i

PEPARTMENT OF THE INTERIOR.

iz.

II E r 11 T

UNITED 8TATE8 GEOLOGICAL SURVEY

THE TEBRITOPvIES

UNITED STATES GE OL G IST-IN-CH AEGE.

VOLUME 11.

WASHINGTON:

G <) V K U N M E N T P R I N T I N O O !•' F 1 CI E ,

1875.

LETTER TO THE SECRETARY.

Office of the United States
Geological Survey of the Territories,

Washington, D. C, October 5, 1875.

Sir: The very vnliiablc niemuir, l)y Prol". K. D. Cope, on the " Verte-
Iirala of the Cretaceous Formations of the West," is respectfully submitted for
your approval and for publication. The great interest which has been excited
among the peojjle, as well as among scientific men, l)y the discoveries of the
remarkable extinct vertebrate remains, within a few years past, in the num-
erous ancient lake-basins of the West, will render the publication very
opportune at this time, and one of great value to the intelligent world. As a
contribution of materials toward the solution of the numerous problems
involved in the geological structure of our great West, as well as the unfold-
ing of its ancient life, it must take the highest rank.

In a certain sense. Paleontology, or the history of ancient organic remains,
lies at the very foundation of geological science. It oftentimes stands as the
arbiter on doubtful questions, and is a very important aid to the stratigrapher
in unfolding the age of strata, and in fixing the great timc-bnnndaries o(
groups, as well as divisions or sul»divisioiis. Xo geological survey can be
considered complete without its cooperation. It has l)ecn with this idea
that so much of the strength of the survey has been given to this department.

So far as the opinions expres-sed by the author in regard to the age of
the Lignitic group are concerned, he aloni' is responsible to the scientific
world. In all the publications of the survey under my charge, prepared by
collaborators of established reputation, I have thought it best to permit the
broadest latitude in the expression of dpininns, w hetlier they liarnionize with
my own conclusions or not. While Professor ( 'oiie reads, in the teachin"-s ot
the extinct vertcbra'.e fauna of tlie Lignili>' irroiip, its Cretaceous age, Pro-

IV

fcssor Lesqucreux, with equal sincerity and force, decides, from tlic lessons
of tlie fossil vegetation, that it belongs to the Tertiary epoch.

The present volume will be followed in due time by two others by the
same author, on the Vertebrata of the Tertiary Lake-Basins of the "West.
I have the honor to be, your obedieut servant,

F. V. HAYDEN,

United States Geologist.
Hon. 1>. R. CowEN,

Acting Secretary oj the Interior.

UNITED STATES GEOLOCICAL SURVEY OF THE TEIMIITORIES.

THE VERTEBRATA

CRETACEOUS FORMATIONS

THE WEST

By E. D. cope.

WASHINGTON:

GOVERNMENT PRINTING OFFICE.
18 75.

CONTENTS.

Letter to Professor Hayden

lutroduction : on the significance of paleontological science

PART I.

On tbc classification and distribution of tlio Cretaceons deposits of the West.

PART II.

Descriptions of the Cretaceons Vertebrata of the West

1. Description of the localities and fauna

2. Literature of the Reptilia

3. Descriptions of the Reptilia

4. Li terature of the Fishes

5. Descriptions of the Fishes

PART III.

Synopsis of tlu- known Cretaceous Vertebrata of North America

Page.

15

42
42
51
53
179
180

245

27889

LIST OF WOOD-CUTS

Kic. 1. Diagram of teeth a{ Cionndon arctatun, page 59.

8. Sketch of bones of I'lolontena gigcs, taken at time of disrovery, i>.i<j;e '.)9.
n. Costal bones of a young Tinliiiio poli/phcmiiK, page ll'i.
4. Ramns of lower jaw of Clidastcs i}rop!ilh«n, page \'irl.
T). Ramus of Loxoccmn.i hivnlor, page Vli.
0. Ramns of Ertji johnii, page Viri.

