of two millimeters. I refer to this species the frag-

112 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

ment of fig. 3, which shows the same kind of nervation, but more distinct
than in figs. 1 and 2. In fig. 1, however, the impression of the right side of
the specimen shows the rays flat, or nearly so, like those of fig. 3, with distinct
though very thin veinlets. Fragments of this kind are very common at
Golden, at the same locality where I could compare them in place; but
I was unable to find any larger part of the fronds, which, considering the
petiole, do not seem to have been of great size. The species is compara-
ble to F. Lamanonis, Brgt, as described in Sap., fit., i, p. 70, pi. iv, fig. 5, and
especially to F. {Sahal) Andegaviensis, Schp., Pal. Vc'g(it., ii, p. 490, a species
not yet figured, but which, according to the observation of Saporta, is
closely related to that of Golden, and is found in the Upper Eocene of La
Sarthe, France. F. Lamanonis is from the same formation, the upper part
of the Gypses of Aix. The position of the rays at the top of the rachis, not
passing lower and not narrowed to an acute base, seems to indicate the refer-
ence of this species to Flabellaria rather than to Sabal.

Habitat. — Golden, especially common at the locality called Table Mount-
ain, south of the School of Mines; Black Buttes, where I found the specimen

of figs. 1 and 2

^ SABALITES.

Sabalitcs Orayanus, Lesqz.

TtiWe XII, Figs. 1,2.
Sabdl Grayana, Lesqx., Trans. Am. Pbilos. Soc, xiii, p. 412, pi. xiv, figs. 4-6.

Frond large; racbis dilated under the rays, taper-pointed and acuminate; rays very numerous;
primary veins thick, nearly at <qual distance; intermediate veins distinct, few, one lo four.

This species is represented by fronds of large size, of which, however,
we have not seen any fragments better preserved than that figured here. The
rachis is broad, flattened and split at the top by compression, as seen on the
right side of a casually superposed fragment of a ray, gradually narrowed,
acuminate, fourteen centimeters long from the dilated part under the rays,
and here six centimeters broad, regularly striate, as seen only near the
borders, under the lowest rays, where the epidermis is not destroyed,
merely convex downward, at the part where it is broken. The very
numerous rays, ninety to one hundred, attached all along the point of the
rachis, are deeply carinato-costate, slowly enlarging upward, distinctly nerved,
at least where the epidermis is preserved ; nerves thick with few intermediate
distinct veinlets. In the specimens figured here, where we have only the

DESCRIPTION OF SrECIES— PALMiE. 113

base of the rays, these intermediate vcinlcts are only one to three (figs. 1 a
and 1 Z) ), but of course the nerves arc more distant and the intermediate
veins more numerous in the upper and enlarged part of the rays, where they
are generally four to six. In some fragments of large rays from the Missis-
sippi flora, I have counted as many as ten of these veinlets. The relation
of this species to that of Mississijipi is especially marked by the form of the
rachis, enlarged under the rays, and gradually narrowed to a long acumen.
This gradual narrowing of the rachis and its enlargement are distinctly seen
in fig. 4 of the Mississippi flora, as also the large number of rays and the
nervation in separate fragments, figs. 3 and 4, showing also the very slow
increase of tlie width of the rays, and therefore the large size of the frond.

Fig. 2 of our plate represents a part of a large petiole found at Golden
with other remains of Palms, and perhaps referable to this species. The
specimen is twenty-two centimeters long; its form is triangular, obtusely cari-
nate; the size eight centimeters tliick at or near its base, and only half this at
the part where it is broken. The rachis under the rays of fig. 1 seems to
indicate a still thicker petiole; but it is here flattened and therefore enlarged;
moreover, the stalks of species of Sabal are rather narrower at a distance from
the fronds than under the rays. The character of coarse regular striae is
the same on the border of this petiole as remarked under the rays of fig. 1.

Habitat. — Golden, Colorado, hard sandstone, between coal banks; Point
of Rocks, Wyoming {Dr. F. V. Haydev). It was first described from Van-
couver specimens.

Sabalitcs Caiiipbcllii, Newby.

Sahal CamphelUi, Newby., Notes on Extinct Floras, p. 41, pi. x, (ined.).

Leaf very l.irge, eight to ten feet in diameter, with fifty to eighty folds; petiole long, one and a
half to two iuclics wide, flat above, without a central keel above or below, unarmed ; nerves numerous
and fine, about fifty in each fold ; six principal ones on each side of the midrib, with three intermediate
ones between each pair, the middle one being strongest.

As this species has been carefully described and finely figured by its
author, and as his plates may soon be published, I have abstained from repre-
senting it in this memoir. Moreover, the very numerous specimens which I
refer to this species, from a comparison with a fine one from Yellowstone,
labeled by Dr. Newberry, and furnished me by Dr. F. V. Hayden, merely
represent fragments of rays or rachis, and are not all perfectly concording to
the above description. The more striking characters of this species are: the
sharp folds of the rays, with keel deeply, narrowly grooved, and their upper
8 T ].•

1 14 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

part also sharply aiul acutely Mcrved; the nervation obscured by a rougli
epidermis, in such a way that the primary veins arc scarcely discernible
even with the glass, and the intermediate ones totally obsolete. From the
former species, it greatly differs by the abruptly narrowed, shorter, though
acuminate point of the rachis, which is convex on the under surface, rough
in the middle, smooth only on the borders, where it is marked by parallel obtuse
strijE, three millimeters distant; by the less numerous rays more distinctly
and sharply carinate; and by the obsolete nervation. It is from this character
especially that I referred to this species, in Supplement to Annual Report,
1871, a large number of specimens from Fischer Peak, of the Raton Mount-
ains, where they were collected by Dr. Hayden's expeditions, and later by my-
self It is there very abundant, for, except one, all these specimens of Palms
of Dr. Hayden, forty-seven in number, represent it in fragments of leaves,
of petioles, of stalks, of racemes, and fruits. My own specimens are of the
same kind. As mentioned in Annual Report, 1872, p. 375, I found it at the
Gehrung's coal, near Colorado City, then abundantly at Golden, with fruits of
Palms (Annual Report, 1872, pp. 383, 391), and at Black Buttes, in the Sau-
rian bed, where I obtained even fragments of leaves glued to fragments of
bones o^ Agathaumas sylvestris (Annual Report, 1872, p. 398). In all these
specimens, the character of the nervation is the same: numerous primary
veins, ten to twenty in. each half-ray, indistinctly perceivable under the
rough epidermis, and no trace of intermediate veinlets. As seen in the
description of the fruits which I refer to this species, they have been obtained
in most of the localities where fragments of leaves were found. I have also,
referable to it, the base of a petiole ten centimeters broad, merely convex,
and rays four and a half centimeters wide, the largest which I have seen of
this species.

Habitat. — Fischer Peak, Raton Mountains, New Mexico {Dr. F. V.
Hayden); Gehrung's coal, Colorado, Golden, Black Butte, etc. Dr. Neivherry
has it from the Fort Union Lignitic, Bellingham Bay, etc.

S a b a I i t e s f r ii c t i f c r , sp. nov.
Plate XI, Figs. 3, 3 a.

Flabellariat friicHfera, Lesqx., Animal Report, 1873, p. 396.

PYond palmate ; rays numerous, from a Ion;; acuminate racliis, acutely carinate, nerved ; fruits oval-
obtuse, narrowed to a short pedicel, borne iu a. loose raceme.

The fragment of the lower part of a palmate frond seems to be refera-

DESCRIPTION OF SPECIES— PALM^. 115

ble to Sabal by the acuminate rachis, wliicli l)ears on l)otli sides numerous
rays enlarging and diverging upward. The rachis is not very distinct, but as
far as it can be seen it looks narrowed into a point, or witli an acumen, along
which very numerous rays are attached; they descend lower than the acu-
men in diverging, as if it had been long and linear. The generic relation of
this Palm is therefore uncertain. The very numerous rays, enlarging upward
from a narrow base, crowded, pressed upon each others, and folded in their
direction, as if the leaf was young and undeveloped, are sharply keeled and
distinctly nerved, though narrow; the primary nerves one millimeter apart,
have apparently two or three intermediate veinlets, which are slightly dis-
cernible. The fruits are probably referable to the same species, as they are
not only seen at the base of the leaf, but more numerous still imbedded
into the stone. Their form is like that of small obtuse spindles, attached by
short pedicels to a loose raceme. These fruits are fifteen millimeters long,
five millimeters thick, surrounded by a thin, shelly pericarp, which, more or
less distinctly and thinly ribbed lengthwise, is crossed by short wrinkles or
splits. They are comparable in form and size to those of some living species
oi' Astrocariufn, like A. Sauri, Mart., to which, however, the leaf has no relation.
Habitat. — Golden, South Mountain.

GEONOMITES, Lesqx.

Frond large, palm<ato-pinnate ; rays connected in the lower part, separating outside, joining the
rachis by thoir whole, sometiojes balf-sheathing base, obscurely carinate.

Geonomites Goldianiis, Lesqz.
TX,' Plate ^^, Fig. 9.

Palmadtes CoWioMMS, Lesqx., Annual Report, 1874, p. 311.

Leaves flabellate; rays flat and without carina, joined in an acute angle of divergence, and, by
their whole base, to a narrow, linear, flat rachis, with narrow furrows marking line of separat ion ; pri-
mary veins generally distinct, with ten intermediate veinlets, sometimes discernible with naked eyes.

The leaf, represented by one fragmentary specimen only, is subcoria-
ceous, the surface being covered by a thick, pellucid, and shining epidermis,
■ which, destroyed in some places, leaves the nervation quite distinct. The
primary veins, however, are visible through the epidermis. The rjiys average
one and a half centimeters broad, are united, their borders being indicated
by a narrow furrow ; fiat, joining the rachis by their whole base, neither nar-
rowed nor decurrent to it; their angle of divergence being about 20°. The
primary nerves, two and a half millimeters distant, have generally ten inter-

11(5 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

mediate veinlcts, wliich, thougli ve4y close, are clearly seen where the epi-
dermis is destroyed. The flat, narrow rachis, apparently linear, is five milli-
meters wide, nerved, the middle veins being stronger than the two secondary
ones on each side of it. This part of frond is closely allied by its characters
to Flahellaria Zinkenii?, Heer, described above from specimens found at the
same locality. As no account can be taken of the rachis, which was not
observable in F. Zinkenii, we have only the rays for point of comparison.
In the present species, the rays are quite flat, broader, all connected ; the
primary veins are distinct, but neither prominent above the surface, nor as
thick; and, discernible through the epidermis, they are more distant, and
have generally ten intermediate veinlets, visible only where the epidermis is
removed. This fragment therefore represents a distinct and peculiar species.
Its relation is to Ludoviopsis geonom<zfolia, Sap. (S(iz. Fl., p. 339, pi. iv,
figs. 1, 3). In considering fig. 1 of this flora as a young leaf, whose narrow
linear rachis is indicated from above the petiole, and fig. 3, regarded by the
author as pertaining to another species, L. discerpta^ as a fragment of an old
leaf of the same, both united together would show us some of the characters
which have been described above, viz, beside the rachis, the narrow flat rays,
which in their conjunction join the rachis by their whole base in a very acute
angle of divergence. The primary nervation of the fragment (fig. 3) is, it
seems, the same as in the American species; it is, however, too obscurely
indicated for a reliable comparison.

Habitat. — Golden, Colorado; South Mountain.

G c o II o HI i t c s S c li i ni p e r i , sp. nov.

Plate X, Fig. 1.

Frond largo ; rays flat, obtusely carinate, balf-sheathing or decurront to a narrow rachis, obscurely
nerved; primary veins thick or inflated ; intermediate nerves few, three or four.

The specimen represents a portion of a splendid, evidently long and
large leaf. The rachis is narrow, not more than four millimeters thick at the
lowest part of the specimen, very slowly decreasing in size, as, at the top of
the specimen, which is twenty-two centimeters long, it still measures three
millimeters. It appears half-round and striate, but is only distinct at a few
places, the rays apparently covering it by their decurrent or half-sheathing
base. The rays are scarcely narrowed at the base, enlarging a little above
it, and connected by their borders, thence decreasing upward, soon disjointed,
and thus palmato-pinnate; they average in width two and a half centimeters

DESCEirXION OF srEClES— PALMiE. 117

a little above the racbis. Marked in tbc middle by a deep midrib, and thus
subcarinate, they show on both sides of it ten to fourteen inflatcul primary
veins, witli few intermediate veinlets, two to four discernible only, and even
rarely, atler abrasion of the epidermis. The substance of the fronds is
thin, membranaceous, of a dull red color, a character which may be casual.

I do not know of any fossil species of Palms to which this one may l)e
compared. Flahellaria longirachis, Ung. (Iconog., p. 19, pi. viii, ix, fig. 1),
has a very long and narrow rachis, but its smooth surface, as well as the
characters of the palmate rays, are far different.

Habitat. — Divide between sources of Snake River and the southern

shores of Yellowstone Lake, with Gymnogramma Haydenii {Dr. F. V.

Hayden).

Geouomitcs tcuuiracliis, sp. nov.

Plato XI, Fig. 1.

Flahellaria loiirjirachuf, Dug., Lesqx., Annual Report, 1873, p. 396.

Frond elongated, apparently linear in outliuo ; rachis very narrow, grooved in the middle ; rays
joining the racbis by a decurreut base, obtusely carinate; nervation obsolete.

The only specimen seen of tliis species is figured. It appears to repre-
sent the upper part of along, linear-lanceolate frond, palmato-pinnate, with a
very narrow rachis, to which the rays are attached in an acute angle of
divergence, scarcely 20°. The rachis is about two millimeters thick, smooth,
and grooved in the middle. The rays, obtusely carinate, narrow, about one
centimeter wide, including both faces, become flat and slightly decurrent
toward the rachis, curve inward in narrowing, and seem to become free or
cut from each other toward their points. The substance is thick and coarse,
the nervation nearly totally obsolete, except where the rays, destroyed by
maceration, have left indistinct traces of nerves, as marked upon the right
side of the figure.

I referred, with doubt, this form to Flabellaria longirachis, Ung. {loc. ci(.),
from the size of the rays, their obtusely carinate and rough surface, together
with the obsolete nervation. But in Unger's two figures, whicli, fine as tliey
arc, show more than is remarked in the description, tlie rachis is of a differ-
ent character, the rays being half-cylindrical, very long and linear, connected
in their whole length. Our too small specimens may, however, be the point
of a long frond, whose base would be represented by l)oth Unger's si>eci-
mens. In this case, however, as in otiiers, where identity with European

118 U^'1TED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

species is doubtful, it is advisable to use different names in consideration of

suggestive modifications of characters by influence of geographical distribution.

Habitat. — Raton Mountains, near Fischer's Peak {Dr. F. V. Hnydcn).

Gconomites Ungeri, sp. nov.

Plate XI, Fig. 8.

Frond large, flabellate or flabollato-piniiate(?) ; r.ays numerous, undivided, half-round, narrow,
joining \>y tbeir wholo base a broad, nerved rachis.

The fragment is comparatively small, but some of the characters of
the leaf which it represents are clearly defined. The frond was a very large
one, as seen from its broad rachis, which, though broken in its length, is still
two centimeters at its base, apparently gradually decreasing upward, and
distinctly striate, at least toward the base, where it is somewhat concave.
The leaf accordingly seems to have been broadly linear-lanceolate. The
numerous narrow, inflated rays join the rachis by their whole base, neither
narrowing nor decurring to it, passing up in an acute angle of divergence,
25° to 30°, and slightly curving inward. They are nearly linear, five milli-
meters at the base, flat underneath, as seen in the upper part of the specimen,
which is merely a counterpart of the underside of the rays, and here marked
by nerves about two millimeters distant, with four or five thin intermediate
veinlets. The veinlets upon the petiole are of the same kind, but without
primary nerves.

As far as can be seen, this species is distantly related to Manicaria
formosa, Heer (Fl. Tert. Helvet., i, p. 92, pi. xviii), where is figured a
splendid specimen, whose rachis is, however, totally destroyed. From the
distance of the rays at the base, this rachis seems to have been broad.
The relation is especially in the size, the inside curve, and the nervation of
the rays, which, however, in our specimen, join the rachis at a more acute
angle of divergence than in Heer's species, are half-round above, the line of
separation being marked by a deep groove. In considering the generic rela-
tion of his species. Prof Pleer remarks that it is not referable to Geonoma,
whose fronds have a narrow rachis, with rays in an acute angle of divergence.
This American species therefore would be related to Manicaria by its broad
rachis only, but differ from it by the more acute angle of direction of its half-
cylindrical rays, which relates it to Geonoma.

Habitat. — Raton Mountains, New Mexico {Dr. F. V. Hayden). The
same specimen bears the leaf described as Ficus Smithsoniana.

DESGKirTION OF srECIES— PALM^. 119

PALMOCARPON, Lesqx,

Fruits of variouH size and forms, gCDerally sunouiideil by a shelly pericarp, and found in connec-
tion with reniaiuBoC I'aliiis.

Paliiiocarpoii composituiu, Lesqz.

Plate XI, Fig. 4.

Catpoliihcs comjweitus, Lesqx., Supplement to Annual Keport, 1871, p. 16.

Fruits oval, obtusely poiuted, narrowed to the base, where they are joined five together, striate in
the length.

The specimen represents a fragment of a short pedicel, to which tive oval
nutlets are attached close together, the upper ones larger and apparently
crushed, the middle ones oval, obtusely pointed, one and a half centimeters
long, seven millimeters thick in the middle, the lower one smaller, all dis-
tinctly striate in the length. This species seems related to the fruits of
Sabal ? fructifera (fig 3 of the same plate). . The upper part of the speci-
men is, however, crushed. Their union to a short pedicel of a close raceme
relates them to the fruits of some Palms, as seen when enveloped in their
spathe.

Habitat. — Placi^re Mountain, New Mexico {Dr. F. V. Hayden).

P a I ni ocarpou jnexicaiium, Iieeqz.

Plate XI, Fig. 5.

CarpolHhes Mexicavua, Lesqx., Supplement to Annual Report, 1871, p. 17.

Fruit rounded on one side, rapidly narrowed to a point, surrounded by a shelly envelope, smooth
or without distinct strise except a few near the point.

This fruit is broadly ovate, pointed or round on one side, narrowed to a
truncate point, twenty-five millimeters long and sixteen millimeters across
the middle; its surface is smooth, without striae, but with a few irregular
splits, which show a thin shelly pericarp.

This fruit is comparable to those of many species of Palms, especially
those of Astrocaryum and Badris, like Bactris inacrocarjm, Wall., Astrocaryum
acaule, Mart., etc.

Habitat. — Same as the former, with fragments of large rays of Sabal?,
which measure four centimeters across {Dr. F. V. Hayden).
P a I ni o V a r p u n commune, Lesqx.
Plato XIII, Figs. 4-7.

CarpoUthes palmamm, Lesqx., Supplement to Annual Report, 1871, p. 13 (in part); Annual Report, 1873,
pp. 382, 3D8.

Fruits large, orbicular when surrounded with the shelly exocarp, round-oval, slightly truncate on
one end, broadly, obscurely pointed at the other, and very minutely and indistinctly vcine<l when this
envelope is destroyed.

These fruits, about three centimeters in diameter, and nearly globular

120 UNITED STATES GEOLOGICAL SUEVEY— TERTIARY FLORA.

when covered with the outer envelope, are broadly oval and a little smaller,
two and a half centimeters long, and two centimeters across when deprived
of tlieir exocarp. This shelly covering is thin, straw-colored or yellowish,
smooth, and easily crushed, as represented in figs. 4 and 5; the endocarp
appears also thin and, like the kernel, soft and easily yielding to compression;
therefore few of these fruits are preserved in their original form. As seen in
figs. 6 and 7, the endocarp is very thinly lined in the length, a character
w^iich is remarked only with the glass; and on one side they are indistinctly
marked by scars resembling the point of a chalaza with the raphis and the
hilum, as seen on the endocarp of some seeds.

At first, considering only the specimens of the Raton, deprived of their
exocarp, and comparing them to the description of Carpolithes lineatm,
Newby., in Notes on the Later Extinct Floras, I supposed that these species
were perhaps identical. But after the examination of a number of fruits
found at Evanston (represented in pi. Ix, figs. 1 to 1 d), which are more
distinctly related to Dr. Newberry's species, those described above were
perforce recognized as referable to far different kinds of vegetables. They
pertain to Palms by their form, their size, their double shelly pericarp, and,
indeed', resemble the fruits of some- species of Iriartea, I. setigera, Mart., for
example; or of Leopoldtna, like L. pulchra. Mart. They are found very
numerous in certain localities of limited areas, as if they were derived from
a common support or a raceme. At least, the specimens of Golden and of
the Ratoh are mixed with cylindrical fragments, like branches and branchlets,
or clustered peduncles half destroyed by maceration, and to which these fruits
seem to have been originally attached; even with pieces of textile filaments,
like remnants of decomposed spathes.

Habitat. — Raton Mountains {Dr. F. V. Hayden) ; Golden, very common,
with fragments of Sahal; Black Butte, above main coal.

Palinocarpon trnncatum, sp. nov.
Plate XI, Figs. 6-9.

FruitB Bubglobular, slightly flattened, truncate on one side, covered with a brownish, smooth,
eholly envelope.

In regard to their size, these fruits represent two species. As their
characters are, however, the same, and as the difference in size may result
from the relative position upon racemes of the same species but of different
age, I have considered them as varieties: var. major and var. minor.

DESCllIPTION OF SrECIES— PALM^ 121

The largest specimen of var. major (fig. 7) measures seventeen milli-
meters across, from side to sidc^ and ten millimeters only from the truncate
base to the top; one specimen, which seems intermediate between both varie-
ties, is only fourteen millimeters broad and eight millimeters high. The
shelly pericarp, of a darker color and smooth, is perhaps slightly stronger tlum
in the former species; nevertheless, the specimens are generally crushed and
flattened by compression. The var. minor (figs. 8 and 9) is represented by
nutlets one centimeter across the widest part and five to six millimeters in
the other direction, generally in a better state of preservation, but with the
same thin pericarp of the same color. The characters of these small nuts
refer them to Sabal, some species of which have racemes of fruits of the
same form, surrounded with an outer envelope easily crushed and of soft
texture. Figs. 8 and 9, for example, closely resemble the seeds of Sahal
Mexlcana, Mart. The connection of these fruits with fragments of leaves of
Sabal, especially of S. Campbeliii, seems also to point out their reference to

species of this genus.

Habitat.— Golden, Colorado; not as commonly found as the former.

P a I III o c a r |> o II c o r r it g a t u ni, sp. nov.
Plate XI, Figs. 10 and 11.
Fruit hard, enlarged iu the middle, truncate at one end, slightly narrowed aud obtuse at the
other, ribbed iu the leuffth, deeply rugose across.

This kind of nutlet seems, at least in fig. 11, to have its pericarp half
destroyed l>y maceration; for fig. 10, which has the same characters, differing
merely by its smaller size, has the epicarp smooth. The one (fig. 1 1) is marked
in the length by ten narrow costaj, and across by deep irregular wrinkles.
It is one and a half centimeters in diameter, and from the exact similarity
of form I consider it as representing the same species as fig. 10. The rela-
tion, however, of these fruits to living species of Palms is as yet uncertain.

Habitat.— Golden, Colorado; with the former.

P a I in o c a r p o n s ii b c y I i ii d r i c ii iii , sp. nov.

Plato XI, Fig. 12.

Fruit oblong or subcylindrical, truncate at ouo end, split at the other iu two diverging or slightly
recurved, poiuted lobes, distantly and obscurely veined toward the base.

As seen from the figures, these fruits vary in size from ten to fifteen
millimeters broad in the middle, though generally of about the same length,
or two centimeters from the border of the truncate base to the points of the

122 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

lobes, whicli, in all the specimens obtained, are flattened. A number of these
arc crushed and disfigured by compression, and the splitting at the top might
therefore be considered as casual or a 'i-esult of mere mechanical agency.
But I have carefully detached from the matrix some well-preserved nutlets,
like those of fig. 12, and all have the top flattened and split, though they
are cylindrical from above the middle to the base. I therefore consider this
character as original, and indeed, but for this, these fruits should be referable to
Oaks, their form being like that of species of acorns of the present flora; or to
nutlets of A^elu?nbiiim, the Sacred Bean, which they also resemble. I believe,
however, that they are referable to Palms. Their pericarp is of the same
color and consistence as that of Palmocarpon commune; it is in the same way
obscurely lined toward the truncate base as in P. truncatum, and the fruits are
found mixed with the specimens of these two species, and also in connection
with fragments of Palm or Sabal leaves. The truncate base is a character
of the fruits of some species of Sabal.

Habitat. — Golden, Colorado ; Table Mountain.

DICOTYLEDONES.

Some of the fossii species of this division of the vegetable kingdom have
been found in Europe with leaves and fruits in connection with stems and
branches, even with flowers. In a few cases, therefore, an exact determina-
tion of these plants has become possible. But generally, and with scarcely
any exception in this country, the fossil species of dicotyledonous plants
are represented by their leaves only, and therefore their determination is
subject to a degree of uncertainty. The leaves, however, afford distinct, even
specific characters, by their form, their thickness, or consistence, especially
by their nervation. A number of botanists of celebrity — A. P. De Candolle,
Leopold de Buch, d'Ettingshausen, Heer, etc. — have attempted to determine
by rules the essential characters of the nervation of the dicotyledonous leaves,
and to represent them under peculiar names, in order to facilitate researches
and the study of palaeontology. Though the laws governing the distribution
of the veins in the leaves in comparison to their forms have not yet been fully
discovered, we have, in the description of our fossil leaves, to rely on and to
describe all the characters which render them identifiable, and therefore to
use all the materials obtainable for that purpose. The terminology serviceable
for the description of the forms of the leaves is generally known by botanists,

DESCRIPTION OF SPECIES— DICOTYLEDONOUS LEAVES. 123

or is found explained in text-books, especially in Gray's Lessons in Botany.
But that which relates to nervation forms a separate section, considered only
in works on vegetable palaeontology, rarely accessible to the student. I there-
fore give herewith an abridged explanation of the terms which may be used
ill describing the American Tertiary plants in this memoir. For some authors,
as for d'Ettingshausen, who has given to the subject a careful and long study,
the groups fixed by the characters of the nervation are numerous, and their
distinction is somewhat embarrassing in many cases. Heer's classification is
by far the simplest, and more comprehensible. It has been admitted by
many European authors, especially by Schimper in his Paldontologie V(ig^tale.
According to it, we have : —

1. Leaves penninerved. — These have a midrib, or primary nerve,
brandling on each side. These branches, the secondary nerves, have to be
considered in their position, as alternate or opposite, in their respective dis-
tance, and especially in their angle of divergence, from the middle nerve and
their direction toward the borders. Their branches are tertiary nerves, or
veins, except when they join the secondary nerves, either immediately or by
subdivisions, when they become nervilles or veinlets. When these veinlets
are in right angle to the secondary or tertiary nerves, and pass across them,
they are percurrent nerviUes; when they are curved, or broken, or connected
with veinlets of the same order, they become ivjlected or broken nerviUes.