7. Pelvis anil adjacent vertebnc of Liodoii prui-Kjer, as exposed in the matrix, page 171.

8. Cranium of Porthens molosmts, page 184.

!>. Femoral bones and ventral tin of a Poi-IIuiin, from above; a, from below; (/, from tlie side, page I'.l-J
10. Scapnla, ete., and base of ])eetoraI fin of I'llecoiilrnm diinii-tjKs, Cope, |iage 244 L!.

Philadelphia, January 20, 1875.

SiK : The accompanying pages embrace my final report on the vertebrate
paleontology of the Cretaceous formations of the West. The greater number
of species described has been derived from the beds of the Niobrara (No. 3)
and Fort Union (No. 6) epochs. The material has been obtained from the
explorations in Kansas by the writer in 1871 ; from similar explorations in
Kansas by Prof. B. F. Mudge in the years 1870 and 1S72 ; from the explora-
tions by the writer in Wyoming in connection with the United States Geolo-
gical Survey in 1872; and from a similar expedition in Colorado in 1873. I
wish to express here my indebtedness to various friends who have assisted
me on these occasions ; especially to General John Pope, commanding the
Department of the Smoky Hill, and Captains Butler and Lyman, and Dr.
King, stationed at Fort Wallace at the time of my expedition in 1871 ; to
Dr. John H. Janeway, United States Al-my, of Fort Hays, and Profs. B. F.
Mudge and George Merrill, of Kansas, for invaluable specimens of the fossils
of the Niobrara group ; and to Capt. E. 0. Clift, Dr. Joseph Corson, and Judge
W. E. Carter, of Fort Bridger, Wyoming, for many kindnesses. I am also
under obligations to George M. Dawson, geologist of the British North Amer-
ican Boundary Commission, for the opportunity of examining fossils from the
Fort Union lieds of British America ; and to the Smithsonian Institution for
facilities in the use of specimens and books.

AVhere it has been possible to throw light on questions of stratigraphy,
this subject has also been touched upon.
I am, with much esteem,

E. D. COPE,

Paleontologist.

Dr. F. V. Hayden,

Director of the U. S. Geological Survey of the Territories.

^CT'

I ISr TR O D U C T I O N^

ON THE GENERAL SIGNIFICANCE OF THE SCIENCE OF

PALEONTOLOGY.

I.

Paleontology is an exact science. It embraces generalizations or laws
obtained by indnction, which may be dednctively applied to the unknown.
Its first law is an illustration of the uniformity of nature's methods, namely,
the law of the persistence of type. An organized structure once created,
and existing under circumstances not hostile to its working, is adhered to
with the greatest fidelity, and extended in time and space. This constant law
is the key to this as to tlie other biological sciences, and occasionally surprises
the student of evolutionistic proclivities. On this basis, the possibility of
reconstruction of the extinct forms of the past will always rest ; and the cer-
tainty of the law is unconsciously admitted by every paleontologist who
determines, names, or classifies a fossil from anything less than a perfect
specimen. It is assumed every day, and universally allowed, although occa-
sionally even an expert is found who sometimes questions it, and still more
frequently an inexpert who does not read nature aright.

The application of tlie law is, however, various as the given terms, i. c,
the remains preserved, ditfer in significance. Thus, to illustrate, certain parts
are common to all stoves, and distinguish them from all other article?t of furni-
ture; but certain other parts not only belong to a stove, but mark a given
pattern of stove, since they belong only to it. A still more minute range of
appearances is found only in one man's make of stoves, and others in tiiat of
another man. Hence, a person acquainted with stoves, sewing-machines,
&c., can readily determine the origin of a very small part by referring it to its
projicr kind and make.

The application of this law of persistence presupposes a knowledge of
the pattern as essential to its deductive application. Hence, a difliculty at
once suggests itself as arising when a portion of an aninial belonging to a new

8

pattern is discovered. That patterns quite distinct from those known to zool-
ogists have existed in past ages has been well proven by paleontologists.
How can the structures of a species of such a kind be inferred from a frag-
ment ? Another law equally true with that of persistence has been developed
from the facts, but it is much more difficult of application. This is the one
already defined by the writer^ under the name of the law of " successional
relation.'' It is absolutely certain that the types of nature, whether pri-
mary or subordinate, form series of steps passing from one condition of rela-
tions to another. The natural deduction is, that if a portion of an animal
exhiljits a form intermediate between two known forms or types, the remainder
of the animal's structure possesses the same kind of intermediacy. This
law is tacitly admitted, and employed by paleontologists ; but there is a diffi-
culty of application in consequence of the existence of other laws now to be
considered.