According to the distribution of the nerves, the surface of the leaves is
divided in are^s of different orders. Those which are limited on one side by
the primary nerves, and fill the space between two secondary ones, are named
areas of the first order. Those which are surrounded by secondary and ter-
tiary veins are areas of the second order. When the veinlets divide in thinner
branches, they surround areola, which may be subdivided by descriptions in
areola or meshes of the second, third, fourth order, etc. Sometimes the primary
areas are traversed, as in the leaves of Willows, by shorter secondary nerves,
which soon divide into areolae or join by their branches the secondary nerves
of tiie first order. These are generally named tertiary nerves, or shortened
secondary, or pseudo-secondary nerves. The areas of the different orders are
often clearly defined, as in the leaves of the Maple, for example ; but some-
times they gradually become undiscernible by the thinning of the veinlets,
and lose themselves in the netting. Li this case, the ultimate <livisions are
called dissolved nervilles.

124 U2^ITED STATES GEOLOGICxYL SURVEY— TERTIARY FLOEA.

Considering the secondary veins in relation to tlieir mode of expansion
into the leaves, they are named : —

1. CraspeAodrome, when they reach and penetrate the borders in passing
up from the middle nerve. The leaves with a nervation of this kind arc more
generally dentate or lobed. These nerves are either simply craspedvdrome,
when they join the borders without branching, as in the Chestnut, the Beecli,
etc., or doubly craspedodrome, when they branch and join the borders with all
their divisions or some of them only, as in the Birch, the Hazel, the Elm, etc.

2. Caiyiptodrome, when the primary areas which they limit do not reach
to the borders. Then the nerves curve inside of the borders, and thus they
inclose the outside borders of these areas. The secondary nerves either curve
quite near the borders, and the veinlets which go out of the flexure imme-
diately enter the margin, or the primary areas do not extend so far, the
secondary veins curving and becoming united nearer to the midrib, and thus
a series of secondary areas is formed between the borders and the bend of the
secondary nerves by successive divisions ; these are named marginal areas,
and, when in many rows, may be distinguished as inner, middle, and outside
areas. They may be named also, in considering especially the nerves, festoons,
or hows, single or multiple, with the same appellation for their range, except
for the border area, which is generally called marginal. When, by anasto-
moses, the veins form a series of meshes, which, distinct, decrease in size from
the lower part of the areas to the borders, this nervation is named brachio-
dronie.

3. Marginal nerves are those which, going out from the midrib at the
base of the leaves, follow close to the borders and parallel to them. They are
not secondary nerves, for these, passing up along and from the midrib, join
the marginal veins along the borders; they form a true border line, which
generally curves inside at the point of union to the secondary veins. This
nervation characterizes a large number of leaves, especially of the Myrtacece,
the Proteacece, the MtjricacecE, etc. A fine representation of it is seen in
Myrica Torreyi (pi. xvi, figs. 4, 8, and 9, of this volume).

4. Mixed nerves are seen upon leaves where the secondary nerves are on
one part craspedodrome, while on the other they are camptodrome. This is
often seen in lobatc leaves, where some of the secondary nerves ascend to
the point of the lobes, while, above it or under it, others curve along the
borders, as in the leaves of the Tulip tree, for example.

DESCItlPTIOX OF SPECIES— DICOTYLEDONOUS LEAVES. 125

5. A hyphodrome nervation is that of thick, coriaceous leaves, whoso mid-
rib only is strong and distinct, while the secondary nervation is indiscernible.
In this case, the nervation is about like the reticulate one, which, however,
has the second and tertiary veins about of equal thickness, distinct, united,
forming a prominent netting.

G. The aerodrome nervation is that of the leaves with secondary nerves,
all at an acute angle of divergence, curving along the borders, the upper ones
passing up to the point of the leaves. Of this kind there are two divisions : —

a. When all the secondary nerves are of the same order, witli the upper
pair only ascending to the point of the leaves, as in the Dogwood ;

h. When the two lower secondary nerves are stronger than those above
them, ascend to the point, and branch outside, as in Ccanothus Americanus,
the common Jersey Tea.

This subdivision makes a transition to the triple-nerved leaves.

2. Leaves palmate-neeved. — In this division, the primary nerves, three,
five, even seven, go out together from the base of the leaves, or from a short
distance above it, generally diverging fan-like, either all of the same thick-
ness or the lateral ones of different thickness, as in some Poplar leaves. With
palmate leaves, the areas are described as for the penninerved ones, the
primary areas being limited between the primary and secondary nerves on
both sides, and the name of cardinal areas being given to the large ones
inclosed between the primary nerves on two sides and the borders on the
other. The palmate-nerved leaves are subdivided also according to the char-
acters of their nervation in —

1. Craspedodrome, with primary nerves going out to the borders : —

a. With secondary and tertiary nerves running also to the borders ;

b. With secondary nerves camptodrome ;

c. With the same in part camptodrome, in part craspedodrome (mixed).

2. Camptodrome, when the midrib only reaches the borders, and all the
others curve along them, as in the Poplar.

3. Aerodrome, when t he lateral primary nerves ascend to the point of the
leaves.

3. Peltate-nerved leaves. — They have their primary nerves radiating
all around the top of the petiole, which may be cither central to the leaves,
or on one side of them. The subdivisions have the same characters as those
remarked above.

126 UNITED STATES GEOLOGICAL SURVEY— TESTIAUY FLORA.

APETALyE.

AMENTACE^.

MYRICACE^.

MYRICA, Linn.

This genus is now represented by about forty species, distributed over
the whole world, with the exception of Australia. The continent of America
has the largest number. Besides seven species belonging to the flora of the
United States, ten others inhabit Mexico and the Pacific coast as far south
as Peru. In Africa, the Cape of Good Hope has nine and Abyssinia one.
Asia has six species between the tropic and the equator, two species being
ascril)ed to Java. Cuba has two, Japan one, and Europe one, Myrica gale,
indigenous also in this country. A peculiar section of the Myriccs, Comp-
tonia, which had many representatives in the Tertiary floras, has only one spe-
cies living, M. Comptonia, D. C, the well-known Sweet Fern, which exclusively
belongs to the Atlantic slope of North America. The present distribution
is worth remarking in comparison to what is known of the genus in the geo-
logical times, in order to come, if possible, to the solution of the question
referable to the origin or dispersion of vegetable types.

The Cretaceous flora of North America has two species of Myrica recog-
nized by their leaves, and one by seeds, the latter referable either to one of the
two first species or representing a third. In Europe, two species also have been
described by Heer from the Quadersandstein of Quedlinburg, Germany, and
a third from the Upper Cretaceous of Greenland. Therefore the origin of
the genus is positively recorded in the flora of the Cretaceous, and at this
period its presence is about equally indicated in Europe and in this country.
But ascending higher in the geological ages, we find forty species of Myrica
described from the Lower Tertiary of Europe, half of them in the Eocene,
the other part in the Paleocene, and more than forty species in the Miocene.
From the Eocene, or Lower Lignitic, of the United States, we know as yet
only two species of this genus, one of which is still of doubtful reference;
and from the groups of Carbon and Evanston, considered as Upper Eocene
and Miocene, we have none. It is only in the fourth group, that of the
Parks and of Green River, that we find the genus represented by a propor-
tionally large number of species, twelve, six of which are referable to the
section Comptonia, -whwh/^n the European Tertiary, has twenty-three species.

DESCRIPTION OF SPECIES— MYKICAOEiE. 127

Except for tliis last group, therefore, there is no rclalion whatever in the
distribution of species of Mijrka in both the Tertiary floras of Europe and of
America; and if we are able to trace on this continent some analogy of
distribution in the section of the Comptonice o{ the Upper Tertiary and of the
present flora, there is evidently none of this kind in Europe.

Though the leaves of Myrica seem distinctly characterized, some of them,
nevertheless, have in their form and nervation a degree of relation to those
o( Froteacece, of the genera Dryandra and Lomatia especially. Some pale-
ontologists, d'Ettingshausen above all, have described a number of these leaves
under the generic name of Dryandroidcs and Lomatites. Since then the
seeds of some of those supposed Froteacecs have been found, and their refer-
ence to Myrica is positive; and now the opinion is prevalent that, if not all,
at least the largest number of these species belong to Myi'ica. Saporta, in
considering this question in a geological point of view, hypothetically demands
whether, as the Myricce and the FroteaeccB have originated in the Cretaceous,
and as there is now a kind of parallelism in some of their characters, seem-
ingly recording a common birth, they could not have been at the outset iden-
tical in their generic characters, and have branched through the geological
times into those two divisions, now so widely separated by some of their
more important botanical characters, as by their geographical distribution.
The celebrated author remarks on this subject that in the Cretaceous flora
of Belgium, Debey and d'Ettingshausen find the Myricce. with the Froteacece,
the first, however, in a very subordinate number, and that, from his own
observations, the relative position of both these groups is remarked to be the
same as high as the Gypses of Aix, when the Froteacecu began to decline and
the Myrica to take the ascendency. On this subject, the little we know of
the recent geological floras of this continent does not give any evidence for the
solution of hypotheses of this kind. The Froteacece of the Dakota group
(Cretaceous), Froteoides of Heer, are too indistinct in their characters to
afford a reliable point of comparison. These leaves, known especially by
their coriaceous consistence and their form, for the nervation is scarcely dis-
cernible, may represent a single species of plants of a group far different from
that to which they are hypothetically ascribed. At any rate, in passing up
to the Tertiary, we do not find any leaf comparal^le to any Froteacece except
in the Lower Yiocene, Myrica Torrey?', which resembles a Lomatia liy its
nervation, but with which seeds of Myrica have been found ; and, in tlie

128 UNITED STATES GEOLOGICAL SUKVEY— TEKTIAKY FLORA.

Upper Miocene, or fourth group, a very small leaf described as Lomatia
micropJiylla, which even may belong to a species of Myrica. I have there-
fore until now considered as very doubtful the presence of true Proteacete in
the geological floras of tliis continent, and for this reason I do not find any
fact in support of the hypothesis of a common origin and parallel develop-
ment of the MyriccB and the ProteacecB. On the contrary, I believe that the
deeper we go into the study of the fossil plants of North America, and
the more intimately we become acquainted with the characters of the ances-
tors of our present flora, the greater will we find to be the analogy of the old
types with the present ones, and that we shall see gradually disappear those
foreign types bearing tlie so-called Austro-Indian and Australian characters.
On the present subject, it seems established: that in Europe i\ie Myricce
take a gradual and considerable importance during the whole Tertiary period,
and then nearly entirely disappear from the flora of that continent. Until
now, no species of this genus is described from the Pliocene. And per
contra: tliat in North America the representation of this group of plants is
scanty from the Cretaceous up to the Upper Miocene, where it takes that higher
degree of predominance which is preserved, it seems, in our present flora.

I have formerly considered as referable to 3Iyrica a fragment of a leaf,
remarkable indeed for its size, as a representative of a genus whose leaves
now are all of a comparatively small size. This leaf is by its form and ner-
vation in exact concordance of characters with those of Co?)iptonia, as seen
from pi. Ixiv, fi^. 1. The fragment is already much larger than Ung'er's
Myrica grandiflora, considered till now as the giant of the genus, and there-
fore its relation is contested on that ground. Now, the discoverer of this
fine specimen, taken from a shaft sixty feet deep, asserts that he has seen at
the same place leaves of the same kind of a flir larger size, some of them
about ten feet in length. If so, we should have to look to another group of
plants for the relation of these leaves. I must, however, remark that, con-
trary to what is asserted by some paleontologists, the Tertiary leaves are
generally of a diminutive size in comparison witli those of the flora of our
time. The size seems greatly exaggerated to the eyes by the fossilization,
or the flat position of the leaves upon stones, and is therefore considerably
reduced by measurements. The largest fossil dicotyledonous leaf wliich I
have ever seen is that of an Aralia, from the Upper Miocene of California,
appearing indeed of a monstrous size. It measures twenty-seven centimeters

DESCRIPTION OF SPECIES— MyRIOACEJD. 129

across the point of the outer lobes. Now, I have a loaf of Aralia papijracea
of Jajjan, which closely resembles tlie fossil species, and wliicii 1 selected
of the smallest possible size, in order to have it in herbarium. It measures
fifty-eight centimeters across, also from the point of the lobes. The largest
leaves of Platanus nobilis, twenty centimeters broad, and of P. Haydenii,
Newby., sixteen centimeters across the lobes, are of the largest and finest
obtained fossil specimens. Leaves of our Platanus occidentalism thirty centi-
meters broad, are not rare. Of living Magnolia, the leaves measure as high
as thirty centimeters long and fourteen centimeters broad ; the Papavv, when
growing under the shade, has still larger leaves, thirty-two centimeters by
fifteen, and so on. As yet I have not seen, except for this Myrica, any fossil
leaves of a size superior to those of the present flora in comparing speci-
mens of the same generic type. This case would therefore look like an
anomaly, and suggest the probability of a different relation. I should have ad-
mitted this view, if, in the original determination of this species, I had not had
for point of comparison Myrica {Comptonia) grandifolia, Ung. (Fl. of Sotzka,
p. 31, pi. viii, fig 1), whose preserved fragment, the upper part of a leaf only,
measures fourteen centimeters long and five centimeters broad, while a whole
and large leaf of the living Comptonia asplenoides is only seven centimeters
long and one centimeter broad. As tliere are some species of the same group
with leaves intermediate in size, M. hakecefoUa, Sap., M. insignis, described
here below, etc., I cannot see in the difference of size a sufficient reason to
separate from this genus a leaf which by its other characters seems evidently

referable to it.

§ 1. — Leaves entire, undulate, or dentate.

Myrica Torreyi, Lcsqz.

Plate XVI, Figs. 3-10.

Myrica Torreyi, Lesqx., Aanual Report, 1872, p. 392. — Scbimper, Pal. V^g6t.,iii, p. 586.

Leaves luembranaceous, narrowly lanceolate, tapering np to a long linear acumen, gradually nar-
rowed to a short broad petiole, distantly dentate ; nervation marginal.

The leaves are variable in size, from four to fifteen centimeters long,
including the petiole, and from eight to thirty millimeters broad below
the middle. Gradually decreasing upward to a long narrow acumen, they
are more rapidly narrowed to a short, broad, sometimes half-winged,
petiole. Their nervation is marginal, a distinct though narrow vein fol-
lowing up quite near the borders, where it is joined by the secondary

nerves, simple or branching, and slightly bending inward at the point of
9 T F

130 UNITED STATESGEOLOGICALSDRVEY— TERTIARY FLORA.

connection. The secondary veins are united by cross-nervilles in right
angle, forming large quadrate areolae ; the ultimate nervation being obso-
lete. The borders, irregularly and obtusely dentate from a little above the
petiole, become entire from the base of the long acumen up to its point.
In the numerous specimens obtained of this species, and those examined in
situ, I have seen the acumen of the leaves preserved, *and measuring one to
three centimeters, linear, entire, gradually sharply pointed. The divergence
of the secondary veins is about 60°. By the form of the leaves and the
slightly obtuse teeth of the borders, this species is like Banksia Ungeri, Ett.,
described in Har., Fl.. p. 54, pi. xvii and xviii, as Myrica hanksiafolia, Ung.
Its nervation,however, relates it very intimately to Lomatia latior, Heer, Bait.
Fl. p. 80, pi. xxiv, fig. 16. The fragment figured by Heer is regretably too
small for a close comparison. It seems, however, remarkably like our fig. 9,
base of fig. 8, both by the nervation and the denticulation of the borders.
Species of the same type are described from Sotzka and Mount Balca, or from
the Eocene. I cannot consider these leaves as referable to a species of
Lomatia, for the reason that the shaly sandstone where they abound have also
numerous scattered seeds and scales like those of Myrica. They are figured
in pi. Ix, fig. 8, and described hereafter as Carpolithes myricarum.

Habhat. — Black Buttes, Wyoming Territory, in soft sandstone above
main coal; also in the burned red shale top of the hills.

myrica acuminata, Uug.
Plate XVII, Figs. 1-4.

Myrica acuminata, Ung., Flor. of Sotzka, p. 30, pi. vi, figs. 5-10 ; pi. vii, fig. 9.— Heer, Fl. Foss. Arct., i, p.

102, pi. iv, figs. 14-16; pi. vii, fig. 6 i ; Mioc. Bait. Fl., p. 33, pi. vii, fig. 1; Fl. v. Bornst.,

p. 13, pi. ii, fig. 1. — Lesqx., Annual Report, 1873, p. 411.
Dryandroides acuminata, Ett., Proteac. d. Vorw., p. 32.— Heer, Fl. Tert. Helv.. ii, p. 103, pi. xcix, figs. 17-21 ;

pi. c, figs. 1-2.

Leaves coriaceous, shining, linear-lanceolate, acaminate, irregularly dentate ; nervation obsolete.

Variable in size, from four to eight centimeters long, and comparatively
narrow, these leaves are linear-lanceolate, gradually tapering upward and
acuminate, more rapidly narrowed to a short petiole, irregularly sharply
dentate on the borders. Their consistence is coriaceous, the surface shining,
and the nervation obsolete, at least toward the borders, where the veins
seem directed to the teeth, but not entering them, linger says of the leaves
of his species: serraturis equalibus, minimis, approximatis, a character which
does not seem £o be in accordance with what is represented in figs. 1 and

DESCRIPTION OF SPECIES— MYRICACE^. 131

4 of our plate, however agreeing with the denticulation of the leaves in figs.
2 and 3. This appears to be a mere variation; for the specimens sent for
determination are numerous, and show many differences in the forms of the
leaves and the denticulation of the borders, even upon the same leaf; one of
them, for example, represents a long acuminate leaf, with equally serrate
borders on one side, while on the other the teeth are close and unequal. As
the specimens from one of the localities mentioned below have much smaller
leaves, shorter and narrower than those described from Europe, ours may
represent a different species. Saporta considers it as rather allied to M. ar-
guta, Heer, and M. Zacchariensis, Sap., both Miocene.

Habitat. — Middle Park, Colorado {Dr. F. V. Hayden). One mile west
of Florissant, Colorado {Dr. A. C. Peale). Mouth of White River, Utah
{Prof. W. Denton).

iTIyrica Copeana, Lesqz.

"piatc XVII, Fig. 5.
Mjjrica Copeana, Lesqx., Annual Report, 1873, p. 411.

Leaves laDceolate, taper-pointed, rounded to tlio base, deeply doubly serrate ; nervation penni-
nerve, craspedodronie.

A fine leaf, of which there is only one specimen, about eleven centime-
ters long (the point and base are destroyed), three and a half centimeters
broad below the middle, lanceolate, gradually acuminate, rounded or truncate
to the base, doubly serrate, with sharp long teeth turned up, and alternate
smaller ones, the first entered by the secondary veins, the others by their
branches or hy thinner shortened secondary ones. The lateral nerves are,
near the base, nearly at right angle to the midrib, and some of them branch-
ing once; the upper veins at a more acute angle of divergence are all simple,
and the border teeth become simple also and equal. The areolation mostly
obsolete, discernible only at a few places, where a thin coating of coaly matter
covering the surface is erased, is in small quadrangular meshes, oblique to the
secondary nerves. The consistence of the leaf is not very thick, not coria-
ceous. This species is distantly comparable to Myrica Graffil, Heer, FI.
Tert. Helv., iii, p. 176, pi. cl, figs. 19, 20.

Habitat. — Near Florissant, Colorado {Prof. E. D. Cope).

myrica undiilata}, Heer.

Plate XVII, Figs. C-8.

Myrica undalaia,Y{w.r, Fl. Tert. Helv., iii, p. 188, pi. cliii, figs. 21-22. — Lesqx., Annual Report, 187.3, p. 412.

Leaves subcoriaceous, ovate- lanceolate or liuear-lauccolate, acuminate, narrowed to a short peti-
ole; borders regularly and deeply undulate; nervation camptodrome.

Leaves variable in size, about five centimeters long, one to two centime-

132 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLOllA.

ters broad, linear-oblong or ovate, acuminate, narrowed in a curve to the short
petiole; borders more or less deeply, regularly undulate; secondary veins
curving near the borders, joining the midrib at a more or less acute angle of
divergence according to the width of the leaves, with shortened ones between
them; fibrillse slightly oblique or nearly at a right angle to the veins; areola-
tion irregularly quadrangular. From the first specimen which I had for
examination (fig. 8), I considered the leaf as positively referable to that of
Heer; but from the comparison of other and more perfect specimens, the
leaves seem to differ, not only in their more enlarged size, but in the nerva-
tion, the secondary veins being closer, more regularly parallel, separated
generally by dissolved intermediate ones, and the areolation more distinctly
quadrangular. Heer also describes his species as coriaceous. These leaves
of ours are not quite coriaceous, though somewhat thick; their identity with
the European species is therefore doubtful. In comparing M. undulata to
his M. obtusiluha, represented in pi. Ixx, fig. 10 {loc. cit.), Heer remarks, as
essential difference, the thickness of the midrib in this last species. By
this character, the leaf of our fig. 8 is identified with M. undulata, while by
its nervation, close secondary veins, and intermediate shorter ones, as also the
linear shape, it is more distinctly allied to 31. obtusiloba.
Habitat. — Elko, Nevada {Prof. E. D. Cope).

JXIyrica nigricans, Lesqx.

Plate XVII, Figs. 9-12.

Myrica nigricam, Lesqx., Sapplement to Annual Report, 1871, p. 6.

Leaves nearly sessile, alternate, oblong or linear-lauceolate, acuminate, ronnded-cnneate to the
base, obtusely dentate ; nervation caniptodrome.

The leaves of this species are much like those of the former, only nar-
rower, all narrowly lanceolate or linear-lanceolate, narrowed to a long acumen,
and about sessile or with a very short petiole about one millimeter long.
They are more or less unequal at the base, distantly obtusely dentate in the
middle, with a nervation of the same character as that of the former species,
or like that of the living Myrica CaUfornica or Myrica gale. The substance
of these leaves is not coriaceous; but all the fragments are blackened on
their surface, and apparently dotted with oil-points, as in the common Myrica
cerifcra. This color renders them easily discernible upon the yellow sofl
sandstone of the locality.

Habitat. — Green River, Wyoming, northwest of the station, above fish-
bed {Dr. F. V. Hayden).

DESCRIPTION OF srECIES— MYKICACEJE. 133

myrica Bolnndcri, Lesqz.

Plate XVU, Fig. 17.

Jlex undulata, Lesqx., Annual Report, 1873, p. 410.

Leaf oblanceolate, obtusely pointed, dentate at the top ; secondary veins camptodrome.

A single leaf, about six cenlimeters long, one and a half centimeters
broad toward the point, coriaceous, obtuse, and obtusely pointed at the top,
which is marked by a few teeth on each side, entire downward, and gradually
decreasing to a broad, wingedC!) petiole (broken); middle nerve narrow; sec-
ondary nerves at irregular distance and on an acute angle of divergence, thin,
camptodrome; areolation obsolete.

Though this leaf recalls by its form and texture species of Ilex of the
present time, some of its characters, the enlarged base of the leaf the narrow
middle nerve, the disposition of the secondary veins, seem to refer it more
probably to this genus.

Habitat. — Locality unknown. Check-number H 1295, sent with spe-
cimens from Castello's Ranch, Colorado, and apparently from the same place

(Dr. F. V. Hayden).

Myrica Liiidwif^iii Scbp.

Plate LXV, Fig. 9.

Myrica Ludwigii, Schp., Pal. V<5g^t., ii, p. 545.

Myrka longifoUa, Ludw., Palasont., viii, p. 94, pi. xxviii, figs. 8, 9; xxix, figs. 1, 3, 5, 6, 7; xxx, figs. 1,
19; Ix, fig. 15.— Lesqx., Annual Report, 1874, p. 311.

Leaves of middle size, subcoriaceous, linear-lanceolate, gradually tapering into a long entire
acumen, distantly deeply dentate in the middle ; midrib thick; secondary veins subopposite, open, par-
allel, curved in passing to the borders, camptodrome.

This leaf, of which we have only the uppor part, a little more than one-
half, seems evidently referable to the European species described from a large
number of specimens; the form of the leaf, with its long acuminate entire
point, corresponds to that of Ludwig {loc. cit., pi. xxviii, fig. 8); the midrib is
somewhat broader in the American specimen, and the secondary veins at a
slightly more acute angle of divergence; but these characters are reproduced
in the leaves figured by the German author, the broad midrib in fig. 1 of pi.
xxix, and the more oblique secondary veins in pi. xxx, fig. 1. As far as it is
seen upon our fragment, whose areolation is obsolete, the details of nervation
have the same characters, the secondary veins curving along the borders,
and the teeth being entered only by veinlets outside of the primary areas.
As remarked by Schimper, this species goes to the same group as Myrica
acuminata, etc., its nearest relation among the living species being M. cerifera,

l.M UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

L., of North America. The fragment indicates a leaf membranaceous or
rather subcoriaceous.

Habitat. — Green River gruiip, near t-he mouth of White River, Utah

{Prof. W. Denton).

§ 2. — Leaves pinnaieh/ lohed.

COMPTONIA.
niyrica latiloba, Heer, var. aciitiloba.

Plate XVII, Fig. 13.

Myrioa latiloba, Heer, Flor. Tert. Helv., iii, p. 176, pi. cl, figs. 12-15.
Myrica latiloba, var. acutiloba, Leaqx., Annual Report, 1873, p. 412.

Leaf membranaceoiis, oblong-lauceolate, cuueate at the base to a comparatively long slender
petiole, pinnately divided in short angular acute lobes; secondary veins distinct, open, mixed.

This leaf has from the base to the middle, where it is broken, the same
character as those described by Heer, it being apparently short, abruptly
narrowed to the point. As the nervation is of the same character, the lower
secondary veins very open, caraptodrome, the upper ones at a more acute
angle of divergence and entering the teeth or craspedodrome, and interme-
diate, short, very thin; secondary veins, soon dissolved into the reticulation,
there is no appreciable difference but the acute lobes. Even this may be
casual and unimportant, for in figs. 12 and 13 of Heer {loc. cit.), the lowest
lobes are mere acute teeth, and, in our figure, the upper ones are apparently
obtuse, though very short. I am the more inclined to consider this leaf as a
mere variety of the European species, that I have from the Miocene of
Oregon, John Day Valley, near Bridge Creek, a specimen of a leaf preserved
in its integrity, and exactly concording in all its characters, even the size, with
Heer's species.

Habitat. — Florissant, near Middle Park, Colorado (Dr. F. V. Hayden).

IWyrica parti ta, Lesqz.

Plate XVII, Fig. 14.

Myrica partita, Lcsqx., Annual Report, 1873, p. 413.

Leaves subcoriaceous, narrow, linear, alternately equally divided to the middle nerve into equal,
broadly lanceolate, acuminate lobes, slightly denticulate on the lower borders ; secondary veins three in
each lobe, parallel, the upper one longer, passing up to the point of the lobes, the lower ones to the teeth.

A mere fragment, too small to indicate satisfactorily the characters of
the leaf It is comparable to the species of Myrica whose type is M. O^niti-
gensis, Al. Br., which are common in the Miocene of Europe But for the
small dcnticulation of tiie borders, it would be identical with M. incisa,ljndw.
(Paleont., viii, p. 9o, pi. xxx, figs. 7-15). This ditlerence is an important

DESCRIPTION OF SPECIES— MYRIGACE^. 135

one, as most of the species of fossil Comptonia have their lobes entire. M.
Matheroniana, Sap., of the Armissau of France, has its lobes somewhat den-
ticulate toward the point.

Habitat. — Elko, Nevada {Prof. E. D. Cope).

iUyrica Br o !!(;■•■:> >'<■!« Ett

riato X\n, Kif;. 15.
Driandra nrongmarti,Ett., Foss. V\. voii HUr., ]). 55, pi. xix, fig.s. 1-26.
Myriea Ilronijniurti?, hesq^., Auuual Report, 18715, p. 418.

Luaf linear, i)iiiuately divided iuto sliort, slightly obtuse lobes; nervation obsolete; secondary
veins craspedodrome.

This fragment is still too incomplete for a satisfactory determination.
The leaf is coriaceous, the details of nervation obsolete, and by its lobate
borders it is intermediate in characters between the leaves described as Dri-
andra Brongnarti, Ett. {loc. cit.), especially like fig. 20, and those of Myrica
ophir, Uug. (El. v. Sotzka, p. 30, pi. vi, figs. 12-16).

Habitat. — Elko, Nevada (Prof. E. D. Cope).

myrica insigiiis, Lesqz.
Plate LXV, Figs. 7, 8.
ilyrica insignis, Lesqx., Auuual Report, 1874, p. 312.

Leaves large, membrauaceous, narrowly oval or oblong, acuminate, narrowed to the base, piu-
nately lobed ; lobes short, deltoid, acute, turned upward ; middle nerve thin ; secondary veins open, par-
allel, alternately passing up to ihe poiut oi the lobes or to the base of the sinuses; areolation large,
polygonal.

The two fragments of this beautiful leaf sufficiently represent its char-
acters. The size is about ten centimeters long, nearly four centimeters broad
in the middle, where the lobes are equal, divided to about one-third of the
space between the middle vein and the borders, the two upper pairs being
much shorter and longer, and the terminal one sharply acuminate, two centi-
meters long. As far as it can be seen at the base of fig. 7, the lower lobes are
rapidly diminishing in size downward, and the lowest one is narrowed down-
ward and slightly decurrent to the petiole. The nervation is perfectly dis-
tinct; the secondary veins, on an open angle of divergence of about 60°,
mostly parallel, are mixed, the principal ones passing up in a slight curve to
the point of the lobes; the others, cpiite as thick, going up to the base of the
sinuses, where they divide into two branches, curving ami anastomosing along
each border, with fibrillaj, which, broken and branching in the middle of the
areas, form large quadrate or irregularly polygonal areolae. This nervation
has the true character of that of tiie Comptonia, ])ut no fossil species offers
a point of comparison ibr this one.

Habitat. — Florissant, Colorado (Dr. F. V. Hayden).

136 UNITED STATES GEOLOGICAL SURVEY— TERTIAEY FLORA.

niyrica? Lessigii, Lesqx.
Plate LXIV, Fig. 1.
Myricaf Leaalgii, Lesqx., Annual Report, 1874, p. 312.

Leaf coriaceous, very largo, oblong iu outline, deeply pinnately lobed ; lobes opposite, ovate-
lanceolate, taper-pointed, slightly broader iu (be lower part, at an open angle of divergence, separated
to near the midrib, where they are joined in broad obtuse sinuses; middle nerve very broad ; secondary
veins proportionally thick, mixed.

If the leaf represented by the figured fragment does belong to the
Comptonia, it is indeed of an enormous size, for the preserved part, which
seems to be one-half of the leaf only, is twenty-two centimeters long, and
the lobes, from the middle nerve to the top, measure more than nine centi-
meters, showing the width of the leaves to be at least eighteen centimeters.
The midrib is very thick; the secondary nerves are of two orders: those of
the first are strong, ascend to the point of the lobes, and branch on each side;
those of the second are narrower, and come out of the middle nerve also.
They are either short, passing up to the base of the sinuses, there diverging on
each side, and following the borders in festoons, anastomosing with fibrillse,
or longer, traversing the large areas between the base of the secondary veins
and the borders of the lobes, dissolving either in branches or fibrillse, in right
angle, as in the former species, which it much resembles by the characters of
the nervation. The ultimate areolation is formed, as represented in the
middle of the lower lobe of the figure, by subdivision nearly in right angle
of the primary areolae, in a very small quadrangular or polygonal reticulation.

Though the characters of nervation are those of Comptonia, remarkably
similar indeed to those of Myrica {Comptonia) Matheroniana, Sap. (fit., ii, 2,
p. 93, pi. 5, fig. 7), beautifully represented in the enlarged figure (7 a), it
is difficult to suppose a leaf of this genus as large as the one represented
here. I have already explained what reasons induce me to describe it in this
section. The celebrated author of the Etudes, quoted above, objects to this
reference, and considers the fragment as part of a leaflet of some kind of
Araliacea, like Aralia mulfifida. Sap. (fit., i, 1, p. 115, pi. xii, fig. 1), a leaf
palmately divided nearly to the top of the petiole in nine-lobed leaflets, vary-
ing from six to twelve centimeters long. The mode of division of these
leaflets has indeed some likeness to that of our fragment, but the characters
of nervation are somewhat different. I am unable to decide the question,
from want of materials for comparison. The consistence of this leaf seems
to have been hard, thick, and membranaceous at the same time, the nervation
being clearly defined in black lines upon llx; brown color of Ihe specimen.

I have latelv received, from Rev. A. Lakes and from Golden, a number

DESCRIPTION OF SPECIES— BETDLACEiE. 137

of specimens, mostly fragments of leaves, which represent a species inter-
mediate, by its characters, the form, and the size, between this and the former.
The nervation is of Ihe same type. These fragments show such an intimate
relation between M. Lessigii and M. insignis that both appear necessarily
referable to the same generic division.

Habitat. — Coal Creek, Colorado, in clay overlying coal, reached by a
shaft sixty feet deep {Gen. W. H. Lesdg).

BETULACE^,

BETULA, Linn.

The distribution of this genus is limited at our epoch to the northern
regions of Asia, Europe, and America, a few of its species ascending to the
Arctic zone. Of the twenty-nine species described in the Prodromus of De
Candolle, eight inhabit North America, four of them exclusively belonging
to its flora. The numerous species of Betula described from the Tertiary of
Europe, thirty-nine, are especially related to the present North American
forms, as. are also the few recognized in our geological formations.

The generic type appears to have originated in the Cretaceous period;
for we have already two species described from tlie Dakota group formation:
Betuliles dentlculata, Heer; Betula beatriciam, Lesqx. The genus is repre-
sented also in the Eocene of Europe by three species, two of them in the flora
of S^zane; by four in the Paleocene, and thirty in the Miocene. Of these, of
course, a large number are uncertain, the specific determinations from leaves
only being perhaps more unreliable for this genus than for any other. In
this country, one leaf only has been found in the lignite of Golden, doubtfully
referable to Betula gracilis, Ludw., which by itself is already of uncertain
relation, the only leaf which represents it being related to Populus rather than
to Betula. Therefore we do not have as yet any positive record of this genus
in the North American Lower Eocene. It is present, however, at Evanston,
or in Upper Eocene, by two species, one of them new; also at Fort Fetter-
man, a Miocene formation, where some leaves of a new species have been
found in connection with a profusion of remains of Taxodium miocenicum.

Bctiiia Vogdesii, Lesqx.

I'lato XVII, Figs. 18, 11).

Beinla Vogdeaii, Lesqx., Annual Report, 1874, p. 312.

Leaves sm.all, thin, t.val, acnt(Oy point.-d, narrowed, and ronnded to the petiole, minutely serru-
late, penninerve ; lateral veins parallel, opposite at or near the base, simple or the lowest ones sparingly
branching, craspedodronie.

These leaves vary in size from three to four centimeters long aiwl from

138 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

sixteen to twenty-five millimeters broad in the middle; their form is nearly
oval, more enlarged, and rounded at the base, which seems to pass down
abruptly from near the petiole as decurring to it; they are minutely apiculate
and serrulate. The species is related by the nervation to B.dentkulafa, Goepp.,
(Schoss. Fl., p. 12, pi. iii, figs. 14, 15), a species considered by European
authors as identical with B. caudata, Goepp.

Habitat. — Fort Fetterman, Indian Territory {Lieut. Vogdes).

Betula gracilis?, Ludw.

Plate XVII, Fig. 20.

Betula gracilis, Ludw., Pateont., viii, p. 99, pi. xxsii, fig. 4. — Lesqx., Aniiual Keport, 1873, p. 398.

Leaf small, ovate, obtusely pointed, distantly serrate; middle nerve thick; secondary veins mixed,
some of them passing up in a curve to the teeth, simple.

The form of the leaf is ovate, apparently rounded at the base, which is
destroyed; the secondary nerves, simple and curved in passing up to the
borders, have the same character as in Ludwig's figure (loc. cit.), being, how-
ever, less distinctly camptodrome, and more generally running to the point
of the small distant obtuse teeth. The European species is already of doubt-
ful reference as remarked above, and therefore this fragment of a leaf is still
more uncertain in its determination.

Habitat. — Golden, Colorado.

Betula Ooeppcrti, Lesqz:.

Plate XVII, Figs. 21-23.

Betula caudatal, Goepp. — Lesqx., Annual Report, 1871, p. 293.

Leaves large, ovate, subcordato or rounded at the base, lanceolate-acuminate ; borders irregularly
crenato-serrate ; secondary nerves half open, subcamptodrome, thin, joined by close nervilles in right
angle.

These leaves are referable to Goeppert's species by their form, their size,
and their nervation; the acumen is also generally inclined on one side, as in
the European species. But, as remarked from a number of specimens, with
borders more distinctly preserved, they have the teeth of the borders of a
different character, not turned out and spinulose, but inclined upward and
rather obtuse, as seen in fig. 23 a, enlarged. The veins, whose angle of diver-
gence is 30° to 40°, are obsolete toward the borders, appearing either to
enter the points of the largest teeth, or to be effaced and lost in the areolation,
which is obsolete. As seen in fig. 21, the lower veins are opposite.

Habitat . Evanston, Wyoming, where Dr. A. C. Peak collected the
first specimens.

DESCRIPTION OF SPECIES— BETDLACE^. 139

Bet II I :i Steve nsoni, Lesqz.

Plato XVIII, Figs. 1-5.

Setula Stevensoni, Lesqx., Annual Rejiort, 1871, p. 21)3; 1S7-J, pp. 38G,401.

Leaves of medium size, ovate, tapering or rounded to an obtuse point, subcordate at tbe slightly
unequal base, Kliort-petioled, serrulate; nervation craspedodrome.

The leaves, varying in size from four to seven centimeters long and three
to four centimeters broad, are ovate, round-pointed, serrate, with equal short
teeth (fig. 3, enlarged); a half-round or subcordate base, and a short petiole;
the secondary veins, six to eight pairs on each side, opposite, at or near the
base, pass up to the borders nearly straight, scarcely branching under an angle
of divergence of 40°, joined nearly at right angle by strong curved nervilles,
interrupted by the veins. I found at Evanston, in connection with these
leaves, a few bracts of cones of Bctula, one of them similar to that figured
by Heer (Fl. Arct., pi. xxv, fig. 25), which the author refers to B. prisca.
Another, with three short-pointed divisions, appeared of the same character as
that of fig. 30 {loc. cit.), named Betula Forshnmmeri. One of them or perhaps
both forms may belong to our species, whose leaves are abundant at the same
locality.

Habitat. — Evanston, Utah ; Carbon, Wyoming.

ALNUS, Tournef.

As seen in the supplement to the Cretaceous Flora of Nebraska, in
Dr. llayden's xViinual Report for 1874, p. SSf), the two forms of leaves pre-
viously referred in the Cretaceous Flora, p. 62, to Alnus and Alnites are
considered by Saporta as ratlier referable to Hamcnnelis than to Alnus, and
have been accordingly described in that supplement under the generic name
of Hnmamelites. At the same time, another leaf of H. {Alnus) Kansaseanus,
found in a better state of preservation, has been represented in pi. vii, fig. 4,
of that same supplement. The reference indicated by the name of Hama-
melites is, however, quite as uncertain as that to Alnus, the more so that it is
not confirmed by paleontological records ; for, in the lowest Eocene of Point
of Rocks, a species of Abms or Alnites has been discovered, while as yet
no species of HamameUs has been found in the North American Tertiary
flora, and none also in that of Greenland. The same can l)c said of the
Tertiary flora of Europe, where one species only, doubtfully referable to

140 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

Hamamelis, is described in the Eocene Flora of Sdzane. Considering,
then, tlie geological records in regard to the present distribution of
species in the North American flora, it would be more rational to refer
to Alnus those Cretaceous leaves, and to regard the origin of this genus as
Cretaceous.

The paleontologists of Europe have to the present time described
twenty-nine species of Alnus, seven from the Lower Tertiary (Eocene and
Oligocene), and twenty-two from the Miocene formations. We have as yet
only five species referred to this genus, one from the Lower Eocene, and
tour from the Miocene; of these, one is described by Dr. Newberry from
the Fort Union group, and two have been found in the Miocene of CaHfornia
and Oregon. This apparent diflference in the distribution of this genus is
ascribable to our limited acquaintance with the North American Tertiary
floras

At the present time, fourteen species oi Alnus are known and scattered
over the boreal hemisphere, except two inhabiting the mountains from South
Mexico to Chili. Two species are predominant in Europe, one south along
the Mediterranean shores from Italy to the Caucasus, another a northern one,
which also goes eastward to Western Asia; two others are still found in
Europe, more rarely, however, and none exclusively limited to that continent.
North America has five species, two of which exclusively belong to its flora,
one from the western slope, the other from the eastern slope only. This
distril)ution is therefore in accordance with that indicated by the Tertiary
flora of this confinent, while it is the contrary for Europe, which counts
twenty-eiglit species in its Tertiary, and has none at our time exclusively
pertaining to its flora.

Alniis Kcfcrsteinii, Gopp.

Plate XVIII, Figs. G-8; Plate LXIV, Fig. 11.

Alnites Kefersteiuii, Giipp., Nov. Act. N. C, xviii, 1, p. 364, pi. xli, figs. 1-19.

Alnus Kefersteiuii, Ung., Chlor. Prolog., p. 115, pi. xxxiii, figs. 1-4. — Hcor, Fl. Tert. Helv., ii, p. 37, pi. Ixxi,

iigs. (i, 7. — LikUv., Paluiout., viii, p. 97, pi. xxxi, figs. 1-5, xxxii, figs. 1, 2. — Ett., Foss.

Fl. V. Bil., p. 47, pi. xiv, figs. 17-20.— Heer, Fl. Foss. Arct., ii, p. 146, pi. xxv, figs. 4-9; Fl.

Foss. Alask., p. 28, pi. iii, figs. 7, 8 ; Mioc. Bait. Flor., p. 67, pi. xix, figs. 1-13. — Lesqx., Annual

Keport, 1871, p. 292; 1872, pp. 386, 401, 405.

Leaves of medinni size, ovate, obtusely pointed or acuminate, rounded-subcordate at base, simply
or doubly serrate ; lat(!ral nerves and their divisions craspedodrome.

The leaves of tliis species are very variable, especially in the denticula-
tion of the borders. The most common variety is that represented in pi. xviii,

DESCRIPTION OF SPECIES— BETULAL'E^. 141

figs. 6-8, with borders doubly serrate, the teeth Ijeiiig small and sometimes
obsolete. This is the form recognized in the Miocene of Alaska, Greenland,
and the more common in Eur()i)e. The base of the leaves is rounded-
subcordate, the lower secondary nerves more or less branching The other
variety, a fragment of which is represented in pi. Lxiv, fig. 11, has the borders
either simply serrate or with a few irregular, large, more acute teeth ; all the
teeth, however, being larger and more obtuse than in the former variety.
The characters of the nervation are the same as seen in the figure ; the
leaves are obtusely pointed, not acuminate. By its larger obtuse teeth, the
leaf is more intimately related to Alnus nostratum, Ung., as described by
Ludw. (Palseont., p. 98, pi. xxxi, fig. 8). But this last species has the leaves
rounded at the top, and those of A. Kefersteinii, represented in Fl. Bait.,
loc. cit., especially fig. 9, agree entirely, in their form and the denticulation
of the borders, with the fragment under consideration.

Habitat. — Evanston, Wyoming; not rare The fragment represented
on pi. lxiv, near Florissant, Souti) Park, Colorado (Z)r. F. V. Hayden). Nine
miles southeast of Green River, "Wyoming {Win. Clehurn).
A.lnite8 iiisequilatcralis, Lesqz.
Plate LXII, Figs. 1-4.
Alnites inocquilateralis, Lesqx., Annaal Report, 1874, p. 307.

Leavej rather thin, apparently membranaceous, very variable in size and form, broadly oval,
obtnse or obtusely acuminate, rounded to the short petiole, distantly crenato-serrate ; lateral nerves
curving to the borders, either entering the teeth by their ends, or passing under them to follow the
borders in simple festoons, joining the teeth by small branchlets.

The leaves vary in size from four to eight centimeters long and from
three to six broad, one of the sides measuring generally one-fourth in width
more than the other. The irregularity in the number of the veins is corre-
spondingly great; one of the leaves, the smallest (fig. 4), for example, having-
six lateral veins on the left side, the lower much branched, while the other
side has ten, all simple. There are a number of fragmentary specimens of
the largest-sized leaves, like figs. 1 and 2, and these appear all related by
their outlines and the nervation to Populus Lebrunii, Wat., a species which
Saporta considers identical with his Alnus cardiojihylla, S(iz. Flor., p. 55, pi. iv,
fig. 9, and pi. xv, fig. 8. This last figure especially is much like fig. 1 of our
plate, merely diflfering by the form of the teeth, which, in the American
species, are broader and more obtuse. In this also the nervation is more
distinctly pennate, and the disposition of the veins to enter the teeth by their

142 UNITED STATES GEOLOGICAL SUltVEV— TEKTIAKY FLOKA.

extremity is more marked. The inequilateral shape of the leaves aud the
irregularity of nervation are not of frequent occurrence in the living species of
Alnus; these characters are seen, however, in the leaves of a number of fossil
species, like Alnus cychidum, Ung., A. sporadum, Saj)., A. cardiophylla, etc.
Habitat. — Alkali Station, Wyoming {IVm. Clehurn).

CUPULIFER^.

OSTRYA, Michx.

By their form and nervation, the leaves of this genus resemble those of
Carpinus and Betula. The teeth of the borders are smaller than those of
Carpinus, and they do not bear any secondary teeth upon their anterior face.
As yet, we have no fossil remains referable to this genus in the North Ameri-
can fossil floras. The capsule of the fruit, which is veined in its length and
vesicular, is easily recognized, though the leaves may not be distinct from
those of Carpinus. In the European Tertiary Flora, the genus Ostrya is
represented by six species, one of them Eocene, another of doubtful refer-
ence; the others all Miocene. The present flora has only two species, one
eastern, in southern Europe, extending from France to the Lebanon Mount-
ains; the other, O. Virginica, exclusively belongs to this continent, having
also a wide range of distribution, or from New Brunswick to Lake Winnipeg,
in 30° of longitude, and from 55° of latitude north to 20° in Mexico, where
the species has been found near Jalapa, and still more south, in the mountains
of Orizaba. It is, therefore, probable that one species, at least, of Ostrya
may be found in the American Tertiary.

CARPINUS, Linn.

The geological distribution of this genus does not agree in Europe
and in North America, as far as we know it, at least. While here two spe-
cies only are known by their leaves, from specimens obtained in the Upper
Miocene of the Parks, the paleontologists of Europe have described seventeen
species from leaves, and eight from the fruits or from involucres. It is pro-
bable, as Schimper supposes, that a number of these species will have to
be eliminated by more careful researches. However, the predominance of
this genus is marked in a high degree in Europe in com])arison to what it is
here. The genus appears already in the Eocene of Sezanne by the leaves
of two species; the others are Miocene; only one is referred to the Plio-

DESC'KII'TION OF SPECIES— CUPULIFER^. 143

cene. Of the five living species of Carpinus, one is now found in luiropc,
passing east into Asia, following about tlie same geographical distribution as
Ostrya carpinifolia. One also, C. Americana, is exclusively limited to the
North American continent, having apparently the same range as Ostiya
Virginica, of which it is a constant associate. It ranges toward the north as
far as Lake Superior, and its presence is recorded in Florida by Cliaj)man.
Gray, in his Statistics of the Flora of the Northern States, places it in the list
of the species which range through 15° to 19° of latitude.

Carpinus g:i'andis, Ung.

Plate XIX, Fis. 9; Plate LXIV, Figs. 8-10.

Carpinus grandis, Ung., Sillog., iii, p. G7, pi. xxi, figs. 1-13; Icouogr., pi. xx, fig. 4. — Heer, FI. Tert.
Helv., ii, p. 40, pi. ixxi, figs. 19 b, c, d, e, Ixxii, figs. 2-'M, Ixxiii, figs. 2-4 ; Fl. Foss. Arct.,
p. 103, pi. xlix, fig. 9; Fl. Fobs. Alask., p. 29, pi. ii, fig. 12, etc. — Lesqx., ADDual Keport,
1874, p. 313.

Leaves oblong-ovate or ovate-lanceolate, taper-pointed, doubly serrate ; secondary nerves close,
parallel, straight to the borders, simple or scarcely branching.

The leaves vary in size, from three to ten centimeters long and from two
to five centimeters broad: their nervation is sharply marked, for the second-
ary veins at least, which, nearly always simple, parallel, close, pass straight to
the borders in a more or less acute angle of divergence, according to the width
of the leaves. These are mostly oblong and taper-pointed or acuminate; their
base is generally round, or subtruncate. In fig. 9 of plate Ixiv, the base a})pears
cuneate; but the leaf seems to have been lacerated on both sides. This, how-
ever, if even the base of the leaf was wedge-shaped, could not separate it from
the species, as some of the leaves figured by the authors are more or less
acutely cuneate to the base. The nervilles are extremely thin and scarcely
perceivable upon our specimens; fig. 9 of pi. xix has the lateral veins more
distant, and the substance of the leaf is apparently of a thicker consistence,
membranaceous or subcoriaceous. For this reason I considered this leaf at
first as a Quercus, under the name of Q. Elkoana (Annual Report, 1873,
p. 413). Heer, however {loc. cit.), has in fig. 24 a leaf with the veins nearly as
distant; and the doubly serrate borders, with simple secondary veins, more
evidently relate this fine fragment to this species, especially comparable to
fig. 2 b, pi. Ixxiii, of Heer, Fl. Tert. Helv.

Habitat. — Near Florissant, Colorado {Dr. F. V. Hayden). Elko Station,
Nevada {Prof. E. D. Cope).

144 UiilTED STATES GEOLOGICAL SUKVEY— TERTIARY ELOEA.

CORYLUS, Toum.

This genus at our time is represented in the northern hemisphere only.
Of the seven species by which it is now known, two inhabit eastern Asia;
three are found in Europe, none being confined to that continent, however, for
two extend to Asia and one to Algeria. Two belong exclusively to North Am-
erica: Corylus Americana, wiiose range of distribution is from Florida to
Saskatshawan, and C. rostrata, which does not go southward far beyond the
Alleghany Mountains, and ascends only to the great lakes. In the fossil flora,
three species only. are known by their leaves in the Miocene of Europe.
One of them is nearly exclusively Arctic, its remains being found in great
abundance in the Miocene of Greenland. It is, however, quite as common
in the same formation of Alaska, and appears more rarely in the Miocene of
the Rocky Mountains. The specimens figured as marked below are from
Carbon and the Washakie group; some fragments, in a bad state of preserva-
tion, and therefore somewhat uncertain, have been found at Evanston. Four
species of Corylus are described from the Union group by Dr. Newberry.
From the similarity of the leaves of C. Mac Quarrii with those of our present
species, it appears certain that these had their origin in geological times as
far up as the Miocene at least, and also that their present characters and their
distribution agree with those of their ancestors.