The first difficulty arises from our possible ignorance of one terminus of
the series or hue in which our fossil represents a stage. This objection is
more theoretical than real, because the living classes and orders are the struc-
tural extremes of the lines of succession ; nevertheless, among divisions of
lesser range, many have reached their culmination, and disappeared in times
past. These points of culmination must be known in order to ascertain the
direction of the succession. Every discovery, however, is not that of an
advanced position on such lines ; intermediate positions being necessarily more
numerous than termini. Hence, this difficulty is of only occasional recurrence.

The preceding considerations all express different phases of the law of
uniformity. I now refer to the law of variation, which is in apparent conflict
with it. It is the law which expresses evolution as opposed to persistence of
types. It especially limits the application of the last law, that of uniformity
in succession ; i. c, that when one portion of structure occupies a position
intermediate between two already known types, the remaining parts of the
same animal or system of organs will occupy the same relation of structure
to the corresponding parts of the known. This is not uniformly true. The
law of variation intervenes, which states that it may occur that, while one
part of an organization occupies a relation of intermediacy, the other parts do
not exhiliit exactly the same relation. It is by the unequal mingling of
structural points that new lines of succession are marked out. Thus it is

' Pcuii Monthly Magazine, 1872, p. 229.

that the power of reconstruction from fragments is limited, but not sufficiently
so as to justify the epithet " pretension," which has been applied to the claim
made. Besides, two other laws remain, which are of great importance to the
paleontologist.

Illustrations of tlic preceding laws may first be given. If a fragment of
an animal be found, which contains a certain type of teeth known as the true
selenodont, it is certain, in accordance with the law of uniformity of type,
that the first bone of the hind foot of that animal (the astragalus) possessed
two pulley-grooved fiices, one above and one below, and not one only, as in
most animals; also, that the lower pulley-face was succeeded by two sub-
equal toes, and that the lateral toes were either reduced in form or wanting.
There is no mechanical relation between the structures of the teeth and foot ;
their accordance is simply a fact of type of a selenodont Artiodactyle.' Again,
if I find a portion of a foot which presents a joint between the first and
second rows of bones which ibnn the sole, I am absolutely certain that the
animal had the two outer ear-bones external to the skull, forming a part of
the lower jaw and the connecting-rod by which the latter is attached to the
skull. This is a type-law of the bird and reptile. Again, if I find a part of
a foot of the structure just named, where the first row of bones of the sole
is united into one mass, and closely embraces the leg-bone without being
continuously united, I know that I have an animal with teeth, with a very
long hip-bone and a very long series of united vertebra? (or sacrum) resting
upon it — in other words, a Dinosaurian.

The law of uniformity in successional relation is well illustrated by the
genus Loxolophodon. The first bone of the foot (astragalus) of this animal
exhibits characters intermediate between that of the elephants {Prohoscidia)
and odd-toed hoofed mammals (example, tapir) ; the remainder of the skeleton
does the same ; the neck-vertebrse are similar to those of the former, while
portions of the skull reseml)le corresponding ones of the latter. The toot of
a dinosaur is intermediate between that of a reptile and that of a bird ; so
are the sacrum and pelvis. The sternum of a frog of the family Discoglossidce
is intermediate between those of ordinary frogs and salamanders ; so are the
vertebrae and ribs.

Examples of the limitation of the latter rule are still more numerous.
They may "be produced from the three cases cited. Thus, in the dinosaur, it

' Represented by a Kuiiiiiiiiiil.

10

might once have been said that tlic jaws did not partake of the intermediacy,
because they all present teeth, and are never smooth, like those of birds. Yet
birds with teeth have recently been discovered, which deprives us of the use
of this character as a definition. In the discoglossid frog, the cranium is not
intermediate in structure between the frog and salamander, but is that of a
trog. Ill tiie Loxolophodon, the toothless front of the upper jaw is not a
general character of either of the orders between wliich it stands.