Corylus mac Quarrii, (Forbes) Heer.

Plate XVm, Figs. 9-11.

AlnitesT Mac Qitarrii, Forbes, Qn.irt. Journ. Geol. Soc, 1851, p. 103.

Cwylui Mac Quarrii, Heer, Fl. Arct., p. 104, pi. viii, figs. 9-12, ix, figs. 1-8, xvii, fig. bd, lix, fig. 7o, p.

138, pi. xxi, fig. 11 c, xxii, figs. 1-6, xxiii, fig. 1, p. 149, pi. xxvi, figs. 1 a, 2-4, xxi, fig. 5 ; Fl.
' Foss. Alask., p. 29, pi. iii, fig. 9, pi. iv; Foss. Fl. of N. Greenl., p. 469, pi. xliv, fig. 11a,

xlv, fig. 66; Spitzb. Fl.,p. 56, pi. xi.figs. 10-13, xiii, fig. 35 6.— Lesqx., Annual Report, 1871,

p. 292 ; Supplemeut, p. 9.
AInu» pseudoglutinosa, Goepp., Tert. Fl. d. Polar Gcg., 1861.
Corylus grosae-serrala, Heer, Fl. Tert. Helv., ii, p. 44, pi. Ixxiii, figs. 18, 19.

Leaves of medium size, very variable, oblong or oval, pointed or acuminate, ronnded, truncate or
omargiiiate at the base, triply serrate, penninerved ; lateral veins thick, branching, craspedodromo, like
their divisions.

The immense number of specimens, some of them in a p.3rfect state of
preservation, which have been examined by Prof Heer, have enabled him to
compare the various forms of these leaves, and to refer them to the same spe-
cies. Seen separately in two or three fragmentary specimens, Hke those fig-
ured here, it is difficult to find their points of identity, and therefore easy to
refer each leaf to a diiferent species. The surface of these Corylus leaves is

DESCRIPTION OF SPECIES— CUPULIFER^. 145

always rough; the secondary veins and their divisions are traced deep into the
derm, and the nervilles are also dccj) and distinct, thougli the details of areola-
tion are rarely so. It is always ditficult, in our specimens at least, to recognize
tlie triple denticulation of the borders, the teeth sometimes appearing siujple,
and being generally more or less destroyed or erased.

Habitat.— Carbon Station, Wyoming; Washakie group {Dr. F. V.
Hat/den). I have not seen as yet any specimens of the species from the
Upper Miocene of Green River and the Parks. Those from the Lower
Lignitic are indistinct and scarcely determinable.

FAGUS, Toumf.

One of the more clearly defined generic types of the Cretaceous Dakota
group flora is that of Fagus. Two species, described from leaves (they
may, however, represent one only), have been published from this formation,
Fagus polydada (Lesqx., Report of the U. S. Gcol. Surveys of the Territo-
ries, vol. vi, p. G7, pi, v, fig. G) and F. crelacea (Newby., Notes on the Later
Extinct Floras of N. A., p. 23). From this it seems that a genus whose origin
is recognized in the Cretaceous, and which, by the wide distribution of one
of its species at the present time through the eastern slope of the North
American continent, furnishes one of the principal constituents of its forests,
should have left its traces through the intermediate geological formation
by an abundance of its fossil remains. As yet, it is not the case. No posi-
tively determinable fossil leaves referable to Fagus have been observed in the
Lower Lignitic of the Rocky Mountains, and none either in the higher stage
of the Eocene at,Evanston or in the Miocene of Carbon. The speci&s
described here below as Fagus FeronicB is from the Upper Tertiary (Miocene)
of the Rocky Mountains, and even, as it may be seen from the remarks in
the description, it is not certain that the leaves referred to it truly belong
to this genus. We have to go higher still in the Tertiary formations, or in
the Pliocene of the Chalk Bluffs of California, to find vegetable remains as
distinctly identifiable with Beech as are those of the Cretaceous. This may
seem a strange inconsistency of distribution. It may be accounted for, as in
other analogous cases, by our insufficient acquaintance with the geological
floras of this continent, represented as they still are by comparatively scanty
materials. The proof is, that fruits of Fagus, perfectly similar to those of
the living species, have been found in the Lower Lignitic of Tennessee and
10 T F

14G I" SITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

Mississippi, and described as Fagvs ferruginea, in the Geological Report of
Tennessee, 1869, 11.427, pl.K, fig.ll. Prof. Heer also identifies two species of
Fagus in the Miocene of Alaska, whose synchronism with the Lignitic dcp<js-
its of Carbon is evinced by a number of identical specific types. One of
those Alaska species, Fagus Antipoji, is in the Chalk Bluifs formation of Cal-
ifornia, and another, jP. castantcefolla, Heer, is recognized in the Miocene of
Oregon. Therefore the persistence of the generic type in the subsequent
stages of the North American Tertiary seems sufficiently established.

Besides the two uncertain species of the Eocene of France, European
paleontologists have recognized tea fossil species of Beech, mostly Miocene;
two of them ranging from Greenland to Italy and Syria. All except one are
intimately allied to the living species of North Europe and North America,
Fagus ferruginea. Ait., and Fagus sylvatica, Linn., to which may be added
the Japanese F. Siebo/di, Endl., or to the northern types. The other living
species, twelve in number, all inhabit the austral hemisphere, and to one of
these, F. obligua, Mirb., of Chili, one Miocene species only, F. pi/gnma, Ung.,
from Euba;a, is comparable.

Fat; us Fcronise, Ung.
Plate XIX, Figs. 1-3.

Fagus Ftronia, Ung., Cblor. Protog., p. 106, pi. xxviii, fig8. 3,4.— Ett., Bcitr. z. Fobs. F1. v. Tokay (Sitz.-
Ber. d. K. K. Ak. d. Wiss.), xi, p. 799; Foss. Fl. v. Bil., i, p. 50, pi. xxv, figs. 12-20.— Lesqx.,
Annual Report, 1873, p. 413.

Fagus Deucalioiiis, Ung., Icouogr., p. 38, pi. xviii, fig. 24.

Vhnus qitercifulia, Ung., Chlor. Protog., p. 96, pi. xxv, fig. 5.

Qttercus myricafolia, Ung., Iconogr., p. 37, pi. xviii.

Leaves ovate or oblong, pointed or acuminate, narrowed or broadly wedge-form to a compara-
tively long petiole; borders irregularly dentate; secondary veins simple, parallel, craspedodrome,
scarcely carved in passing to the borders.

The three leaves referred here to this species agree so exactly in all
their characters with those published by the European authors, as seen above,
and also with the description in Schimper (Pal. V^get., ii, p. 603), that it
is scarcely possible to doubt their identity. Their size is variable; their
ovate-acuminate form, their irregular dentation, the comparatively long peti-
ole, the distribution, and also the degree of divergence of the secondary veins,
are the same. The American forms appear only slightly more acutely nar-
rowed to the petiole, but a difference of the same kind is remarked between
the leaves figured by the authors, as between figs. 3 and 4 of jjI. xxviii of
Ung. Chlor. Protog., and therctbre cannot be considered as specific. It is,
however, questionable if all these leaves may be referred to Fagus. This

DESCRIPTION OF SPECIES— CUPDLIFEK^. 147

uncertainty is evident from Unger's synonymy, and is still more forcibly
suggested by the fact that no species ol Fagvs has leaves with doubly den-
tate borders. This is a question, however, which has to be decided I)y
European authors.

Habitat. — Elko, Nevada {Prof. E. D. Cope). The two specimens from
Golden described under this specific name in Annual Report, 1872, p. i578,
are mere fragments, too imperfect for satisfactory determination. The same
remark is applicable to the single leaf referred to Fagus Deucalionis, Ung., from
Evanston specimens, in Annual Report, 1871. p. 292.

aUERCUS, Linn.

The distribution of the Oaks at (he present time affords .some interesting
data, which may be considered with reference to what is known of the devel-
opment of the genus in the geological times.

Of the two hundred and eighty species described in the Prodromus, the
North American continent has for its share one hundred and twenty-one,
tliirty of which belong to the United States, including New Mexico, and
eighty-five to Mexico, most of them, at least ; for a few species descend far-
ther south to Costa Rica, even to New Granada, one of these being found in
the mountains near Bogota, 5° north of the equator. Of the twenty-five
species which are credited to Europe, nine are Mediterranean, passing to
North Africa and to Western Asia. None but these Mediterranean species
have been found upon the continent of Africa. From Asia, one hundred and
twenty species are described, a large number from the southern peninsula
and from the depending islands, Java and Sumatra; even a few from the
Philippines and Celebes Lslands. In that continent, the genus extends its
representatives to 5° south of the equator. Japan is credited with nineteen
species. From this we see that the Oaks belong to the northern hemisphere
nearly exclusively, as very few species pass below the equator in Asia, and
as South America, Australia, and Africa (except those of the southern shores
of the Mediterranean) have none. This fact is worth remarking, as the range
of habitat of the Oaks is as variable as the characters of their leaves; and
therefore they should have, more than any other genera, by their facility of
adaptation to every kind of climatic circumstance, a world-wide, or at least
a more general, distribution. Considering the species of the eastern slope of
the United Stales, Gray's Statistics of the Flora gives to Quercus j>aluslris a

148 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

range of 6° only, while Q. rubra, Q. alba, and Q. obtusiloba have their place
in the list of the species ranging from nineteen to twenty-nine degrees of
latitude.

The disposition of Oaks to constitute varieties by hybridity is well
known. This has rendered the specification of many of their forms very
difficult and uncertain. It is therefore not surprising to find the same uncer-
tainty in regard to a number of the fossil forms referred to this genus, and
which are identifiable by leaves only, the most variable and diversified organ-
ism of all.

Considering what is known of Oaks in geological times, species of
Quercus are positively and distinctly represented in the Cretaceous; at least,
if we admit as representatives of this genus the leaves of the Cretaceous of
Belgium, which have been separated by Debey and d'Ettingshausen, under
the generic name of Dryofhyllum, for reasons which have not been as yet
satisfactorily explained. These- leaves have the same characters as certain
species of Oaks; and, compared for their nervation, their form, even their
consistence to those of some of the North American living species, they do not
present any mentionable diflference. The fine leaf, for example, described in
the Supplement to the Cretaceous Flora, Annual Report, 1874, as BryophyUum
{Quercus) latifolium, p. 340, pi. vi, fig. 1, is like a counterpart of some leaves of
Quercus bicolor; and that of Quercus primordialis of the Cretaceous Flora (Rep.
U. S. Geol. Surv. of theTerr., vol. vi, p. 64, pi. v, fig. 7) is equally similar to leaves
of Q. prinos. The section of the Willow-oaks, the Salicifolioe, established by
Schimper for species with entire leaves, is less definitively recognized in the
Cretaceous. Tlie characters of Quercus salicifolia and Q. cuneata, Newby.,
like those of Q. Ellsworthiana, Lesqx., do not exclusively pertain to Oaks;
therefore the relation of the leaves representing these forms is somewhat
doubtful. However, the evidence afforded to this question by geological
records seems to prove the origin of at least two different types of Oaks in
the Cretaceous ; for the Paleocene of France has five species, described by
Watelet, and, besides those which are mentioned here below, five others
have been recently observed by Saporta in the Gelinden formation, which,
considered for a long time as Upper Cretaceous, is now "definitively recog-
nized as the lowest member of the Eocene of that country. Three of these
Oaks, according to the author's remarks, go to the section Lepklobalanus,
the two others represent that of Cerris. Higher into the Tertiary, the

DESCKirTION OF SPECIES— CUPULIFERiE. 149

European authors describe a few species fioin the Middle Eocene, and still
a larger number from the upper measures, the Oligocene or Armissan. It is,
liowever, in the Miocene that the genus reaches its full development; and, judg-
ing from data obtained from the study of tin; North American Tertiary floras,
the predominance of Oaks increases even into the Upper Miocene and the
Pliocene, for, in fifty species which represent the flora of the Chalk Bhiffs of
California, nine, or more than seventeen per ceut., belong to Quercus. This
proportion, as far as it can be estimated, is far above tliat of the Oaks in the
present North American flora; for, counting the arborescent, or woody hard
species, which could be preserved by fossilization, in supposing analogous
circumstances, the Oaks of the eastern slope of the United States would not
constitute more than six per cent, of its flora.

Prof W. P. Schimper, in his Pak^ontologie Vdgdtale, describes one
hundred and sixty fossil species of Oaks, which he distributes in tive different
sections : —

In the first, that of the Salicifolia, he has forty-seven species, four of
which are Cretaceous, twelve Eocene, and tliirty-one Miocene. Among the
Eocene species, he places three described from Vancouver's Island.

In the second section, which includes the si^ecies with serrate or dentate
leaves, he has thirty-nine species, three of which are Cretaceous and five
Eocene ; of these, three are North American. This disproportion in the distri-
bution is rather uncertain, however, and explainable in part by the fact that the
Eocene floras of Europe, that of Mount Bolca, for example, are still unknown,
and also J)y the reason that some species referable to this section are described
by Schimper under the generic name of Dryophyllum. Of these, the S(izanne
flora (Lower Eocene) has four; that of Gelinden, a still lower formation, has
five (these published since the Pal^ontologie Vc^gt'tale was out); and two
more have been more recently discovered in the Point of Rocks measures
of Wyoming, which, l)y its flora, characterizes the lowest Tertiary of this
country.

Schimper's third section has the species with coriaceous, few-nerved,
or indistinctly nerved leaves, which are generally of difficult and uncertain
determination. It contains thirty species, three Oligocene, one Quaternary,
the other Miocene. Quercus aculiloha, a well-characterized Oligocene species
of this section, has been found also at Golden in the Lower Lignitic of
Colorado.

150 U>'ITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

The fourth section, for lobalo-crenate leaves and craspedoclrome secondary
veins, has only fonr species, which, I think, should l)e placed in the second
division ; for Quercvs castanea and Q. furcinervis are evidently derivations
of Dryophyllum, or referable to the same section ; while Q. ehpnodrys and
Q. Deuterogeva, two species of Unger, are comparal)le to Q. prinoides and
Q. montana, a type of the North American flora, which, as seen above,
appears also derived from Dryophyllum.

To the fifth section pertain the species with sinuate-lobed borders, like
those of our White Oaks. It has only twelve species, all Miocene, especially
Upper Miocene. As yet, this type is recognized in the Tertiary flora of this
continent by one species only from the Pliocene of California. However,
Q. Furuhjelmi, Heer, from Alaska, is placed by Schimper in this section, and
as this species is closely allied on one side to the Cretaceous Dryophyllum
(Quercus) latifolimn, on the other to the present Q. hicolor, we may, without a
too hazardous hypothesis, consider this group as identical with the former, the
leaves passing, by the deepening of the obtuse teeth into lobes, to Q.. alba,
Q. ?nacrocarpa, etc., and, by narrowing them into more acute teeth, to Q.prinos,
or to the fossil species Q. pseudo-cast anea, Q. furcinervis, etc.

The last section of Schimper is reserved for the doubtful species, twenty-
four in number, two of them known only by fruits.

From this it may be assumed that the different groups of Oaks of the
North American present flora, at least those of the eastern slope, have their origin
recorded in the Cretaceous by some related specific types; that the original
characters, those pertaining to the fourth section, are recognized in species
of the Eocene of the Rocky Mountains; and that perhaps even the genus
takes its preponderance in North America, as in Europe, during the Miocene
period, especially in this country in the Pliocene.

§ I. — Leaves with entire borders.
Quercus neriifolia, Al. Br.

Plate XIX, Figs. 4,5.

QuercM neriifolia, Heer, Fl. Tert. Helv., ii, p. 45, pi. Ixxiv, figs. 1-6, 16 a, b, c, d (acorns), pi. Ixxv, fig.
2; iii, p. 178, pi. clii, fig. a.— Ung., Gen. et Sp., p. 403.— Ett., Foss. Fl. v. Bil., p. 54.—
Gaud., Cont., vi, p. 12, pi. ii, fig. 1. — Lesqx., Annu.il Report, 1873, p. 413.

Qucreus Jignitum, Al. Br., Stizenb. Verz., p. 77. — Heer, Ueber iler Tert. Fl., p. 53.

Quercus commutata, Heer, Fl. Tort. Helv.. pp. 14,21.

Leaves coriaceous, with polished surface, narrowly lanceolate, gradually tapiriug toward the
base; middle nerve thick and deep; secondary veins on a very open angle of divergence, sparingly
branching toward the borders, where they are effaced ; generally separated by shorter intermediate
tertiary veins.

DESCRIPTION OF SPECIES— CUPDLIFER^. 151

The two specimens referred 1o this species are very imperfect. One
(fig. 5), with coriaceous texture, smoolh surface, gradually narrowed down-
ward, has, however, the characters of this species, especially as marked by Heer
{Joe. cit., figs. 3, 4). The secondary veins are close, open, on an angle of
divergence of 60°, with shorter intermediate tertiary veins. The fragment
represented in fig. 4 is less distinctly identified with this species by the coria-
ceous substance of the leaf, its open veins nearly at right angle to the thick
midrib, toward the base, and more evidently branching near the border.

Habitat. — Near Florissant, west of Pike's Peak, Colorado {Dr. A. C.
Peak), specimen represented in fig. 4. Raton Mountains, New Mexico {Dr.
F. V. Haijden), specimen represented in fig. 5.

Querciis straiuinea, Lesqz.
riate XIX, Figs. 6,7.
Qutrcim siraminea, Lesqx., Annual Report, 1872, ]>. ,373.

Leaves variable in size, broadly ovate, obtusely pointed, ronnded downward, or narrowed to a
short petiole; secondary veins parallel, on an open angle of divergence, camptodrome.

These leaves seem to represent two different species. They have, how-
ever, a common character, that of the yellowish, shining, secondary veins,
which I have not seen on any other species of fossil plants of this locality.
The leaf in fig. 6 is smaller, with the secondary veins at a more acute angle
of divergence; it has, however, the same character of nervation, shorter
intermediate tertiary veins, and more or less distinct veinlets, oblique to the
secondary veins. In both leaves, also, tlie basilar veins are marginal, and
ascend to the branches of the secondary nerves above. This species is dis-
tantly related by its nervation to Quercus Des/ocsi, Heer (Fl. Tert. Helv.), as
represented in pi. Ixxviii, fig. 7, a species which shows in different fragments
(figs. 6 and 14) a marked diversity in the size of the leaves and in the
angle of divergence of the secondary veins.

Habitat. — Golden, Colorado.

Quercus ciilorophylla, Ung.

Plate XXI, Fig. 3.

Quercus thlorophylla, Ung., Cblor. Protog., p. Ill, pi. xxxi, lig. 1.— Heer, Fl. Tert. Helv., ii, p. 47, pi.
Ixxv, figs. 3-9.— Massal., Fl. Foss. Senog., pi. xxx, fig. 4.— Sism., Mater., p. 41, pi. ix, fig.
5.— Lesqx., Trans. Amer. Phil. Soc, xiii, p. 41t), pi. xvii, figs. 5-7; Annual Report, 1869, p.
196; Supplement to Annual Report, 1871, p. 14; Annual Report, 1872, pp. 383, 407.

Leaves coriaceous, ovate, obtuse, rounded in narrowing to a short petiole; borders entire, undu-
late, apparently recurved; secondary veins thin, obsolete, camptodrome.

Though a number of fragments referable to this species have been found,
none represent a full leaf, and the more complete specimen is here figured.

152 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

Generally the leaves are more distinctly narrowed to the short petiole than
in this one; hut even in the European specimens, as for example in the fine
branch figured by Sismonda {loc. cit.), this character is remarked quite as
frequently as that of the more acutely cuneate base. The differences are
especially appreciable in the three leaves from the Mississippi Lower Lig-
nitic, in Trans. Amer. Phil. Soc, loc. cit., where fig. 5 has its base still more
obtusely rounded than in Ihe leaf figured here; while fig. 7 has its lower part
more narrowed, as it is in tlie leavQg described by Heer, and fig. 6 is inter-
mediate between both. In all the fragments which have come under my
examination, the substance of the leaves appears coriaceous; the secondary
veins are scarcely perceivable and the borders seem narrowly reflexed. The
leaves do not vary greatly in size, averaging eight centimeters long and three
to four centimeters broad, being always, as seen in the figures given by the
authors cited, either obtuse or rounded, or even slightly emarginate at the
point.

Habitat. — Marshall's, Erie, Colorado. Fischer Peak of the Raton
Mountains, New Mexico (Dr. F. V. Hayden). Six miles above Spring
Canon, near Fort Ellis, Montana (Dr. A. C. Peak). The specimen from
this last locality is obscure.

Qnerciis ciiiereoides, ep.. nov.

Plate XXI, Fig. 6.

Leaf small, subcoriaceous, entire, narrowly ovate-lauceolate, acuminate, rounded to the petiole ;
middle nerve narrow; secondary veins curving toward the borders, camptodrome.

This fine leaf, nearly five centimeters long, two centimeters broad in the
middle, its widest part, is narrowed by a curve- to the base, and seems to be
short-petioled (petiole broken). The secondary veins at unequal distance,
but parallel, join the narrow middle nerve under an angle of divergence of
40°, first nearly straight, then curving near and along the borders, or campto-
drome; the areas are all primary ones and undivided except by nervilles,
which, at right angle to the middle and secondary nerves, branch nearly in
the same direction, and form irregularly square or rectangular areolae. The
relation of this species is marked with the North American Quercus cinerea,
Michx., of the pine-barrens of the South, or to Quercus jthellos, \ar. maritima,
for its nervation at least. Among the fossil species, it is especially comparable
to Quercus salici7ia, Sap. (lilt., i, p. 24, pi. vi, fig. 6), of the Gypses of Aix, France.

Habitat. — LocaUty unknown.

DESCllIPTION OF SPECIES— CUPULIFEK^. 153

^ II. — Leaves serrate, dentate, or crenate.
Quel- ens Valdeiisis, Heer.

Plate XIX, Fig. 8.

Querent Valdensis, Heer, VI. Tcrt. Helv., ii, p. 49, pi. Ixxviii, fig. 15; iii, p. 178, pi. cli, fig. 17.— Ett.,
Fo88. Fl. V. Bil., p. 56, pi. xvi, figs. 5, C, 7.
Leaves eoriaceous, ovate, acuminate, rounded to the petiole ; secondary veins close, parallel, more
open toward the base, subcamptodronie ; borders obscurely dentate.

This fragment, representing llie lower part of a coriaceous leaf, whose
base is rounded to a slender, short petiole, has rather the appearance of a
leaf of Betula, or of Ostrya, than of a Querats. It resembles closely, how-
ever, the fragment described by Heer {loc. cit., fig. 15) by its coriaceous
consistence, its form, and nervation. The .sharp denticulation of the teeth is
not distinct in this specimen. I have seen another, from Contra Costa,
California, which is more perfect, representing a whole leaf, acuminate and
sharply dentate, with the same type of nervation as this one, and evidently
referable to the same species. The secondary veins are subcamptodrome,
either following along the borders in festoons and entering the teeth by
nervilles, or some of them passing up to the point of the teeth, as marked
by d'Ettingshausen {loc. cit., fig. 7). This species is evidently extremely
variable; for the leaves seen by this last author are obtuse, some of them
round-oval; in all, however, the secondary veins are more open toward the
base, and the nervilles distinct, nearly simple, and continuous in right angle to
the veins.

Habitat. — Lignites of Rock Creek, Laramie Plains, Wyoming, -with
Quercus acroclon, Populus subrotundata, etc. {Dr. F. V. Hnyden).

Qiicrciis Godctil, Heer.

Plate XX, Fig. 1.

Quereut Godeii, Heer, Fl. Tert. Helv., ii, p. 50, pi. Ixxviii, figs. 10, Jl ; iii, p. 179, pi. cli, fig. 11.— Lesqx.,
Annual Report, 1871, p. 297.
Leaf snbcoriaceous, lanceolate, tapering upward, rounded to the base; borders serrato-dentate;
secondary nerves numerous, parallel, branching near the borders, camptodromo, separated by tertiary
shorter veins.

Of the two specimens which we have of this species, the best preserved
one is figured here. The leaves appear coriaceous, or, at least, subcoriaceous,
their surface being blackened by an adhering thin coating of coaly matter
and undulately wrinkled across, as in coriaceous leaves ; the stone is, however,
too coarse to show the thickness of the vegetable remains imbedded into it.
These leaves are unequal at the base, more enlarged and rounded on one

154 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

siilo, and tapering to a jniint (?) (broken). The only difference remarked
between this leaf and the one fignred by Hecr {loc. cit., pi. li, fig. 11), is in
the size of the teeth, which are shorter in the American form ; they are,
however, sharply i)ointed and unequal. The secondary veins are exactly of
the same character as in Heer's species, branching toward the borders, camp-
todrome, with intermediate, short, tertiary veins, sometimes on a more open
angle of divergence. These leaves, even if their identification with the Mio-
cene species of Europe were positively ascertained, do not seem to be refer-
able to Oaks. By their small, sharp teeth, their unequal base, their nervation
also, they closely resemble the leaflets of some compound leaves, like those
of some Aralia, our A. sjnnosa for example, wliich, however, are not thick.
Habitat. — Six miles above Spring Canon, Montana, with Cinna?fiomum
lanceolatum, Rhamnus rectinei'vis, etc. {Dr. F. V. Hat/den).

Q ti e r c II s C I e b u r ii i , Lesqz.

Plate XX, Fig. 2.

Quercua Clebnrni, Lesqx., Annual Report, \fi73, p. 399.

Leaf coriaceous, linear-oblong, narrowed to the base, obtusely dentate; secondary veins sbort and
obsolete, nearly at right angle to the midrib.

The upper part of the only leaf which represents this species being
destroyed and the nervation obsolete, its characters are indefinite. Its pecu-
liar form and nervation, however, and the flat, slightly obtuse teeth of the bor-
ders, even the unequilateral base, are characters identical with those of Quercus
urophylla, Ung. (Fl. v. Sotzka, p. 33, pi. ix, fig. 9). As in our species, also,
the lateral veins arc in right angle to the midrib. From the numerous
figures given of that Quercns by the German author, the leaves are seen to
be extremely variable, especially in the characters of the denticulation of the
borders and the direction of tlie secondary veins.

Habitat. — Black Buttes, Wyoming.

Quercus? f raxinif olia, Lesqz.
Plate XX, Fig. 3.
Qito'cus Laharpi, Gaud. — Lcs(jx., Annual Report, 1871, p. 297.

Leaves membranaceous, lanceolate, rounded and narrowed to the entire base, tapering to the
obtusely dentate point; secondary veins close, numerous, cuiving in passing up to the borders, cauipto-
dronie.

This leaf, known only from one specimen, is about eight centimeters
long, two centimeters broad in its lower part, entire from the base to near the

DESCRIPTION OF SPECIES— GUPDLIFEK.^. 155

middle, obtusely dentate above, witli mimeroiis parallel, camptodronie, sec-
ondary veins, diverging about 40°, slightly more open in tlie lower part of" the
leaf. The reference to Quercus Lahnrpi is contradicted by the total absence
of nervilles, the more numerous, closer, secondary veins, and the curve of
the borders of the leaf to the base, or to the petiole. It resembles some
species of Fraxinus, like F. denticulata, Heer (I\rioc. Bait. FL, p. 89, pi. xxiv,
figs. 2fi and 27). The absence of the details of nervation |)revents a satis-
factory comparison and identification of this leaf".

Habitat. — Six miles above Spring Canon, Montana {Dr. F. V. Hayden).

Quercus Ellisiana, Lesqx.

Plate XX, Figs. 4, 5, 7, 8.

Quercus EUisiana, Lesqx., Annual Report, 1871, p. 297.

Leaves broadly ovate or ovate-lanceolate, obtusely pointed, obtusely dentate in the upper part,
broadly cuneate to a short petiole ; secondary veins at au open angle of divergence, slightly curving in
passing to the borders and entering the teeth.

All the leaves of this locality are ujion coarse metamorphic sandstone,
which the vegetable substance seems to have penetrated, and upon which the
outlines and the essential parts of the nervation are preserved as painted in
black. For this reason, the original consistence of the leaves is inappreciable.
These leaves are numerous, variable in shape, broadly rhomboidal, and entire,
or with undulate borders, as in fig. 7 ; or ovate and broadly wedge-form
toward the point and the base, and distinctly obtusely dentate in the upper
j)art, as in fig. 4; or oval, undulate, or dentate above, as in figs 5 and 8.
The nervation is variable in the same degree, the secondary veins being
either forking, distant, open (angle of divergence 40° to 50°), as in figs. 4
and 7, or more numerous, closer, simple, and more oblique, as in fig. 5.
A number of fragmentary specimens unite all these forms in one species by
transition between their characters. Their general outline is that of leaves
of Alnus, a genus to which they might be referable ; but the narrow lanceo-
late leaves, like that of fig. 5, rather resemble Qmicns leaves, at least some
of those described as Quercus by European authors. Thus this species is
closely allied to Quercus 'pseudo-alnus, Ett. (Foss. Fl. v. Bil., p. ,59, pi. xvii,
figs. 3-6), merely differing by the character of the denticulation, which is gen-
erally more obtuse in the American form, and does not descend so far down.

Habitat. — Six miles above Spring Canon, near Fort Ellis, Montana
{Dr. F. V. Hayden).

156 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

Quo re (IS Pralei, Lesqx.

Plate XX, Fig. 6.

Quercus Pealei, Lesqs., Anuual Report, 1871, p. 297; 1872, p. 406.

Leaf coriaceous, small, cuneate, and entire from the middle to the base (petiole broken), more
abruptly narrowed to an obtuse point, and undulato-dentate in the upper part ; lower pair of secondary
veins from the base of the leaf, opposite, at a more acute angle of divergence, and camptodrome ; upper
pair alternate, distant, sparingly branching, craspedodrome.

This small leaf, four centimeters long and two centimeters broad in the
middle, may be referable to the former species. The character of the
nervation appears at first very different; but, comparing fig. 4 to fig. 6 of
the same plate, the analogy in the position and direction of the secondary
veins becomes more evident, for in fig. 6, as indicated by the festoons of
the tertiary veins or branches along the borders above the broken part,
the leaf has, like that of fig. 4, a marginal vein, whose place is taken on
the other side by a camptodrome secondary vein. The more marked differ-
ence which induced me to separate this species is the more evidently
coriaceous substance of this leaf and the discernible nervilles. It appears
intimately related to Quercus fagifoUa and Q. triangularis, Goepp, (Schoss.
Tert. Fl.,pp. 14 and 15, pi. vi, figs. 9-17), — two species described by Schimper
as Farrotia fagifolia.

Habitat. — Six miles above Spring Canon, Montana {Dr. A. C. Peale).

Near Fort Ellis, Montana (Jos. Savage).

Quercus Haidin$feri, Ett

Plate XX, Figs. 9, 10.

Quercus Eaidingeri, Ett., Fobs. F1. v. Vien., p. 12, pi. ii, fig. 1.— Heer, FI. Tert. Helv., ii, p. 53, pi. Ixsvi, figs.
5, 7, 8, 10, 14. — Gaud., Contr., ii, p. 42, pi. iii, fig. 6. — Lesqx., Annual Report, 1874, p, 313,

Leaves subcoriaceous, broadly lanceolate, acuminate, narrowed to the base; borders creuulato-
serrate; lateral nerves uuequidistant, sparingly branching, effaced toward the borders, camptodrome.

The American leaves representing this species are slightly larger than
those of the European Tertiary, especially that of our fig. 10. They agree,
however, so well by all their characters with those described under this name,
that I consider the identity as undeniable. The size is from ten to fourteen
centimeters long, and three and a half to five centimeters broad below the
middle. The largest leaf in Heer (loc. cit., fig. 4) is as long as that of our fig.
9, and slightly narrower. Tlie narrowly ovate-oblong shape of the leaves,
tapering upward to an acumen and slightly more rapidly narrowed in curving
to the base, the crenulate or dentate borders, are the same, and the inequi-
distant secondary nerves, some of them simple, some sparingly branching,

DESCRIPTION OF SPECIES— CUPULIFEB^. 157

with tertiary intermediate veins, arc also of the same character as seen in
Heer (Joe. cit., figs. 7 and 14). This species is related by some of its characters
to the two former ones. As Heer and d'Ettingshausen, also, have found and
described the fruits of this species, its relation to this genus seems sufficiently
established.

Habitat. — Green River, Wyoming, with Ficus arenacea, F. Gmidini,
Populus arctica, etc., described hereafter 'Dr. F. V. Haydcn).

Qiicrciis drymcja, Ung.
Plate XIX, Fig. 14.

Quo-cms drijmfja, Ung., Chlor. Protojg., p. 113, pi. xsxii, figs. 1-4. — Heer, Fl. Tert. Helv., ii, p. CO, pi. Ixxv,
figs. IS, 19, 20.— Ung., Fosa. F). v. Sotzka, p. 33, pi. ix, figs. 1, 2.— Gaud., Contr., i, p. 17,
pi. vi, fig. 4; vii, lig. 4; ii, p. 44, pi. iv, figs. 1-10. — Ett., Foss. Fl. v. Bil., i, p. 58, pi.
xvi, lig. 9. — Maesal., Stud., i>. 18l>, pi. xsiv, fig. 7. — Sism., Mater., p. 40, pi. xvii, fig. 1, etc.

Leaf linear, apparently acuminate; borders equally distantly dentate; lateral veins in an acute
angle of divergence, nearly straight to the borders, craspudodrome.

There is perhaps no sufficient evidence of the reference to this species
of the fragment figured here. It has the t()rm and size of the leaf of Un-
ger's Fl. v. Sotzka (Joe. eit.), fig. 2; the nervation is of the same character, the
secondary veins passing up in an acute angle of divergence and nearly straight
to the point of the teeth, and also the subfalcate shape of the linear-lanceolate
leaf is similar. In fig. 9 of the Bil. Fl. {loe. eit.), the linear leaf, the angle of
divergence, 30°, of the veins, and the distant teeth, are of the same character
also; and, indeed, I do not know any other species, eilher fossil or living, to
which this fragment might be more legitimately referred. The ditference in
the form of the teeth, either very acute or somewhat obtuse, is remarked also
upon the fragment, as well as npon the leaves figured by European authors.

Habitat. — Near Castello's Ranch, Colorado (^Dr. F. V. Hayden).

Q u c I' c II s II a y d e II I ■ , Lesqz.

Plato XIX, Fig. 10.

Qutrciia Haydenii, Lcsqx., Annual Report, 18G9, p. 190.

Leaf lyrate, sharply dentate toward the enlarged base and at the abruptly narrowed point;
nervation palmato-pinnato ; lower lateral veins branching, open, the upper one distant, parallel, all
craspedodronie.

This leaf has no relation to any fossil ones known to me. Its general
outline is somewhat like that of the leaves of Liriockndron. Enlarged at
the nearly truncate base, and there round-lobed, with dentate borders, it is
narrowed or strangled in the middle, and still enlarged upward, with its top
broadly deltoid and sharply dentate. The middle part of the leaf only has entire

158 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

borders. The nervation is as irregular as the form, being tripalmate at the
base, with the lateral veins, one only and branching on one side, while on the
other they are double and parallel, with three pairs of secondary veins above,
at a distance from the primary nerves; a nervation related to that of Plata-
nus. Indeed, by its sharp teeth, turned upward, this peculiar leaf might be
taken for a deformity of the common F. aceroides or P. Guillelma, but for
its remarkable shape. As it comes from the same locality as the specimens
of the following and apparently very variable species, it may belong to it. This
possible relation indicates its place in this genus. It seems to represent a
transitional form between the dentate and the lobate leaves of Quercus.

Habitat. — Laramie Plains, near Rock Creek, Wyoming {Dr. F. V.
Hay den).

<t II e r c 13 s a c r o d o n , Lesqx.
Plate XIX, Figs. 11-13.
Qttercus acrodon, Le.sqx., Annual Report, 1869, p. 196 ; 1872, p. 389.

Leaves oval or obovate, more or less rapidly narrowed upward to an acute point and downward to
the petiole; borders sharply deeply dentate; lateral nerves parallel, equidistant, craspedodrome.

The three fragments figured here expose the general characters of the
leaves of this fine species. The form is ovate, either narrowed in about the
same degree to an acute point and to the petiole (broken), as in fig. 12, or
obovate, rounded upward to a short point, and narrowed cuneiform to tiie
base. The borders are sharply dentate, sometimes with unequal teeth, as in
fig. 13, but the teeth are all turning upward and deep, separated by obtuse
sinuses. The secondary nerves, mostly simple, some of them, however,
branching near the borders, pass up, all in the same angle of divergence, 40°,
either straight or slightly curving, entering the teeth as well as their branches.
The substance of these leaves is not coriaceous, the areolation being, how-
ever, rendered obsolete by a thin coating of carbonaceous matter. In fig. 11,
only the netting is distinctly seen, formed by veinlets in right angle to the
nerves, close, dividing, either obliquely or in right angle, into quadrangular
meshes. The base of the leaves only has entire borders.

Like the former leaf, these have no direct relation to those of any fossil
species of Quercus published by authors. Anu)ng living species, its affinity
is with Q. agrifolia, Nee., of California.

Habitat. — Same locality as the former {Dr. F. V. Hayden).

DESCRIPTION OF SPECIES— CUPULIFERvE. 159

Quercus Vibiirui folia, Lesqx.

Plato XX, Figs. 11, 12.

Quercus attenuataT, Goepp., Lp8()X., Annual Report, 1873, p. 398

Qucrcua triangularis, Goepp., Lescix. (in part). Annual Report, 1872, p. 377.

Leaves oval-oblong, narrowed and broadly cuneato downward to a slender petiole and npward to
a short point; borders distantly dentate; nervation pinnate, craspedodromo.

Except for the character of denticulation of the borders, the leaf in fig.
11 is quite as similar to Q. attenuata, Goepp. (Tert. Fl. v. Schoss., pi. viii, fig.
4), as that of fig. 12 is to Q. triangularis of the same work (pi. vi, figs.
13—17). The first of tliese leaves of ours (iiffers by the distant, short teeth
of the borders, turned upward and mostly simple, while Goppert's leaves are
doubly dentate, with very small teeth turned outside. The second leaf (fig.
12) has, as seen in the upper part of the left side, small, simple teeth, agree-
ing in characters with those of fig. 11, while Goppert's Q. triangularis has
undulate borders. In all these leaves, the general outlines and the nervation
are the same, and they evidently refer to the same genus. But, considering
its slender, apparently long petiole, the relation of these leaves to the species
of the German author, or even to Quercus, seems uncertain. Count Saporta
considers them as rather representing a Viburnum. Indeed, they have some
affinity to V. inarginatum of pi. xxxviii, figs. 1—5, of this volume, their relation,
especially with fig. 2, being more appreciable. As in the species from Black
Buttes, that of Golden shows a cuneate base, lower secondary veins opposite,
emerging a little above the border base of the leaves, branching outside, all
the nerves aiul primary branches entering the short teeth, and a broadly
angular top, as wide as in V. marginatum, at least as it is in the leaves of figs.
2 and 3 {Joe. cit.). There is a great ditference, however, in the thin, more
distant, secondary veins of this species, and in the general outlines of the
leaves. These vary in size from four to nine centimeters long and from three
and a half to six centimeters broad in the middle, where they are widest,
their shape being rhomboidal ; the petiole of the largest leaf is one and a
half centimeters long to the point where it is broken; the angle of divergence
of the veins, 40°, is the same in both leaves, as is aLso the number of the sec-
ondary veins in comparison to the size, the smaller leaf having four pairs
only, and the large five, all slightly curving in passing up to the borders.

Habitat. — Sand Creek, Colorado {A. R. Marvine), Golden, Colorado,
and Black Buttes, Wyoming. The specimens from Black Buttes are upon
burned red, very hard clay-shale, where the vegetaljle organs have been
somewhat obliterated by heat. They are, however, identifiable.

160 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLOKa.

Que re IIS pi at a III a, Heer.
Plato XXI, Fig. 1.

Quercua platania, Heer, Fl. Foss. Arct., i, p. 109, pi. xi, fig. 6, xlvi, fig. 7; ii, pi. xlvi, fig. 5, Iv, fig. 3 c;
Spitz. Fl., p. 57, pi. xii, figs. 5-7.— Lesqx., Annual Keport, 1872, p. 386.

Leaves membranaceous, very large, round-cordate to tbo base ; borders simply dentate ; secondary
veins thick, the basilar ones opposite, more distant from the upper pairs, and branching, all slightly
curving inside in ascending to the borders, craspedodrome.

This fine leaf indicates, by the preserved fragment, its lower part, a
width of thirteen centimeters and a length of about twenty centimeters with-
out the petiole. It is rounded, auricled, or deeply cordate at the base, with
the borders unequally but simply dentate, the short, outside-turned teeth
varying in distance, following the relative position of the secondary veins and
of their branches, which all enter them. The lateral veins are thick, the
lower ones opposite and somewhat more distant from the first pair of veins
above, as these are from the following ones in ascending. Comparing our
figure with that of Heer {loc. cit., pi. xlvi, fig. 5), the remarkable affinity of
tliose leaves appears striking. Their size, their form, and their nervation are the
same. They merely differ in the character of the teeth, which, pointed along
the borders of the American leaf, are obtuse and rhore prominent, in that of
Greenland. But this difference is negatived by Prof Heer, who, in his
description of the specimens from Spitzbergen, which he refers to this species,
and which have the borders of the leaves cut in short, acute teeth, remarks
(p. 57, loc. cit) that the smaller teeth cannot separate this form from that of
Greenland. The locality, Carbon, where a number of Arctic and Alaskan
species have been found, renders more probable the specific relation of this
leaf to those described under this name from Greenland. We have the same
kind also from the Miocene of Roach Hill, Oregon.

Considering its relation, our leaf as fiir as it is known, has the same
degree of affinity to Viburnum platanoides, Lesqx. (pi. xxxviii, fig. 8), as the
former described leaves of Quercus Viburnifolia have with V. marginatum;
supposing, however, that the upper destroyed part of the specimen is rounded
or abruptly pointed, as the direction and the thinning of the upper secondary
nerves seem to indicate, and not lengthened and lanceolate, as it is in fig. G
of pi. xi of Heer (loc. cit.).

Habitat. — Carbon, Wyoming, with Populus Arctica, Paliurus, Colomhi, etc.

DESCRIPTION OF SPECIES— CUPULIFER^. 161

^ III. — Leaves lohnte ; borders entire.

<tucrciis iicgHiidoidcs, Lesqz.

I'lato XXI, Fif;. 2.

Qmrcue negundoides, Lesqx., Annual Keport, 1871, p. 2U2.

Leaf coriacfDUs, short-pet ioled, cordate at the base, enlarging upward to the three-fourths of its
length, where it is palmatoly cut in three short lobes, the middle taper-pointed, the lateral ones shorter
and obtuse; lateral veins seven pairs, equidistant, parallel, craspedodrome.

This peculiar leaf has no relation to any fossil one known to me. Its
appearance is that of some Acer, or of some deformed leaf of JJlmacea or
BelulacecB. I referred it to Quercus on account of its thick consistence and
of a distant likeness to Q. fagifolia and Q. triangularis, Goepp. The leaf is
five centimeters long and three and a half broad in its widest part, under the
point of the lateral lobes; the petiole is slender, seven millimeters long; the
borders entire, slightly irregularly undulate, and the secondary veins all simple,
and at an angle of divergence of 40°, except the lower pair, which is basihir,
slightly more open, with a few brandies on the lower side.

Habitat. — Evanston, Wyoming, in shale under the main coal {Dr.
F. V.Hayden).

Quercus ungustiloba, Al. Br.

Plate XXI, Figs. 4, .5.

Quercus angusHloha, Ludw., Palseont., viii, p. 103, pi. xxxvi, fig. 3.— Heer, Braunk. Pfl. v. Boernst., p. 14,
pi. 1, fig. 8. — Lesqx., Annual Keport, 1872, p. 378.

Leaves coriaceous, oblong-lanceolate iu outline, uarrowed to the petiole, deeply cut on each side
into two opposite, linear, obtusely-pointed, narrow, entire lobes, and a terminal one of the same size
and form.

This species is positively identified by the two specimens we have
figured here, which, though fragmentary, are more complete than those
known from the Oligocene of Europe. The leaves, varying in length between
seven and fourteen centimeters, are deeply cut on each side in two oblique
linear entire lobes, lanceolate to the point, the lowest four to six centimeters
long and about one centimeter broad, with obtuse sinuses. The middle or
terminal lobe is broken above its base, but is apparently as long as the lateral
ones. The secondary veins are scarcely distinct and the areolation tottdly
obsolete. The consistence of the leaves is evidently thick.

Habitat. — Golden, Colorado, where I found the specimen represented
in tig. 4. The other was discovered at the same locality by Prof. F. B. Meek.
11 T F .

162 UNITED STATES GEOLOGlCxVL SURVEY— TERTIAKY FLORA.

DRYOPHYLLUM, Debey.

Leaves lanceolate, oblong or lineai-obloug, generally dentate, rarely entire, penuincrve; second-
ary nerves (in dentate leaves) snbojiposite, numerous, parallel, more or less curved, snbcamptodromo,
forking near tlie borders; (iu eutiro leaves), camptodrome ; r.ervillts transversely decurreut, simple or
forking, joined by braneblets in opposite direction (Sap.).

This definition by Saporta i.s translated t'loni Pal. Vegt't , ii, p. 613.
Schimper adds that tliis type, which originates in the Cretaceous epoch,
appears to have been the precursor of Quercus, Castanea, and Castancopsis;
that his leaves unite the cliaracters of some Quercus and Castanea species
now living in the Lebanon, the Himalaya, and the mountains of Mexico.
He adds that it is regrettable that the species of the Upper Cretaceous of
Aix la Chapelle, from which this genus has been established by M. Debey,
have not been described to the present time, as they could throw some light
on the true place which they take in the vegetable world. In considering
the geological distribution of the species of Quercus, I have already remarked
that I am as yet unable to discern any sufficient character for authorizing a
general separation of the leaves described as Dryophyllum. I have, however,
preserved this section in deference to the authority of European writers,
submitting to their wider experience and more judicious discrimination.

D r y o i> li y 1 1 u lit ( Q u c i- c ii s ) c r «! ii a t u in , Lesqx.
Plate LXII, Figs. 10 and 11.
Dryophi/llum crenatum, Lesqx., Annual Report, 1874, p. 301.

Leaf subcoriaceous, oblong-lanceolate, obliquely truncate to the base; borders deeply, regularly
undulate or broadly crenate-dentate ; lateral veins thick, snbcauiptodrome ; nervilles thick in right angle
to the nerves.

The substance of these leaves is somewhat thick, and their surface
coarsely marked by the nervation. As far as can be seen from the two frag-
ments obtained, the leaves are of medium size, linear-oblong, either equally
undulate on the borders or dentate, with obtuse, broad, short teeth. The
secondary veins are thick, flat, entering the teeth or reaching the borders by
their ends, while a fork under the teeth or quite near the borders follows
them as a camptodrome division, anastomosing with the nervilles. The sub-
division of these in the middle of the areas is mostly in right angle, forming
ultimately small square or indistinctly polygonal meshes. The midrib is flat
and broad. The relation of this leaf is evidently to the Cretaceous D. lati-
foliuni, Lesqx. (Annual Report, 1874, p. 340, pi. vi, fig. 1).

Habitat. — Point of Rocks, Wyoming {Dr. F. V. Hayden).

DESCltlPTKJN OF SPECIES— CUPULIFElt^.. 163

D I* y o i> li y 1 1 u III ( <| ii c i* c ii s ) sub Ta I c a t u m , Lesqz.
Plato LXIIl, Fit;. 10.

Dryophyllmn suhfalcatum, Lcsqx., Annual Report, 1874, p. ;!01.

Leal" subcoriaceouH, liiirai-lanceolato, narrowly taporpointed ; borders regularly serrate, with
short blunt teeth turned upward; lateral veins very oblique, close, parallel, straight to the point of the
teeth.

We have only a fragmentary specimen of this species, the upper half of
a leaf, which, by its form and nervation, seems at first referable to tlie genus
Castanea or to some variety of the Cliestnut-oaks. The nervation is of the
same character as in the former species, however, the upper branch of the
secondary veins passing from near the point of the vcmiis under the sinuses
and closely following the borders, anastomosing with the tibrilla). These arc
very close, percurrent, mostly simple, and rarely branching, distinct, though

thin.

This species is intimately related to Dryoplujllum Dexvalquei, Sap. and
Mer. (Flore de Gelinden), especially to the fragment represented on pi. iii,
fig. 2, differing merely by the shorter, less acute teeth of the borders, the
slightly flilcate form of the leaf, and the close, thin fibrillae.

Habitat. — Point of Rocks, Wyoming (il/r. Wm. Clcburn).*

CASTANEA, Tournf.

The difficulty of discerning the fi)ssil leaves of this genus from those of
Dryophyllum or Quercus renders uncertain the epoch to which its origin is
referable.

One fossil species has been described by Dunker as Castanea Hausmanni,
from the Cretaceous Quadersandstein of Blankenburg, Ilartz, where it is
found with leaves of Credeneria. On this species, however, Schimper observes
. that it indeed resembles the leaves of C. vcsca, the living species so widely
distributed in Europe and North America, l)ut that it could just as well be-
long to Dryopliyllum. From the Eocene of France, one species is described
by Watelet, another is from the Lower Miocene of Southern Europe, and
four are distributed in higher stages of the same formation. In this country,
leaves of one or perhaps two species of this genus are abundant in the Mio-
cene of Oregon. The fragment published here, and apparently identical with

*To this section of Oaks is referable Qua-cns fiirdiinrrie, Kossm., described in Annual Report, 1873,
p. 398, from very fine speeinieiis of the Cascade Mountains of Oregon and of the Spanish Mountains of
California. The specimens belong to the Geological Survey of this last State, and are figured for its
Report.

1G4 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

one of the Oregon species, is from the Upper Miocene of the Parks of Col-
orado.

It is probable that if the fossil leaves of this kind were known from suffi-
cient specimens, the number of the species would be reduced; for the j)resent
flora has only two species: one, C. vulgaris. Lam., or C. vesca, is represented by
a number of marked varieties, often described as species, and distributed over
the whole temperate zone of the northern hemisphere, especially along the
shores of the Mediterranean Sea in Europe, Western Asia, North Africa, and
also in China, Japan, and the United States; the other, C. pumila, exclu-
sively belongs to North America.

Castanea int e r media, Leaqx.

Plate XXI, Fig. 7.
Castanea intermedia, Lesqx., Annual Report, 1874, p. 313.

Leaf snbcoriaceous, long, narrow, linear-lanceolate, gradually narrowed downward; borders
equally sharply serr.ate ; secondary veins slightly curving, open, close, parallel, simple, and craspe-
dodrome.

The fragment represents a leaf, long indeed comparatively to its width,
which is not much above two centimeters; while it appears to have been
more than twelve centimeters in length. The borders are sharply serrate,
the teeth turned upward, thorny-pointed, equal, and each entered by second-
ary veins, which are simple, close, about four millimeters apart, diverging
from the thick midrib at an angle of 50° to 60°. Every trace of areolation
is erased. Comparing this leaf to the other fossil species published, it differs
from all, being only related by its nervation to Castanea Ungeri, Heer (Fl.
Foss. Alask., p. 32, pi. vii, fig. 1), and even in our species the lateral veins are
still more numerous and more open. It has a more evident likeness to the
leaves of our present Castanea pumila, from which it would be undistin-
guishable, but for its linear form and the gradually tapering base. The
simple nervation, the degree of divergence of the veins, the form and sharp-
ness of the teeth, are all alike. The common C. vesca of the North has
longer leaves, more distant veins; but these leaves, at least when young, and
especially those of the bushy shoots, are gradually narrowed to the petiole,
and linear-lanceolate, like the fossil one. This therefore appears interme-
diate between both living species of North America.

Habitat. — Middle Park, Colorado {Dr. F. V. Hayden).

DESCKIPTION OF SrECIES— SALICINE^. 165

SALICINEJE.

SALIX, Liim.

Together with their narrowly lanceolate form, which they have in
common with many other plants, the leaves of Willows are recognized by
the following characters: — The middle nerve is strong, continued below
the base of the leaves into a short petiole ; the secondary veins are numerous,
close, parallel, generally at an open angle of divergence at or near their point
of union to the midrib, more oblique in coming near to the borders, where
they unite in continuous festoons by their curved points. These lateral veins
are generally intermixed with shorter tertiary ones, whose angle of divergence
is often different, and which, branching in the middle of the primary areas,
form, by subdivisions in right angle, first, large rectangular areolae, and then,
by multiple nervilles, a net of very small irregular meshes. As said above,
the form of these leaves is more generally narrowly lanceolate, more or less
rapidly narrowed or rounded to the petiole, but sometimes also broadly ellip-
tical or oblong-ovate, even ovate-subcordate. The borders of the leaves are
entire or simply crenate, dentate, or serrate.

The origin of the genus seems legitimately referable to the Cretaceous
period. In vol. vi of the Reports of the United States Geological Survey of
the Territories, p. 60, pi. v, figs. 1-4, four leaves from the Dakota Group of Ne-
ltraskaare<lescribed and figured as Salix protecEfolia,'Lesqx., whose reference
to the genus is so clearly indicated by the characters of form and nervation
that it seems ind ubitable. Prof. Heer, whose sagacity of determination of fossil
leaves is so remarkable, has described, from the same fovmatwn, Salix nervil-
losa, a species of a different type of this genus; and Prof Newberry also
has referred to Salix, and from tlie same formation of Nebraska, a number of
leaves which he considers as representing four different species. From the
European Cretaceous of Blankenburg, Hartz, Dunker has described Salix
Hariigii, a leaf which, according to Schimper, may belong to a species of
Quercus of the section of Q. phellos; and Heer recognizes a fine species,
Salix Ga;lziaria, in the formation of Quedlinburg. These are sufficient
authorities in proof of the antiquity of this genus. Schirnpcr remarks, in
Pal. Veget., ii, p. 6(;3, that if the attribution of certain saliciform leaves of the
Cretaceous torniation is correct, we have, in tlu; type of the Willows, one of
the most ancient ibrms of the subdivision oi the dicotyledonous A ugiosperms.

IGG UNITED STATES GEOLOGICAL SURVEY— TERTIAEY FLOIJA.

He adds : — "That some species are also iiidieated in the Eocene, lni( they are
as uncertain as fiiose of the Cretaceous, and it is only with the first deposits
of the Miocene epoch that every kind of doubt is put aside on lliis <|ueslion,
for'then we have not only numerous leaves of which the generic assimilation
is indubitable, but also flowers and fruits, which give to the botanists the
possibility of determining the subdivisions which had some representatives
at that time." It is certain that, according to the remarks of the same author,
it is during the Miocene period that the Willows have attained their maxiinum
of specific evolutions. But this extraordinary development of the generic
type is sufficient to \no\e the great antiquity of the genus, if even the pres-
ence of numerous leaves, as positively identifiable as those of the Miocene,
had not been found in the Cretaceous of both continents. To this we have to
add the continuity of the records from the Cretaceous through the Eocene and
the Oligocene. Schimper describes fifty-seven fossil species of Willows, two
of which are of doubtful generic relation.. Adding to this -S. GcEtziana of
the Cretaceous of Quedlinburg, we find twenty-three species referable to
the section of dentate leaves. Of these, none is Cretaceous and none is
exclusively North American. Four are Eocene, one Oligocene, seventeen
Miocene, and one Pliocene or Quaternary. Salix varia/is, common in Europe,
has been recognized in the Miocene of Alaska and in the upper divisions of
the same formation in California. S. Lavateri and S. macrophijUa are also in
the Miocene of Europe and in that of Alaska. Of the thirty-six fossil
species of Willows with entire leaves, eight are Cretaceous, six from the
American Dakota group, and two from Europe. No Eocene species is known
from Europe, but we have in the American formation one Miocene European
Willow, /Sa/i.r Integra, Goepp., discovered in the Eocene strata of Black Buttes,
Wyoming Territory, and one species peculiar to this continent,/S'.teZ)M/an.s-, fiom
the same Eocene formation in the State of Mississippi. -S'. angustata, S. elongnta,
and S. media are European and American Miocene species, while two others,
S. Rhcana and 8. Groenlandica, belong exclusively as yet to Greenland. An
American form, S. Worthenii, is from the chalk bluffs of tiie Mississippi, a
Pliocene formation. It is represented as yet by a single leaf and may be
recognized, when other specimens are obtained, as identical with one of
our living species,

DESCRIPTION OF srEClES— SALIClNEiE. 167

Salix Integra, Goepp.
Plato XXII, FiKs. 1,2.
Salix inlegra, Goopp., Schoss. Teit. Fl., p. 25, figs. 6, 10, 14.— Lesqx., AdiukvI Report, 187:?, p. 397.

Leaves Bmall, entire, oblong, lanceolate-acuminate, gradually narrowed to a short petiole; lateral
veins thin, close, intermixed with short tertiary ones; primary areolni large, (inadrate.

The leaves of this species are generally small. Those figured hy Goep-
pert vary from one to four and a half centimeters long and proportionally broad ;
those described here l)eing therefore of the largest size. The nervation, as
seen in figs. 1 and 1 a, enlarged, is that of the genus, agreeing entirely with
that represented by the European author, especially in fig. 1. The only nota-
ble difference of characters is in the more gradually tapering acumen of the
Anicriean leaf, and this is not of specific value for a species whose leaves arc
trreatly variable in size and form, some of them being obtuse or even half-
round at the point, wiiile some others are sharply acuminate or pointed.

From Heer's remarks on this species in Fl. Tert. Helv., iii, p. 175, the
leaves represented by Goeppert {loc. cit., figs. 2, 3, 4, 8, 9), with more distant
and stronger secondary veins, without intermediate tertiary ones, arc refera-
ble to Benzoin attenuatum, Heer {loc. cit., ii, p. 82, pi. xc, fig. 10), to which
he refers also the leaves described in the same volume (p. 32, pi. Ixviii, figs.
20-22) as Satix Integra. We have apparently here the two fi)rms described
and figured l)y Goeppert, one (fig. 1) identical in characters with fig. 1 of the
Schossnitz Flora, as remarked above, and one (fig. 2) showing quite as dis-
tinctly the characters of fig. 2 of Goeppert's, for it has the secondary veins
more distant and no trace of intermediate tertiary veins. The habitat of
these leaves being the same, I consider the difl'erence as merely apparent,
resulting from the more imperfect state of preservation of the leaf in fig. 2.
Specimens from Golden, and also from the Miocene of Oregon, have the same
character of nervation, and a short, naked petiole. The typical characters of
Benzoin attenuatum, viz., the border base decurrent along the petiole, which
is therefore winged, and the lower lateral veins at a more acute angle of
divergence, and following the borders to the middle of the leaf, as indicated
in Heer's Joe. cit., pi. xc, fig. 10, are not seen upon any of the American
leaves.

Habitat. — Black Buttes, Wyoming, the two specimens figured here;
Golden, Colorado, a more imperfectly preserved one.

168 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

S a I i X media, Heer.
Plate XXII, Fig. 3.

Salix media, Heer, Fl. Tert. Helv., ii, p. 32; iii, p. 17.'), pi. Ixviii, figs. 14-19. — Al. Br., Stizenb. Verz., p. 78. —
Ludw., PalajoDt., viii, p. 93, jil. xxviii, figs. 1-4. — Lesqx., Anuual Report, 1871, Supplement,
p. C; 1873, p. 411.

Leaves entire, rounded at the base to a short petiole, linear, lanceolate-pointed, or gradually
taperirg to the point; secondary veins open, camptodrome.

The figure represents one of two leaves upon the same* specimen, the
one not figured having the same character, rounded at the base, linear-lan-
ceolate to near the point where it is broken, and without any trace of secondary
veins. These are not generally visible upon the upper surface of the leaves.
Of those represented by Heer, one only (fig. 16) has its nervation discernible;
it is the same with those described by Ludwig. The identity is recognized,
however, by the form of the elongated, lanceolate, entire leaves, rounded to
the short petiole, a character distinctly marked in all the figures of this spe-
cies published by European authors, except in Ludwig's loc. cit., fig. 2, which
seems referable to another. The specimens from Elko agree also exactly
in the same characters, and their secondary nervation is still more obsolete
than in those of Green Kiver.

Habitat. — Green River, Wyoming (Dr. F. V. Harjden); Elko Station,
Nevada {Prof. E. D. Cope). *

Salix angusta, Al. Br

Plate XXII, Figs. 4, 5.

Salixangusta,Ai. Br., Stizenh. Verz., p. 77. — Heer, Fl. Tert. Helv., p. 30, pi. Ixix, figs. 1-11. — Lesqx., Annual

Report, 1H71, Supplement, p. 6; 1872, p. 405.
Salix anguntifoUa, Al. Br., in Buckl. Geol., p. 512.
Salix avgustissima, Al. Br., in Leonh. und Bronn. Jahrb., 1850, p. 169. — Ung., Gen. et Sp., p. 418.

Leaves entire, generally very long, and comparatively narrow, at least twelve times longer than
broad, linear-lanceolate, gradually tapering upward to a long acnmen and downward to a short petiole;
middle nerve thick; secondary veins close, numerous, more distinct than in the former species.

This sjjecies essentially diifers from the former l)y narrower, much longer
leaves, either gradually narrowed to the point and fo the petiole, or linear,
with borders nearly parallel in the middle, tapering into a long point, and
more abruptly rounded to the petiole; the middle nerve is broader, and the
secondary veins more; distinct, in a more acute angle of divergence. The size
of these leaves is very variable, at least for the length, one of our fragments
exposing a leaf of about fourteen centimeters long, while the other specimen
is of a leaf preserved entire, and measuring only six and a half ccnlimelcrs,
though scarcely broader. Another specimen of the same sj)ecies and from

DESCRIPTION OF SPECIES— SALTOINE^. 169

the same locality has oiil}' one-half of a leaf, also of the same width, twelve
millimeters broad in the middle, showing the under part, with secondary
veins distinctly marked, like their divisions, and the snrface evidently villous,
or bearing the impression of a coating of hairs Prof Ileer, in his description
of this species, compares it to S.viminalis, Linn., a living and common species
of this country, remarking, however, that the s[)ecimens do not indicate
whether the leaves are villous, as in the living species. This villosity is not
apparent upon well-preserved specimens received from Oregon.

Habitat. — Green River Station, Wyoming, above fish beds {Dr. F. V.
Hat/den). One fragmentary specimen from six miles above Spring Canon,
Montana, appears referable to the same species.

Salix elougata, O. Web.

Plate XXII, Pigs. 6, 7.

Salix elongata, O. Wel>., Palaeont., vol. ii (Separ. Abdr.), p. 63, pi. ii, fig. 10. — Heer, Fl. Tert. Helv., ji,

p. ;il, pi. Ixix, figs. 15, 16.— Lesqx., Auiiual Keport, 1872, p. 372.
Salix longisaima, W< as. & Web., Palaeont , iv (Separ. Ab<lr.), p. 30, pi. v, fig. 6.

Leaves entire, long, lanceolate, gradually narrowed from the middle toward the point and toward
the petiole; middle nerve thin ; borders apparently revolute or reflexed ; secondary veins open, distant.

The leaves of this species, at least judging from the fragmentary speci-
mens which I refer to it, are larger than those of the former, more distinctly
lanceolate and more rapidly tapering to the point, coriaceous, with the veins
more di.stant, irregular in their relative position, and with a very narrow
midrib. These fragments have the same characters of tbrm and nervation,
as far, at least, as this may be discerned, as the fine leaf described {loc. cit)
by the author of this species. The leaves figured by Heer as representing
it are, however, narrower, linear-lanceolate, of the same form as those of
Salix longa, Al. Br., but distinct by the narrow midrib.

Habitat. — Elko Station, Nevada {Prof. S. W. Garman).

POPULUS, Linn.

Represented merely by more or less fragmentary leaves, as it is as yet
in our American fossil ilora, the genus may be characterized as follows: —

Leaves of various forms; broadly ovate, round or lanceolate, generally
cordate or rounded at the base ; long-petioled ; with borders entire, crenulate
or dentate; palmately nerved from the top of the petiole or from a distance
above the border base. The prinuiry and secondary nerves are caniptodrome,
anastomosing outside of the bends with ncrvilles, either passing directly to

1 70 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

the tectli or divisions of the borders, or curving along them. Tht; areas,
divi(h'(l in right angle by fibrillae, form primary irregularly s(jnare meshes,
and the veinlets branching in various directions ultimately constitute a very
small polygonal areolation. One pair of slender marginal nerves jjass from
the top of the petiole I)elo\v the primary lateral veins, joining their branches
by veinlets. These general characters are, however, modiiied in many
ways Tlie nervation, especially, is very variable and complex, for, by the
addition of one or two pairs of nerves under the primary one, it becomes
five- or seven-palmate, while in other leaves, as for example in those of Popu-
Itis balsamifera var. angusllfoUa, a species especially common in the valleys
and along the base of the Rocky Mountains, the lower primary nerves are
alternate, of the same thickness as the secondary ones, all equidistant and
equally branching underneath, representing thus a pinnate nervation, rendered
remarkably similar to that of the leaves of Salix by the interposition of short
tertiary veins, traversing the lower side of the areas and dissolving in their
middle by subdivision in nervilles. In other cases, as in the so multiform
leaves of Populus alba, an introduced species, too common in this country,
the primary lateral nerves, two or three pairs, stronger and more branching
than the secondary ones, go straight to the |x»ints of acute lobes or of larger
teeth formed by expansion of the laminae, are thus craspedodroiiie, as well as
some of the secondary ones, and give to the leaves some of the characters
and the appearance of leaves of PI at anus.

These variations in the characters of the leaves, even of the same species,
render their determination very difficult and somewhat unreliable; for the
paleontologist has rarely for examination and study a series of specimens so
numerous, and in such a perfect state of preservation, as are those which
have served for the preparation of the admirable monograph of this genus
in Heer's Fl. Tert. Helv., where are represented branches bearing leaves
of various forms, some entire, some dentate, besides buds, bracts, catkins of
ilowers, and seeds.

The difficulty of determination of the leaves of Populus may, in a certain
degree at least, account for the great ditfercnce between the number of spe-
cies of this genus known as living at our time, eighteen only,* and the fossil
ones, of whicii Schimper's Tal. Veget. describes sixty-two, considering, how-
ever, nineteen of them as (h)ubtful Of the species of Poplars living at our

"According to tbo mouograj)!! of Wesniael, in Do Caudolle's Prodiomus.

DESCRIPTION OF SPECIES— SALIGINE^. 171'

time, the North American continent lias tlie largest nnniber, nine, mostly
inhabiting the cold and mountainous regions. Some, however, have a wide
range of distribution: Populua tremuloides, Michx., and P. Canadensis, Desf.,
lor example, extend between the Pacific and the Atlantic from Canada lo
Louisiana and New Mexico. One species, P. Mexicana, Wesm., is ))eculiar
to Mexico, descending as far soutii as Tampico. Anotlier, P. trichocarjia,
Torr. & Gray, is limited to California. Three species t»re common to Europe
and Asia ; even one of them, P. alba, L., has been found in Algeria. China
has two species, Japan one, and three belong to the Orient.

Considering the fossil species, and leaving out (hose which have been
separated as doul)tful, we find described in Schimper's synopsis forty-two, of
which two only are Cretaceous: Populuslitigiosa, Heei-, and P. ekgans, Lesqx.,
both from the Dakota group. To these we should add, though not mentioned
in the synopsis, P. cydofhijlla, Lesqx., and P. Lancasfriensis, Les(|x., from the
same formation, with P. hyperborea, Heer, P. Berggreni, Heer, from the Upper
Cretaceous of Greenland, and P. primceva, Heer, from the Lower Cretaceous
of the same country. Dr. Newberry has described, in his notes on tiie later
extinct floras, four species, two of which are considered by himself as douljt-
fully referable to this genus ; another, P. microplnjUa, is of uncertain relation,
and the fourth, P. elliptica, of a Miocene type, seems to have been referred
to the Cretaceous by a mistake caused by misplacement of labels or of speci-
mens. Though it may be that we have already seven Cretaceous species of
Populus, one of which, P. primceva, represents, by leaves and scales of seeds,
the only dicotyledonous j)lant found in the flora of Comes, a Greenland Lower
Cretaceous flora, composed of seventy-five species of Filices, Selaginece, Cyca-
dea, ConifercB, and a few monocotyledonous, mostly of Jurassic or Wealden
types: this Populus is thus the oldest dicotyledonous plant known as yet.
From this, we cannot be surprised to find the generic tyjie already ])re-
ponderant at a higher stage of the Cretaceous, that of the Dakota group, whose
flora is connjosed mostly of dicotyledonous plants. No speci(;s of Populus,
however, has been described to this time from the Cretaceous of Europe.
This is a remarkable fact, rendered more striking by the scarcity of rejire-
sentatives of the same genus in the Eocene of that continent, which has
until now only two species, both from the lower members of the forniation,
Sezanne and Bellcu. Besides five Tertiary species described by Dr. Newberry
iVuni the Fort Union and Yellowstone Lignitic, and wiiich may be Eocene,

1 72 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA,

\v(> havo six others, positively referable to this formation, one of wliirli is from
Vancouver, Iwo are common to the Mississippi and tlie Colorado Lii,niitic, and
two have been obtained from the lowest Tertiary strata of Point of Rocks,
which immediately rest upon the Cretaceous, and whose flora still preserves
a few representatives of Cretaceous types. In the Miocene of Europe, how-
ever, the genus takes a large predominance, twenty- eight species being there
described from this fwmation. To the present time, we have only ten ; but
considering the number of Tertiary species known from both continents,
this indicates about the same pro])ortion in the geological (iistril)uti()n.
Seven of the American Miocene species arc common with Europe; and, of
these, three are also found in Alaska They represent especially the Middle
Miocene, predominant in Oregon, and on the eastern side of the liocky
Mountains, especially at Carbon, a division from which comparatively few
materials have been obtained until now. As remarked in the descriptions
of the species, the relation of some of them to types of the present flora of
this country is distinctly recognized.

§ I. — MarginatcR.

Pop II I IIS lat ior, Al. Br., var. cordif olia.

Plate XXII, Fig. 8.

Populus latior cordif olia, Heer, FI. Tert. Helv., p. 12, pi. Iv.— Ludw., Falajont., viii, p. 91, pi. xxvi, fig. 7. —
Lesqx., Annual Report, 1871, pp. 287, 289.

Leaves nearly round, broadei than long, sbort-JSointed, subtruncate at base, wavy-margined;
primary veins three, camptodromf.

The form of this leaf and the characters of the borders arc referable to
those described by Heer, and the nervation, somewhat abnormal by the
absence of one of the primary nerves joining the midrib a little above the
border base, is comparable to that of the leaf figured by Ludwig. The upper
j)art of the leaf is destroyed. But another specimen, whose nervation is not
quite as distinct, described in the same Report, p. 287, as Populus latior var.
transversa, Heer, represents a smaller, short-pointed leaf, with more deeply
marked undulations of the borders, which is apparently referable to the same
variety.

Habitat. — Washakie Station, Wyoming {Dr. F. V. Hayden).

DESCKIPTION OF SPECIES— SALICINE^. 173

P o p u I II s »> II b r o t II n d n t :i < Lesqx

Plate XXIV, Figs. n-8.

PopuJua suhrotmda, Leeqx., Aiiiiiial Report, 1807, p. 196 — Sclip., Pal. Y6g6t., ii, p. 680.
Populus attenuata, Al. Br., Lfsqx., Annual Report, 1872, pp. 'JHG, 3s9, 39a.

Loaves long-petioled, nearly round, as broad as long, Bubtruncate at the base, abruptly pointed,
acutely dentate; nervation tripalmate, canii)todrouie.

The first of these leaves (fig. 8) was considered as representing u new
species, distinct from P. attenuata, hS.. Br., especially by the sharp, turned-up
teeth of the borders and the more abruptly narrowed or truncate base. It
has, however, a great likeness to the figure given of this last species in Heer
(Fl. Tert. Helv., pi. Iviii, fig. 1). The examination of other specimens from
Carbon, though more fragmentary, confirms the first opinion in regard to the
specific difference. The primary lateral veins separate from the midrib a
little above the base, being much branched underneath and comparatively
thick ; the secondary ones are somewhat higher up, mostly simjjle, and the
areas are cut by nervilles in right angle to the veins. These characters are
identical with those of P. attenuata; but the round shape of these leaves is
different and the teeth always more acute than in the European species.
This form is related to the North American Populus monilifera, Ait.

HABiTAT.-^Rock Creek, Laramie Plains, Wyoming {Dr. F. V. Hayden);
Cai-bon, Wyoming, where it is not rare, and found in both the beds of shale
above and below the main coal. The specimen in fig. 7 is from Evanston,
Wyoming, procured by Dr. Hayden

Populus in elan aria, Heer

Plate LXIV, Fig. 5.

Populus melanaria, Heer, Fl. Tert. Helv., ii, p. 16, pi. liv, fig. 7 ; Ivii, fig. 1.— Lcsqx., Annual Report, 1874,
p. 302.

Leaves with a long, slender petiole, broadly deltoid or subtrnncato at the base ; borders acutely
serrate; primary lateral nerves emerging from a distance above the base of the leaves, with a pair of
marginal veinlets underneath.

This leaf, considering what can be seen of it by the fragment, which
represents merely its lower half with the long slender petiole, the distinct
nervation, and a few of the border teeth, exhibits characters in accordance
with those described above, and translated from Schimper's Pal. V(ig(it., ii,
p. 684. It agrees especially with the fig. 7 of Heer, loc. cit. This author
remarks that the species essentially differs from Populus latior van subtrun-
cata by the position of the lateral primary nerves at a distance from the

174 UNITED STATES GEOLOGICAL SUKVEY—TERTIAKY FJ.ORA.

border base of the leaves. In our specimens, as seen in the figure, the dis-
tance is still greater than in that of the Fl. Tert. Ilelv. Prof. Ileer remarks
also that he has had for examination a largo number of specimens of the
same species, but that in all, except one, which he has figured, the upper part
of the leaves is destroyed, as it is in ours. lie mentions, as distinctive char-
acters, the acutely serrate borders of the leaves, and the middle nerve thicker
than the lateral ones, the same as seen upon our specimen. I believe, there-
fore, that the identification of this leaf with the European species is fully
authorized. Heer considers this species as allied to the living P.dilatata, Ait.,
and P. nigra, Linn.

Habitat. — Point of Rocks, Wyoming {Dr. F. V. Hayden).

Popiilus melanarioides, Lesqx.

Plate LXII, Fig. 5.

Populus melanarioides, Lesqx., Annual Report, 1874, p. 302.

Le.aves long-putioled, subcori.aceous, nearly round, Bubtruncate at baee ; borders entire, undu-
late ; nervation teruate from above the base ; secondary veins, two pairs, at a great distance from
the primary ones, these much branched outside, the others simple ; divisions passing to near the bor-
ders or entering them.

By its subcoriaceous substance and the long petiole of the leaves, this
species is related to the section of the Trepid<z (Trembling Poplars). As in
Populus tremulcefolia, Sap. (fit., o, 2, p. 26, pi. iii, fig. 4), to which our species
is allied, the veins and their branches pass through the areas to very near the
inflated borders, which they seem to reach, but along which they are really
curved. Our leaf differs from those of this last species merely by less distinctly
undulate borders, by the higher position of the primary lateral veins above
the base, and by the great distance of the secondary less numerous veins.
These two last characters are, however, of no moment in the specification
of Poplar leaves, as can be seen in the examination of a few specimens of
Pupalas alba. In Ibssil species, also, as in P. Mosdliensls., Sap. (Et., 8, 2,
p. 30, ))1. ii, figs. 6-8), the three leaves which represent it have each a difierent
type of nervation. The relation of our leaf to that described l)y Saporta from
the Tertiary of Provence may be therefore more intimate than it appears
from the comparison of a single leaf. It is also comparable to Populus helia-
dum, Ung., by its form, and to P. melanaria, Heer, by its nervation.

Habitat. — Point of Rocks, Wyoming ( Wm. Cleburn).

DESCRIPTION OF SPECIES— SALICINEJE. 175

§ II. — Trepidce.
P o p II I II s t n g: c r i , sp. uov.
Plate XXIV, Fig. 5.
Popvlua heliadum, Ung., Lesqx., Annual Report, 1873, p. 397.

Leaves siibcoriaceous, loug-petioled, bidadly ovate, obtuse; borders entire, slightly nndulate;
secondary veins etiuidlstant with the primary ones.

This leaf, rather membranaceous than coriaceous, with a sh'nder, appar-
ently long, petiole, has the form of Populus mutabilis repnndo-crcnata, Heer,
as represented in Fl. Tert. Hclv., pi. Ixii, fig. 1, and tlie nervation of 7'. halsa-
moides, Goepp., as in pi. Ix, fig 2, of the same work; the primary lateral
veins and the secondary ones being equidistant, and of the same thickness.
From the first of these species it differs by the nervation; from the second,
by the entire borders of the leaves. From P. heliadum, Ung. (Fl. v. Sotzka,
p. 37, pi. XV, fig. 7)j to which I referred it formerly, it differs equally by the
more numerous equidistant nerves, the less enlarged form of the leaf and the
borders less distinctly undulate, Unger describing them as undulate-dentate.
Other specimens may indicate a more positive relation. Except for its slender
long petiole and an apparently merely membranaceous or thin consistence,
this species should go to the section of the Coriaceous Poplars, and be per-
haps referred to P. monodon, represented in figs. 1 and 2 of the same plate.

Habitat. — Golden, Colorado.

§ III. — Balsafnifce.

Populus laevigata, Lesqx.

Plate XXII, Fig. 9.

Populus laxigata, Lesqx., Annual Report, 181)9, p. 195.
Populus cequalis, Schp.,* Pal. V(;g6t., p. 093.

Leaves largo, cordate, aeimiinate, rounded to the base, dentate ; teeth acute, turned outside; midrib
much thicker than the lateral ones; surface smooth or polished.

This fine leaf, thirteen centimeters long, and eleven centimeters wide a
little below the middle, is so closely allied to Populus bulmmoidcs, Goepp.,
extensively distributed in the European Miocene, that it can only be separated
by the form of the teeth of the borders. The size of the leaf is equaled
by that of some of those published by the authors as pertaining to P. halsa-
moides; the middle nerve is in both species comparatively thick; the gen-
eral nervation and the areolation are the same. Many fragments seemingly

* The name Populus aqualis mentioned was originally Kent fo Prof. Schiuiinr with a i)hotographed
figure of the specimen, and changed aftervrard without knowing that he had described the sxjecies.

176 U^riTED STATES GEOLOGICAL SDEVEY—TEKTIARY FLOKA.

referable to F. balsamoides have been found in different parts of the North
American Lignitic formations, at Golden, Black Buttes especially; one also is
from Mount Diablo, California; but these fragments are all too insufficient for
positive identification; and the difference in their stations seems to contradict
an identity with the Miocene species. Good specimens of P. balsamoides
have been, however, procured from the Miocene of Coral Hollow, California,
and have been figured for the Fossil Tertiary Flora of that State. Heer has
described it from specimens of Alaska.

Habitat. — Kock Creek, Laramie Plains, Wyoming {Dr. F. V. Hayden).

Populus Zaddachi, Heer.

Plato XXII, Fig. 13.

PopuUs Zaddachi, Heer, Fl. Foss. Arct., i. p. 98, pi. vi, figs. 1-4; xv, fig. li; ii, p. 468, pi. xliii, fig. 1.'');
xliv, fig. 6; Fl. Foss. Alask., p. 26, pi. ii, fig. 5a; Spitz. Mioc. Fl., p. 55, pi. ii, fig. 13c; x,
. fig. 1 ; xi, fig. 8 a ; Mioc. Bait. Fl., p. 30, pi. v, vi, figs. 1-7 ; xii, fig. 1 c— Lesqx., Annu.al Report,
1871, p. 292.

Leaves ovate, obtuse, rouDded or subcordate at the base, crenate, palmately five-nerved ; upper
primary lateral nerves at an acute angle of divergence, ascending to above tbe middle of the leaves.

This species, very common in the Miocene of the Baltic, of Greenland,
and of Alaska, has, until now, few representatives in this country and none in
the Lower Lignitic. It is found with the following species in the Upper
Tertiary measures, especially abundant in the Pliocene of California. By
comparison, it will be seen that our leaf has exactly the characters of those
figured in Fl. Bait., pi. v, figs. 2 and 5, the lower lateral veins being effaced
and very short, and the base of the leaf being rounded in narrowing to the
petiole The general nervation is also the same, the primary upper lateral
nerves being much thinner than the midrib, indeed, of the same size as the
secondary ones, which are at a comparatively great distance from the basilar
nerves and at a much more open angle of divergence. Tiie teeth of the
borders are of the same form, obtuse or half-round, turned upward. The
size of the leaves of this species is very variable; ours measures six centi-
meters long and nearly five centimeters broad. Heer figures one from the
Baltic Miocene, sixteen centimeters long and twelve centimeters broad.

Habitat. — Green River Station, Wyoming, above fish beds (Dr. F. V.
Hayden).

DESCRIPTION OF SPECIES— SALICINE^. 177

Populiis Richardson I, Heer.

Plate XXII, Figs. 10-1-,'.

Poputua Sichardsoni, Hcer, Fl. Foas. Arct., i, p. 98, pi. iv, figs. l-!r>; vi, figs. 7,8; xv, fig. 1 c; ii, p. 468, pi.
xliv, figs. 7-9; Iv, fig. 3 b; Si)itzb. Mioc. Fl., p. 54, pi. x, figs. 8-12. — Lesqx., Annual Report,
1873, p. 411.

Leaves broadly ovate or nearly round, truncate or slightly cniarginate at the base, deeply irregu-
larly crenate, palmately five-nerved; primary lateral nerves about as thick as the midrib, the upper
ones at an acute angle of divergence, ascending to near the point, branching in right angle or in a
broad angle ot divergence.

As recognized from European specimcn.s, tlie leaves of this species are
very variable in size, mostly broadly oval or round, even broader than long, with
their borders deeply cut in irregular round teeth, a character which is clearly
defined in figs. 11 and 12 of our plate. They are described as acuminate, and
appear to be so in two figures of llic author; but, in others, they arc evidently
obtuse, a character remarked upon the leaf of our fig. 12. There is a great
difference in the relative position of the primary lateral nerves, which gen-
erally come out from the top of the petiole, as in fig. 10, but which in fig. 11
become distant, the internal ones being far above the base of the lamina and
of the lower ones. That this difTerence is unimportant for specification is
evidenced by the nervation of the leaf in fig. 12, whose basilar nervation is
intermediate between that of the two others. This species is, by its char-
acters and its habitat, closely allied to the former. It is one of the most
common of the Artistic Regions, but has not been found either in the Baltic
or the Alaska Miocene.

Habitat. — Elko Station, Nevada (Prof. E. D. Cope). It is represented
in the collection by six specimens, being therefore abundant at the locality.

§ IV. — Coriacea.

Pop ti Ins inntabilis, var. f. o v a I i $ , Heer.

Plate XXIV, Figs. 3, 4.

Pop\iU» mutabili» f. <n>aZi«f, Heer, Fl. Tert. Helv., ii, p. 22 ; i, pi. i, figs. 1,2; ii, figs. 2o, i; ii, pi. Ix, fig.
12 6; Ixi, figs. 1-3,6,9; Ixiii, fig. 4.— Lesqx., Annuiil Report, 1871, p. 292; 1872, pp. 401, 40.5;
1873, p. 307.

Leaves coriaceous, long, petiolate, oval, pointed, narrowed or rounded to the petiole; borders
entire ; nervation three- or five-palmate.

The references to the descriptions in the reports apply to different
varieties of this species, some of which have not been figured, on account of
the deficiency of the specimens, or from their exposition upon large rocks
which could not be displace'd or broken. These varieties arc so numerous

12 T F

178 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

and so intimately allied by their characters that it is extremely hazardous lo
identify single leaves with one of fhe eiglit subdivisiuus of this species in
Heer's monograph. The 6rst of our leaves (fig. 3) is subcoriaceous, entire,
oval, obtusely pointed, and narrowed to the long slender petiole in the same
degree as to the point. The nervation is three-palmate from above the base,
obscure, indeed ; for the lateral primary nerves, narrower than the midrib, are,
as well as the secondary ones, scarcely discernible. For its shape and nerva-
tion, it is like the leaves in Heer {loc. cit., pi. i, fig. 1 , and pi. ii, fig. 2 h). The
other leaf, more distinctly coriaceous, is lanceolate, gradually enlarged toward
the base, and rounded to (he petiole; the borders are entire, and the nerva-
tion, five-pahnatc from above the base, is quite distinct and has the characters
of Poplar leaves. The shape is, however, different from any of the leaves
figured by Heer, its nearest affinity being with fig. 12 h of pi. Ix and fig. 1
of pi. Ixi. This leaf therefore may represent a new species of the division
of the Coriacece. A very fine specimen of the var. e. repnndo-crenata, Heer,
a leaf fully preserved, sixteen centimeters long, without the eight centimeters
long petiole, and eight centimeters broad toward its round truncate base, was
exposed at Evanston upon a block of sandstone prepared for building. I
could only make a sketch of it, and l)y comparison found it perfectly similar
to the fine leaf in Heer {loc. cit., pi. Ixii, fig. 4). Other fragments were recog-
nized imbedded with bones of the Saurian at Black Buttes.

Habitat. — Evanston, Utah, as represented in fig. 3. Six miles above
Spring Canon, Montana, the leaf of fig. 4 {Dr. A. C. Peak). Black Buttes,

Wyoming, etc.

Populus arctica, Heer.
Plate XXIII, Figs. 1-6.

FopuluB ai-cHca, Heer, Fl. Foas. Arct., i, pp. 100, 1.37, 158, pi. iv, figs. 6 a, 7; y, vi, figs. '),6; viii, figs. .5,6;
xvii, figs. 5 b, e; xxi, figs. 14, 15; xxx, fig. 9; ii, p. 408, pi. xliii, fig. 15 a; liii, fig. 4; Spitz.
Mioc. Fl., p. 55, pi. X, figs. 2-7; xi, fig. 1; xii, fig. 6e. — Lesqx., Annual Report, 1871, pp. 289,
aOO; Supplement, p. 9 ; 1872, pp. 385, 401 ; 1873, p. 406.

Leaves tbickish or coriaceous, round-oblong, or sometimes enlarged in the middle, and broader
than long, .abruptly short-pointed, n.arrowcd or truncate to the petiole; borders entire, undulate or cre-
nate; nervation five-palmate from the top of the petiole ; upper primary nerves as thick as the uiidrib,
much branching outside, passing up in an acute angle of divergence and curving inside toward the
point of the leaves ; secondary veius thinner, and distinct from the primary ones.

Comparing the figures which represent this species, it is evident that
their characters, however different they may be, all agree with those of the
leaves described by Heer under this name. Fig. 1 and fig. 3, two leaves of
the same size and of the same form, enlarged- in the middle, about seven

DESCRIPTION OF Sl'ECIES— SALICINEiE. 179

cciilitneiers long iind as broad, witli LMitiie holders, a triiiicato or rounded
base, abruptly short-pointed, seein like a copy of pi. v, Ug. .'5, of Ileer (/oc. cit.).
Tiiis is only a little larger, but tlie ntM-vation is exactly the same. Our fig. 3
has the i)or(!crs (;ntire, not even undulate, but the same character is clearly
marked upon the leaf of pi. x.\i, fig. 14, of the Arctic Flora. The transition
from our fig. 8. a leaf broadly cuneate, to fig. h, narrowed to the base, is indi-
cated l)y the intermediate form of fig. 4, and the leaf of fig. C, with crenate
borders, finds its typical analogy in that of pi. x.\i, fig. 15, of Heer, an analogy
indicated also by the habitat, as the leaves from Troublesome Creek, rep-
resented in figs. 3 and G, are mixed with other intermediate forms upon the
same specimen. It is well, however, to remark the similarity of this last
figure with that of P.paleomelas, Sap. (Et., ii, 2, p. 123, pi. 7, fig. 10), which
differs oidy by the primary nerves being more slender, not curving inside,
and the secondary veins descending lower. The small leaf of fig. 2 is com-
parable to those of the following species, but it has the strong, distinct nerva-
tion of P. arctica, represented in the Arctic Flora l)y leaves still much smaller
than this, and also the peculiar, wrinkled, somewhat shining surface of the
species.

Habitat. — Troublesome Creek, Colorado, Mr. Mitchell, who collected
from the locality twelve specimens only, half of which represent this species,
the others Platanus offinh. Carbon, Wyoming, where the species is common
with Acer, Platanus accroides, etc. Green River, Wyoming, with species of
Ficus of Miocene character. Though abundant in Greenland and Spitzber-
gen, it is not described from Alaska. It has not been seen until now in the
specimens from Oregon and of California.

Populus dccipicus, Lesq:s.

PlateXXIII, Figs. 7-11.
Populun ded-piena, Lesqx., Annual Report, 1872, p. 385.— Sohp., r.-il. V^g^t., iii, p. 590.

Leaves small, coriaceous, entire, broadly rhoraboidal, deltoid to the point, and also to the long,
slender petiole, palmately three- or five-nerved from the base.

The numerous leaves seen of this species have all the same characters.
They are small, as broad as long, varying in size from one and a half to four
centimeters long without the petiole, which is very slender, and as long as
the lamina, if not longer. Broadly cuneate to the obtuse point, and equally
so to the petiole, they are more or h'ss enlarged in the middle; one of the
leaves of fig. 9 being four centimeters broad and less than three and a half

180 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

centimeters long. The surface is not rugose or crumpled, but rather
smootli. T!i(! borders arc perfectly entire, not even undulate, and the
consistence is coriaceous, somewhat less so than in the former species,
which it resembles by the shape of the leaves and by the nervation. The
primary nerves, however, are ranch thinner, the nervilles closer, strong, the
upper ones passing to secondary nerves, or altogether taking their places, as
in the fragment of the left side of fig. 9. In other leaves, however, the
distribution of the secondary veins is the same as in P. a?xtica,oi' which
this new species seems to be like a diminutive form. It has also a great
similarity to Paliurus columbi, Heer, whose leaves are -found both at Carbon
and Creston, mixed with those of this Fopulus, and undistinguishable when
the petiole is destroyed. This remarkaljle likeness is seen in comparing for
example fig. 10, whose petiole is shorter and thicker than in the other leaves,
with figs. 14 and 15 of our pi. 1. The identity of habitat and the similarity
of characters in these leaves has rendered their separation difficult, and for
some of them uncertain.

The relation of the.se two last species to Populus has been controverted,
for the reason that no point of comparison is found at our time among living
species of Poplars. The shape and nervation of the leaves have some like-
ness to those of Cercis, these of P. decipiens resembling for example C. antiqua,
Sap. (fit, i, p. 134, pi. xiv, fig. 4 a). This attribution is, however, contradicted
by the long petiole of both the American Miocene species and by a marked
diflerence in the details of the nervation.

Habitat. — Creston, Washakie group {Dr. F. V. Hayden) ; Carbon, Wy-
oming, shale above the main coal, as common there as P. arctica.

Populus mouodon, Lesqz.

Plate XXIV, Figs. 1,2.

Populus monodon, Lesqx., Trans. Am. Phil. Soc, vol. xiii, p. 413, pi. xv, figs. 1,2 ; Aunual Report. 1871,
Supplement, p. 13 ; 1073, p. 375.— Schp., Pal. V^g6t., ii, p. 699.

Leaves large, corioceous, entire or undulate, broadly ov.ite, I.anceol.ate or taper-pointed, rounded
to the base ; primary nerves b.asilar.

The two first leaves of this species described from the Mississippi have
the borders undulate, one of them being marked by a single obtuse short tooth.
This difference, the only one remarked between them and those figured here, is
• of no specific value. These leaves are large, from eight to sixteen centimeters
long, and from six to twelve centimeters broad toward the base, those of the
Mississippi being still larger. The very thick midrib, the slender secondary

DESCRIPTION OF SPECIES— PLAT ANEJi:. 181

nerves, numerous, equidistant, parallel, on the saino open angle of divergence
of 60°, obliquely crossed by very strong nervillcs, are common characters
to all the specimens of this species. From the position of the lower lateral
nerves in our fig. 1, tiiey appear opposite from the base of the leaf, and
show a tripalmate nervation. In fig. 2, the lower nerves are alternate, a dis-
tribution which, though rare in leaves of Popvlus, is seen, as observed
above, in the living P. balsam if era, var. angustifolia. Therefore, as the con-
sistence of this leaf and its shape are the same as in tliat of fig. 1, I consider
it as representing the same species. This Popuius is very closely allied to
P. Gaudini (F. 0.), Heer (Fl. Tert. Helv., pi. Ixiv), by the form, the size, and
the borders of the leaves entire or undulate. The nervation is also of the
same type, rendered still more analogous by the absence of one of the pri-
mary nerves in Heer's fig. 6, as it is in fig. 2 of our plate. The European P.
Gaudini differs by the midrib being nearly half narrower, and the leaves
abruptly narrowed into a long acumen.

Habitat. — Raton Mountains, base of Fischer Peak, New Mexico (/>?-.
F. V. Harden). I found the small leaf of fig. 2 at the same locality.

PLATAN EJ:.

PLATANUS, Tour.

Four species only of this genus are known at our epoch. Platanus
occidentalism Linn., the Plane-tree, or Buttonwood, as it is generally called, is
common in this country, and one of the largest trees of the North American
continent. Its i)abitat, by predilection, is along the rivers of the rich bottom-
laud fisrtilized l)y iuundations, where it attains such a size that its trunk
measures sometime four feet in diameter, even more, its branches spreading
wide around in a kind of wild, irregular fashion peculiar to it, denoting free-
dom oi' movement, and strength rather than elegance. Like the pioneer of the
western wilderness, it seems uncouth and coarse. Its white bark hangs in
patches along its l>ranches, like the shreds of an old tattered garment, but its
limbs are sound and healthy, always covered in summer with a profusion of
large leaves. Its trunk is often hollow, but the tree is tenacious of life. It
defies the attacks of the wild elements, the devastating hurricane, the cold
storms of the winter, Ihc tropical heat of the hottest summer days, protecting
against the multiple changes of t)ur capricious climate the world of vegetables

182 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

sheltered under its branches. In the wide plains liarren of trees, it is seen
looming far away as a fringe to a distant horizon, inviting the tired and thirsty
caravan of the western prairies to a place of rest, where it finds abundance
of fuel and water.

All the species of Platanus are easily recognized by their leaves, gener-
ally of large size, somewhat thick, even coriaceous, especially in a fully ripe
state, palmately lobed and three- or five-nerved from above the base of the
lamina. These leaves have a strongly marked and a mixed nervation, the
primary nerves reaching the point of the lobes, while their divisions, as also
the secondary nerves, either end into the points of the teeth or curve along
the borders as camptodrome.

Of the four living species, one, P. orientalis, Linn., is indigenous in Asia
Minor, whence it has passed to Europe, and has there become a favorite as
an ornamental tree. If more elegant in the distribution of its branches than
P. OCX iden talis, it is also generally of much smaller size. Two other species,
P. Imdeniana, Mart., and P. Mexicana, Moric, thrive in the valleys of Mexico;
the other, P. racemosa, Nutt., belongs to California.

In the Cretaceous of the Dakota group, we find already four well char-
acterized species of this genus, one of which is by its leaves remarkably
similar to P. aceroides of tiie Miocene, the ancestor of P. occidentalis. Besides
these, three other forms have been ascribed to the same genus with less posi-
tive evidence. The Cretaceous formations of Europe have, to the present
time, no representatives of Platanus. Neither in the Cretaceous floras of
Greenland, of Quedlinburg, and of Moletin, by Heer, nor in that of Nie-
dershoena, by d'Ettingshausen, do we find any vegetable remains ascribed
to the genus. Nor is it mentioned, to my knowledge, in the manuscript
notes obtained from Devey and d'Ettin|shausen on the dicotyledonous plants
of the Cretaceous of Belgium. We find the same difference in passing up to
the Eocene formations. In the Lower Lignitic of the Rocky Mountains,
this disputed ground where Eocene evidence afforded by vegetable remains
is contested by animal paleontology, which points to the Cretaceous, four spe-
cies oi Platanus are recognized, one of which, P. Haydenii, Newby., is closely
allied to our living P. occidentals. The Eocene of Europe has none; at
least, no species of this kind is described from the Lower Eocene of Gelinden
and of Sdzanne, and I do not find any mentioned in the list of the species
recognized at Mount Bolca W(; have to go up to the Upper Miocene of

DESCRIPTION OF SPECIES— PLATANE^. 183

Oeningen, Lobsau, Aix, etc., to find in Europe the first remains of Platanus, —
P. areroides, Heer, iiiul P. Guillelmce, Gocpp., two species so much alike that
they have for a long time been considered as one. In the Upper Lignitic;
of the Rocky Mountains, we iiave the same two .species in Alaska and at
Carbon; and si ill higher, in the Pliocene of California, there are two more,
P.appcndicvUita, Lesqx., which, like P. lindeniana, has bifid deciduous stipules,
and P. dissccta, Lesqx., whose leaves, sometimes three-lobate and less deeply
dentate, have a relation to P. niccmom. P. aceroides being recognized as
ancestor of P. occidental/^!, we find thus in the geological times forms inti-
mately related to those of (he [ircsent llora of this continent, and therefore
a clear historical record of the genus.

Plata II us Guillelmce, Goepp
Pl.ate XXV, Figs. 1,2,3.

Platamia Guillelma, Ooepp., Foss. Fl. v. Schoss., p. 21, pi. xi, figs. 1,2.— Heer, Fl. Foes. Arct., ii, p. 473,
pi. xlvii, xlviii, xlix, figs. 4 i.c, d.— Lesqx., Annual Report, 1871, pp. 289,290; Supplement, p.
9 ; Annual Keport, 1872, p. :)H7.'

Platanu^ CEninghauniana, Goepp., Joe. cit., pi. x, fig. 4.

riatamis aceroides, Heer, Fl. Tert. Helv., ii, p. 71, pi. Ixxxviii, figs. 13, 14 ; Fl. Toss. Arct,, i, pi. xii.

Leaves membranaceous, subtrilobate, with dentate or undulate borders, subtruucate or rapidly
narrowed to a short petiole.

This form has been a long time considered by Heer as a mere variety
of P. accroide^s, Goepp., and is still admitted as such by d'Ettingshausen. Our
leaves, as figured in pi. xxv, do not show a clear distinction of the characters
which ought to separate the species. As far as our specimens indicate it, it
is scarcely possible to admit that they represent two specific forms. Figs. 4
and 5, which I think referable to P. aceroides, have the shar[) and large teeth
of this species; but fig. 4 has the leaves more distinctly cuneate or narrowed
to the petiole, and fig. 5 has not ai>y lobes, and these two characters refer
them to P. GuiUelnite. In the large number of specimens obtained of this
type at Carbon, some fragments have still longer, more acute lobes and teeth
than these two leaves, and therefore are more positively referable to P.
aceroides. In the three leaves which represent P. GuiUeimcE, fig. 3 has the
teeth scarcely marked, indeed, like mere undulations; in fig. 2, they are shorter
than in fig. 4, hut already turned upward, and a slight increase of size and
sharpness of teeth and lobes does not seem to be of account for a specific

* The specimens referred to this species from Placifere,7?ew Mexico, are too obscure for precise
•letermination. That of Black Butt.es, described in Eeport, 1872, is referable to Viburnum platanoides.

184 UNITED STATES GEOLOGICAL SURVEY— TEETIAEY FLORA.

character. Thougli it may be, I have here separated the leaves according to
the descriptions and figures of the European authors, without positively rec-
ognizing this distinction as legitinriate; for P. accroides and P. Guillelmce are
both represented by specimens from the same localities. From Goeppert's
iigures {loc. cit.), the leaves are all much smaller than those of our plate, espe-
cially than fig. 3. In Heer's, however, fig. I of pi. xlvii {loc. cit.) is about
of the same size as ours.

Habitat. — Carbon, Wyoming, where it is the most abundant; Washakie
Station, Wyoming {Dr. F. V. Hat/den).

Plata 11 us aceroides, Goepp.

PlatcXXV, Figs. 4,5,6.

Platanus aceroidct, Goepp., Fobs. F1. v. Scboss., p. 21, pi. ix, figs. 1-3.— Heer, Fl. Ter. Helv., ii, p. 71, pi.

Ixxxvii, Ixxxviii, figs. 5-12, 15 ; FI. Foss. Arct., i, p. Ill, pi. xlvii, fig. 3; p. 138, pi. xxi, fig.

17 fc; xxiii, figs. 2 6, 4; p. 150, pl.xxvi, fig. 5 ; p. 159, pi. xxxii, figs. 1, 2.— G.aud. & Strozzi,

Feuilles Foss., p. 35, pi. v, figs. 4-C, vi, figs. 1-3.— Lesqx., Annual Report, 1869, p. 196 ; 1871,

p. 290 ; Supplenjent, p. 11 ; AuuurI Report, 1^72, pp. 389, 406.
PlatnnuB rtigosn, Ooepp., loc. dt., p. 20, pi. xi, figs. 3, 4.
Platanus cunei/olia, Goepp., loc. cit., p. 22, pi. xii, fig. 2.
Plataims (Eninghausiana, Goepp., loc. cit., p. 20, pi. x, figs. 1-3.
Platanus Ettingshauseni, Mass., ex p. Synops., p. 49 (pi. cit., xvii, fig. 3, xix, fig. 3).
Cissua platanifolia, Ett., Foss. Fl. v. Vien., p. 20, pi. iv, fig. 1.
Qucrcaa platanoides, Goepp., loc. cit., pi. vii, figs. 5,6.
(Juerctis rotundata, Goepp., loc. cit., pi. viii, fig. 9.

Le.ave8 piilmately trilobate, truncate or round-cordate to the petiole ; borders deeply acutely den-
tate, with unequal teeth turned upward.

As said in tlie description of the former species, this Platanus sliows
a great variety in the characters of its leaves. It is the same in our living
species, which, on the same tree, bear leaves from three to thirty centime-
ters broad between the hit(^-al lobes, and from four to twenty centimeters
long. The length of the petiole is equally variable, from one and a half
centimeters to eight; two leaves of the same size, and close to each other,
upon the same branch, having the petiole, one five centimeters long, and tiie
other nine. Most of the leaves taken from grown-up trees are three-, more
generally five-lobed, with acute divisions, the teeth of the borders being also
very acute, generally turned upward, and with the base truncate or broadly
cordate. It is only upon the young shoots growing in thickets in the gravelly
beds of the rivers that we see leaves scarcely lobate, or not at all, merely
with short, irregularly dentate borders. These represent in their outline and
general characters the fossil leaves described as P. Guillelnue, the others
those oi P . aceroides. Both species may be therefore con.sidercd, in an equal

DESCIilPTION OF SPECIES— PLATANEJ5. 185

degree of evidence, as the ancestors of P. occidenlalis. Small round leaves,
without distinct lobes, like our fig. 5, are rarely seen in tlie living species.
It is the form represented by Heer in Fl. Tert. Ilelv., pi. Ix.xxviii, fig. 10, as
P. aceroides. It is also very rare to find in the living state as large leaves as
that of our fig. 3, with nearly entire or scarcely dentate borders, all the
leaves of P. occldentaUs resembling this fossil form by their shape, having the
base narrowed, wedge-form, and the borders distinctly and sharply dentate.
But we have a similar l()rm in Heer (Fl. Foss. Arct., pi. xlvii, fig. 1), referred
by the author to P. GuillelmcB.

Fig. 6 of our plate represents a separate stipule of a different species.
As it is nearly entire or obtusely dentate, it l)elongs probaWy to P.Haijdenii,
Newby., whose leaves are generally very large, either trilobate, with lobes
directed upward and obtusely dentate, or with merely ovate, simply or doubly
dentate leaves, without lobes. Specimens of this species occur in profusion at
Golden, and often both forms are represented ui^on the same block of sandstone.

Habitat. — Same as the former.

Plata 11 us Rayiioldsii, Ncwby.

Plate XXVI, Figs. 4,5; Plate XXVII, Figs. l-A.

Plalaitvs BaynoUsii, Newby., Extinct Fl. of N. Am., p. 69.— Lesqx., Annual Report, 1872, pp. 379, 399.—
Schp., Pal. V^g^t., ii, p. 708.

Vnr. Integrifolin.

Platanus integri/olia, Le8qx.,MSS.

Leaves of large size, suborbiculiir or obscurely triangular in ontliue, more or less rounded and
entire toward the decurrent bise, dentate, serrate or undulate, even entire, subcoriaceous.

The author of this species has had for his description a leaf fully pre-
served, with two short lol)es or points below the more elongated terminal one,
and with borders coarsely doUbly dentate. None of my specimens has the
poin.t preserved ; the general shape only is surmised from the more or less
incomplete fragments, and the denticulation is marked upon all the leaves of
pi. xvii, either in sharp or obtuse, small teeth passing above to mere undula-
tions. Though I have no doubt. that all these leaves represent the same species,
there are some differences, striking enough to warrant the representation of
these leaves of ours, which expose characters not recognized in the specimens
which were in the possession of Dr. Newberry. This difference is especially
in the integrity of the borders of the leaves (pi. xxvi, figs. 4 and 5), a character
which has not been recognized to this time in any species ol' Platan u.s. The
nervation of all the forms is perfectly similar. In pi. xxvii, fig. 2, the leaf,
dentate at or near the base, is merely undulate in its upper part, and, from ihe
direction and thinning of the primary nerves, it is evidently not lobatc, but

186 U^■ITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

merely ronnd(Ml or pointed; fig. 4 of pi xxvi lias the same form, and the
bordens are only .-sliglilly undulate, while in fig. 5 of the same plate the borders
are perfectly entire. Hence, with these distinct modifications of characters
exposed to view, it would not be advisable to consider these leaves under
different specific names, as I did formerly, before I had opportunity to recog-
nize the variations upon a large number of specimens. The size of the leaves
is, like the length of the petiole, as variable as in the other congeners.

Habitat. — Golden, Colorado, in connection with P. Hnydenu; the spe-
cimen of 1)1. xxvii, fig. 3, is from the same locality, by Rev. A. Lakes; the
specimen of pi. Ixxvi, fig. 4, is from Black Buttes, Wyoming.

Platan us rlioniboidca, Lesqx.
Plato XXVI, Figs. 6, 7.
Platanus rltomboidea, Lesqx., Annual Report, 1873, p. 400.

Leaves membranaceous or subcoriaceous, enlarged upward from a narrowed cuneiform base,
obscurely lobed above the middle, entire tovi-ard tbe base, deeply sharply dentate in tbe upper p.-irt ;
lateral nerves in an acute angle of divergence, parallel.

The substance of these leaves is of the same consistence as in the former
species, either membranaceous or subcoriaceous; the shape is rhomboidal in
outline, largest at the middle, hence narrowed and entire to the base, broadly
lanceolate or subtruncate to the point, and there deeply dentate; the lateral
teeth entered by the lateral nerves being a little longer or passing to short
acute lobes. From the two only specimens in my possession, tbe leaves
of this species appear comparatively small, from seven to twelve centimeters
long and from five to nine broad. The nervation is Platanoidal, but the wedge-
shaped base of the leaves does not agree in character with that of Platanus
leaves, at least in a general point of comparison ; for, as I have remarked
already, P. occidcntalis has in some peculiar habitat all its leaves narrowed
to the petiole, but dentate to the base, and even a variety of P. orientalis,
described as P. c«weate, Willd., has them cuneate, and often entire downward,
in the same manner as our fossil species. The fossil leaves, however, are of
a more coriaceous substance.

Habitat. — Golden, Colorado (CajH. E. Berthoud, Rev. A. Lakes).

BALSAMIFLU^.

LiaUIDAMBAR, Linn.

The genus is represented in tlie flora of our e[)och by five species. One
of them, Liquidumbar styraciJLuum, Linn., the North American Sweet Gum,

DESCRIPTION OF SPECIES— DLMACE^. 187

is a tiee of moderate size, with tivc-palmalo, serrulate, dark green leaves,
emitting, when bruised, a pleasant fragrance by the exuding of a sweet-scented
glim. The species has a wide range of distribution, being most frequent in
the southern district of our llora, as marked in Gray's Statistics, even passing
above its northern limits and descending to South Florida and Mexico. It
has, like Platanus, a close relation to an Oriental congener, Liquidamhar ori-
entale, Mill., indigenous of Asia Minor; two other species, with pennincrvate
leaves, not l()l)ate, inliabit the East Indian region, Java, and C'liina. Another,
with tripalmately divided leaves, has been more recently discovered in Japan.
To the present time, no leaves of Liquidmnhar have been recognized in
the specimens from the Lignitic of the Rocky Mountains. The genus is,
however, represented in the Miocene Flora of Alaska by Heer, p. 25, pi. ii,
fio-. 7 and in a more recent formation, that of the Chalk Bluifs of California,
it has numerous leaves of a species closely allied to the living L. styrncijluum.
I have described as referable to Liquh/amhar some leaves from the Creta-
ceous deposits of the Dakota group. As they have the borders entire, they
typically ditTcr from tlie genus, as far at least as it is represented at our time,
and, therefore, this reference is doubtful. Though it may ])c of the origin
o( Liqui/famhar, its presence is positively traced on this continent as tlir back
as the Miocene. -Europe has until now two fossil species with serrate leaves
from the same formation, and a tiiird, L. Gaepperti, Walt., from the Paleocene,
whose leaves have the borders entire, bearing to the norinal form the same
relation as L. integrifoUum, Lesqx., of the Dakota group.

URTlCINEiE.

U L M A C E M.
ULMUS, Linn.

The Elm leaves are short petioled, ovate-acuminate or broadly lanceolate,
pointed, doubly acutely dentate or s(!rrate, with a more or less inequikitcral
base, and a pinnate nervation, of clo.se, deeply marked, secondary veins,
ascending at first straight toward the borders, and then curving up, in enter-
ing the teeth, as craspedodrome. The species of this genus are at our present
time about equally distributed in the northern hemisphere, nine of its eighteen
species being Asiatic (four of them in China), four European, and six Ameri-
can. Except U. Mexicana, which Liebman found in the western declivities of
the Cordilleras, all the American species inhabit the northeastern slope of

188 UNITED STATES GEOLOGICAL SURVEY— TEKTIAEY FLORA.

lliis continent; none has been found until now in California and along the
North Pacific coast. Their range is especially in the whole area between the
great lakes and the gulf shores, except for U. cra.ssifolia, Nutt., which is not
known north of Arkansas, and V.Jhndana, Chap., limited to Florida.

Considering its geological records, the genus seems of recent origin in
this coiintiy. The only species described here is from the Upper Miocene
of South Park. Three others are known from the Chalk Bluffs or Pliocene
of California; but in the Lower Lignitic, even in that of Carbon, no leaves
of Ulmus have been found until now. One species, however, is described
by Heer, from the Miocene of Alaska ( U. plurinervia, Ung), where it is repre-
sented by a single leaf, and another has been found in Oregon. None is
recorded from Greenland. Per contra, in Europe, the genus has a number
of representatives already in the Lower Eocene; three are described by
Saporta, from Suzanne, and twenty-two other species are recorded in Schim-
per's Pal. Vdg^t, mostly from the Paleocene and the Lower Miocene forma-
tions. From all these representatives of old, the present distribution of the
genus upon the old continent seems normal. It does not appear to be the
same in North America, for while we find in the Pliocene of California three
species of Elms, none has been left there in it-s present flora, and all the
American species are now, as remarked above, distributed on the eastern
slope. This fact represents only an apparent anomaly; the existence of the
Elms in California at the Pliocene epoch proving a persistence over the
whole continent of some types locally and more recently destroyed by glacial
agency.

Ulmus tcniiinerTis, Lesqx
Plate XXVI, Figs. 1-3.

Vlmus ienuinervis, Lesqx., Annual Report, 187."?, p. 412.

Leaves small, thin, short-pwtioled, either round and equal, or cordate and inequilateral at the base,
lauceolato, gradually acuminate; borders unequally serrate ; lateral veius thin, more or less flexuuus,
and curved in passing up to the borders.

The leaves of this species are comparatively small, averaging six centi-
meters in length and less than three centimeters in width. Their nervation
is thinner, and the direction of the lateral veins less straight or more curved
than in any other spe<Mes of this genus. By the great unequality of some of
its leaves and their doubly serrate borders, it is related to U. Braunii, Heer,
a very common species of the Miocene, whose leaves arc also small. But in
the European form these are comparatively broader, most generally unequal

DESCRIPTION OF SPECIES— ULMACE^E. 189

at the base, with larger, less mixed toctii, iiiid a longer peliole. Exeept hy the
more curved lateral nerves and the simple inieqiial teeth, Ihcse fossil leaves
could represent a variety of Vlmus Americana like the one growing in Texas,
which has small, either equal or inequilateral leaves, round or cordate at the
base.

Haiutat. — Near Middle Park, Florissant, Colorado {Dr. F. V. Haijden).
Castello's Ranch, Colorado (Capt. E. Berfhoud).

PLANERA, Gmel.
This genus is closely allied to the former. Its only living species, indig-
enous in the southern part of the United States, has been found by Michaux
along the borders of the tower Ohio River, where it is very rare, more
generally inhabiting the swampy bottoms of Georgia, Florida, etc. This spe-
cies, Planera aquatica, Gmel, is a small tree resembling Carpinus, bearing
along its branches small, ovate-pointed, unequally serrate leaves, distichous,
and penninerved, like those of Ulmus. The genus is recognized in the
Miocene of Europe in two species, one of which, P. Ungeri, Ett., is very
common and variable. In the geological formations of this continent we have
already three species, one in the Middle Miocene, PL microphylla, Newby.,
the others described below in the upper stages of the same formation. I have,
besides, mentioned from Bellingham Bay, P. duhia, Lesqx. (Am. Journ. Sci.
and Arts, vol. xxvii, p. 361), probably a small form of P. Ungeri, and discov-
ered leaves referable to the living species (P. aquatica) in the chalk banks,
Pliocene, of the Mississippi River, as remarked in the same volume, p. 365.
We may therefore follow the distribution of Planera from the Vancouver
Eocene and the Fort Union Miocene, as indicated by Dr. Newberry, without
interruption, to our time.

Planera. longi folia, Lesqx.

Plate XXVII, Figs. 4-ri.

Planera longifolia, Lesqx., Annual Report, 1872, p. 371 ; 1873, p. 413.— Scbp. Pal. V^g^t., iii, p. 592.

Leaves small, comparatively thick, obloug, lanceolate, obtusely pointed, cuneato to tlie petiole,
simply obtusely dentate; lateral veins thick, simple, craspedodrome.

The species is represented by a large number of specimens, all with the
same characters. The leaves vary in size from two and a half to four and a
half centimeters long without the petiole (five to eight millimeters long) and
from nine to eighteen millimeters broad in the middle. Fig. 5 represents the
broadest of all the leaves seen as yet. They are oblong, obtu.sely pointed,

190 UNITED STATES GEOLOGICAL SURVEY— TERTIAUY FLORA.

the borders marked by large, blunt teeth, vvhose upper marginal line is
horizontal, and thus scarcely turned upward. These characters, as also the
narrower shajje of the leaves, always equilateral at the base, and their thicker
consistence, seem to constantly and positively separate this species from the
following one. There is also a marked difference in the more straight direc-
tion of the secondary nerves at the same, and generally more acute angle of
divergence, 40° to 45°, entering tlie teeth without curving upward, and some-
what thicker.

Habitat. — Elko Station, Nevada {Prof. S. W. Garman). Castello's
Ranch, Colorado {Dr. F. V.Hnyden, Prof. E. D. Cope). Found there in pro-
fusion. Mouth of White River, Utah {Prof W. Denton).

Planer a Ungeri, Zitt.
Plate XXVII, Fig. 7.

Planera Ungeri, Ett., Foss. Fl. v. Vieu., p. 14, pi. ii, figs. .'j-18. — Heer, Fl. Tert. Helv., ii, p. 60, pl.lssx;
Fl. Foss. Arct., p. 110, pi. is, fig. 14 b; Fl. Alask., p. 34, pi. v, tig. 2.— Ett., Foss. Fl. v. Hiir.,
p. 40, pi. X, tigs. 4, 5.— Uog., Foss. Fl. v. Kumi, p. 24, pi. iv, fig,s. 10-1(5. — Sap., £t., ill,
1, p. 72. — Sisni., Mater., p. 48, pi. xviii, figs. 2-4. — Ludw., Palajout., viii, p. 100, pi. xsxviii,
figs. 9-11 ; pi. xxxix. ix, figs. 3-5, fruits.

Planera dubia, Le.sqx., Am. Joiiru. Sci. and Arts, vol. xxvii, p. 361.

Planera lonyifoUaf, Lesqx., Meut. Auuiial Report, 1H73, p. 413.

Zellcora Ungeri, KovatB iu Uiig., Icouog., p. 42, pi. xx, tig. 19. — Massal., Syn. Fl. FosB. Senog., p. 43.

Ulmus zeJkovafoUa, Ung , Chlor. Prolog., pi. xiv, tigs. 7-12.

Uhnus prwionga, Ung., Gen. et Sp., p. 411 ; Iconog., p. 43, pi. xx, flg. 20.

Vlmus parrifolia, Ung., Iconog., pi. xx, figs. 21 22.

Comptonia ulmifolia, Ung., Foss. Fl. v. Sotzka, p. 32, pi. viii, figs. 4, 5,

Fagus ailantica, Ung., Chlor. Protog., p. 105, pi. xxviii, fig. 2.

Quercus subrubnr, Goepp., Tert. Fl. v. Scboss., pi. vii, figs. 8, 9.

Quercus aemi-elliplica, Goepp., loc. cil., pi. vi, fig. 4.

Quercus orcadum, Web., Palieont., ii, p. 172, pi. xvili, fig. 13.

Castaiicc atavia, Goepp., loc. cit., p. 18, pi. v, figs. 12, 13.

Leaves short-petioled, ovate-acuminate, generally unequal at the base, simply dentate or crenate ;
secondary veins simple or branching near the point, curving upward in entering the teeth.

The leaf (fig. 7) was mentioned as referable to the former species from
a sketch communicated by my friend, Capt. Berthoud. Receiving later from
Dr. Hayden a specimen similar to the figure formerly sent, I had opportunity
to compare it to those of P. longifolia, and to positively recognize the marked
difference in the acute point of the teeth, the nervation, and the general form.
I have also examined, as points of comparison, the figures of all the works
quoted by Schimper in the above nomenclature, and in none of these do I
find any leaf comparable in the characters to those of P. longifolia. We have
therefore two species, P. Ungeri, described already from the Miocene of
Alaska by Ileer, and the former, exclusively referable until now to the Upper
Miocene of Elko and the Parks.

DESCEIPTEON OP SPECIES— OELTIDBJC. 191

Habitat. — Castello's Ranch, Colorado [Crqit. Bcrthoud, Dr. F. V. IIcoj-
den). Very rare.

CELTIDE J5.

CELTIS, Tournf.

Considering its American representatives, this genus is distinctly charac-
terized by petioled leaves, short, truncate or subcordate at the more or less
inequilateral base, ovate or oblong-lanceolate, acuminate, with serrate, crenate,
or entire borders, triple-nerved. The two lateral basilar nerves ascending
to the middle of the leaves or above, are camptodrome, like their division
and the secondary nerves also.

More than seventy species of this genus arc described in De Candolle's
Prodromus as distributed over the whole world The number, however, is
exaggerated; for the forms are disposed to vary greatly by culture and to be
modified by various atmospheric changes. For example, twelve species are
credited to North America, and these, from the opinions of botanists of the
present time, who have had opportunity to compare them and to follow their
variations, are reduced to two species, one, C occidentalis, which ranges in
longitude from the Atlantic to the base of the Rocky Mountains, and in
latitude from New England to Florida, and the other, C. pallida, Torr., which
inhabits Texas and Mexico. The first of these species is locally very abun-
dant.

The geological records of Celtls are for this continent as yet obscure.
The flora of the Mississippi Eocene has two fragments of leaves, described
as Celtis brevifolia, Lesqx. (Trans. Am. Philos.'Soc, vol. xiii, p. 41 (i, pi. xx, figs. 4
and 5). The lower part of these leaves and their nervation sJiow the characters
of Celtis; but the upper part is destroyed, and thus the generic relation is not
positively ascertained. In Europe, which has one living species only in its
flora, paleontology has recognized seven species of Celtis, all Miocene.

M R E ^ .

FICUS, Tournf.

The leaves of species of Ficus are so variable in size, shape, consistence,
nervation, etc., that it is not possible to expose any characters by which their
reference to the genus may be positively ascertained. Schimper remarks on
this subject (Pal. Vegdt., ii, p. 728): — "That most of our determinations of

192 UNITED STATES GEOLOGICAL SURVEY— TERTIARY FLORA.

Ficus leaves merely rest upon more or less evident probabilities, as the largest
number of the fossil leaves referred to this genus could be ascribed to others
bv the same reasons which induced the authors to admit them as Ficus.
And it is probable that the number of nearly one hundred fossil species now
described will be considerably reduced when they have been thoroughly
studied." In this memoir, the reference of the tertiary leaves to Ficus is
essentially based upon their identity, or close relation of characters, to spe-
cies described by European authors, who have access for their comparison
to large collections from the whole world, while I have at my disposal
only specimens from Cuba, Florida, and South America, which, though ref-
erable to numerous species, do not represent, by far, all the types of leaves
pertaining to this genus.

The species of Ficus are extensively distributed between the tropics,
in the humid and warm regions of the equator especially. Many are found
in the West Indian Islands, in Jamaica and Cuba; three are indigenous in
Florida. A single one inhabits the southern regions of Europe.

As far as evidence can be credited, the origin of the genus is Creta-
ceous both in Europe and America. Prof Ileer has two species from
Moletin, one from Greenland, and one from the Dakota group. Three others
are described from this same formation in Dr. F. V. Hayden's Annual Re-
port, 1874, pp. 341 and 342, and four from Niedershoena by d'Ettingshausen.
From the Eocene floras of both continents, numerous leaves are referred to
Ficus, and the number greatly increases in Europe with the Miocene forma-
tion. In the Lignitic of the Rocky Mountains, Ficus leaves are most abundant
in connection with remains of Palms in the Lower Eocene of Black Buttes,
Golden, and the Raton Mountains; they are thus typical evidence of the
temperature of the epoch.

§ I. — Penninerved leaves.

Ficus laiiccolata, Heer.

Plate XXVIII, Figs. 1-5.

Finu lanceoJata, Heer, Fl. Tert. Holv., ii, p. 62, pi. Ixxxi, figs. 2-5; iii, p. 1S2, pi. cli, figs. 34, 35; clii,

fig. 13.— Mass., Fl. Seiiog., p. 223, pi. xxx, fig. 8.— Ett., Fl. Fofs. v. Bil., p. 07, pi. xx, figs.

3, 4.— Heer, Mioc. IJalt. Fl., p. 73, pi. xxii, figs. 1, 2.— Sisni., Mater., pi. xv, fig. 5 ; xxvi,

fig. 2.— Lesqx., Animal Report, 1S71, p. 300; 1873, p. 414.

Leaves coriaceous or subcoriaceons, entire, lanceolate, tapering upward to a long acnnicn, and

narrowed downward to a thick sLort jietiole ; midrib strong ; lateral veins irregular in diHtance, cauip-

todrome.

The leaves of this species are generally larger than those which are fig-

DESCRIPTION OF SPECIES— MOKE.52. 193

tired in our jilate; the essential cliaractcrs, howi'vcr, the lanceolote form, the;
long tapering hase, tlic thick petioh-, and the nervation, arc distinct!}- recog-
nized npon tlie specimens. The subslanre is thick, rather coriaceous; tlie
distance between the lateral nerves variable, and the areolation of the same
type as in fig. 1, pi. xxii, of the Mioc. Bait. Flora, a leaf of the same size as
those of our plate, whose fig. 3 compares also in every point to fig. 13, pi. clii,
of the Fl. Tert. Helvet. Onr fig. f>, however, does not closely agree in its
characters with those of tliis species. The secondary nerves are too close
and equidistant, reaching to near the borders, and following them in a series
of cnrves. 'Its thick midiib refers it to a iVcws, and its narrowed base to this
species. It is from a different locality, and the only fragment which I had
for identification.

Habitat. — The leaves in figs. 1-3 are from the Green Eiver group, Wy-
oming, with Populus arctica and Cyperus Chavanenais (Dr. F. V. Hin/den);
fig. 4 is from a specimen from Willow Creek, Middle Park, Colorado (Dr.
W. H. Holmes); and fig. 5 from Florissant, near South Park, Colorado {Prof.
E. D. Cope).

Ficus Jyiix, Ung.
Plate XXVIII, Fig. G.

i'lc'iis Jtjnx, Uog., Fl. V. Sotzka, p. 16,5, pi. xxxiii, fig. 3.— Ett., Tert. Fl. v. Hiir., p. 41, pi. x, figw.

C-8; Fobs. Fl. v. Bil.,p. 69, pi. xx, figs. 2-7.— Ilcer, Fl. Tert. Helv., ii, p. 63, pi. Ixxxv, figs,

8-11. — Lesqx., Annual Report, 1873, p. 414.
Rhamnus Eridani, Ung., Foss. Fl. v. Sotzka., p. 178, pi. lii, figs. 4-6 (Me Ett.).

Leaves coriaceous, lanceolate, narrowed to the petiole; secondary nerves .it an open angle of
divergence, close, parallel, simple, straight to the borders, where they abruptly curve along them.

The leaves of this species, as represented by Heer and Unger {loc. cit.),
are larger than the fragment figured here, and generally broadly oval and
obtuse. Fig. 7, however, of the Bilin Flora is similar to ours in every point.
The German author refers to this same species three leaves of the Sotzka
Flora (figs. 4-6, pi. lii), which also closely resemble ours, especially fig. 6.
The identity, however, is not absolutely ascertained on account of the frag-
mentary state of our leaf, whose petiole is broken and its upper part destroyed.
The direction of the secondary veins, at the same angle of divergence as in
the European form, and abruptly curving in touching the apparently reflexed
borders, relate this leaf rather to Rhamnus than to Ficus.

Habitat. — Elko Station, Nevada {Prof. E. D. Cope).

13 T F

194 UNITED STATES GEOLOGICAL SUllVEY— TERTIARY FLORA.

Ficus niultiiicrvis, Heer.

Plate XXVIII, rigs. 7, 8.

Ficus mulHnerris, UeeT, Fl. Tert. Helv., ii, !>. G3, pi. Issxi, figs. 6-10; Issxii, fig. 1.— Ett., Fosx. Fl. v.
Bil., p. 63, pi. XX, figs. 5, 6.— Leeqx., Aunual Report, 1871, p. 300.

Leaves coriaceous, very entire, elliptical-laneeolate, acuiuinate, gradually or rapidly narrowed to
the base ; lateral nerves at an open angle of divergence, very close, parallel, curving close to tbe borders ;
areolation very small, quadrangular.

Considering the form of the leaves of this species, we see it represented in
both figs. 7 and 8 of our plate, in coincidence, the first with figs. 7 and 8 of Heer
(loc. cit.), and the second with fig. 6, which represents a leaf of about the same
size, with the base round-truncate. Except that the secondary nerves are
somewhat more oblique to the midrib, generally at least, but not more so than
in fig. 7 {loc. cit.), the nervation has the same characters. The secondary
nerves, very close, running straight to the borders, are separated by interme-
diate tertiary veins, which, though somewhat thinner, are quite discernible
even to near the borders. Joined by nervilles in right angle, divided and
subdivided in the same manner, the result is an ultimate areolation of very
small and quadrangular meshes. This areolation, more distinct upon the
American leaves than upon those described by European authors under this
name, and also the more oblique divergence of the veins, seem, however, to
disagree with the characters of the species, even with those of the genus.
Considering these differences, Count Saporta is disposed to refer these leaves
to Laurus, or perhaps to Nerium, rather than to Ficus.

' HABrrAT. — Green River group, with tlie leaves represented in figs. 1-3
as Ficus lanceolata {Dr. F. V. Hayden).

Ficus oblanceolata, Lesqz

Plate XXVIII, Figs. 9-12.

Fxcua oblanceolata, Lesqx., Annual Report, 1872, p. 387. — Schp., Pal. Y6g6t., iii, p. 595.

Leaves subcoriaceous, entire, obovate or obhmccolate, obtusely pointed or acuminate, cuneate to
the petiole; secondary veins numerous, parallel, camptodrome.

The leaves of this species have some likeness to those of F. lanceolata,
Heer. Their form is comparatively shorter and broader, and their nerva-
tion far more equal, the lateral nerves being close, equidist