These difficulties arise from the existence of the subordinate variations
or subtypes of a general or major pattern, and, for their resolution, require
only a new application of the first law of uniformity on the lower plane. If
the subcharacters defining the subpattern be known, the existence of one
presupi^oses that of the others. The structure of an artiodactyle astragalus
will not enable me to infer the character of the incisor-teeth of the animal;
for this I require some other more minutely-correlated portion. So I can
infer the ribs and vertebrae from the sternum of the discoglossid frog, but n(jt
the cranium ; for this I require some part correlated with discoglossid char-
acters only, and not only significant of the relations to the orders of Batra-
chians, as are the characters mentioned, although it happens, by the accident
of discovery, that none but such frogs are known to possess them to-day.

The two laws vyliich further aid the deductions of the paleontologist are
those of mechanical relations and of embryonic parallelism. One structure
requires another in order that an animal be viable. Thus, long legs in a grazer
presuppose a long neck to enable it to reach the ground with its lips. Hooked
claws presuppose carnassial teeth or a hooked beak. For a horizontal body
to be properly poised on two legs instead of four, the weight of the viscera
must be transferred backward, and the anterior regions of the body lightened.
This we find to be the case with birds and Dinosauria. The lower bones of
the pelvis, with the contained organs, ai-e thrown backward, while the fore-
limbs are lightened and the head reduced in proportionate (not absolute) size.

The parallelism of types with transient embryonic conditions of other
types, aids the paleontologist essentially in the classification or proper location
of a specimen. Its relation to known series must be first determined, as this
obviously precedes in reconstruction all application of tlie law of uniformity.
Such reference having been made, either to a new series or to a place in a
known series, the considerations heretofore adduced come into view, but not
sooner. Hence, the law of parallelism is as essential to the paleontologist, as
it is all-pervading and all-expressive of nature itself.

11

XL

Paleontology in its relation to geology is as yet partially empirical.
Thus, while its indications are definite for one locality, they have not identical
significance for all localities on the earth's surface. The lower we descend in
the scale of being, the more uniform over great areas are its phenomena ; but,
among higher animals, esjiecially vertebrates, the greater the geographical
peculiarities as compared with the stratigraphical. Professor Agassiz once
said that the e.xisting geographical fauusB are more distinct than the extinct
faunae of two consecutive epochs of geologic time; a statement justified by
many facts. Hence, it has been believed by some, that fossil vertebrates
cannot furnish conclusive evidence of the age of the rock-strata in which they
occur; for, say they, we have to-day existing on the Australian continent,
animals that approach more nearly to those found fossil in the Jurassic forma-
tions of Europe than to any now living on the latter continent ; so that, were
Australia to be presently submerged and her strata and fossils again brought
to light, the paleontologist would assert that the sun had not shone on that
land since the days of the Jura. And so he would were he not at the same
time a zoologist; just as the bare zoologist would err in the opposite direction
of assuming the modern age of the European Jurassic beds, because they
contain the living types of Australia. Thus, a foundation-fact of zoology
properly applied is essential to the paleontologist, namely, that the earth
now presents four or more distinct faunal areas, the more prominent among
which are the Australian, the South American, and the temperate lands of the
northern hemisphere. Each of these possesses many pecuHar forms of life
not now found elsewhere. Has this distinction always prevailed? Paleon-
tology answers decidedly in the affirmative, so far as extinct mammalia are
concerned. There seems to be no doubt that the faunal distinctions have a
very ancient origin, and are therefore to be first considered when estimating
the age of strata from the contained mammalian remains. The explanation
of this diversity is not yet attainable ; but an important advance has been made
by the discovery of the great similarity between the extinct forms of the
northern hemisphere and the living or more modern ones of the southern
hemisphere faunae. The Jurassic character of much of the Australian fauna
is known, while prevalent t3'pes of South America and Africa can be shown
to have much relation to Eocene types of the north. In North America and
Europe, tapirs, opossums, coatis, civets, kinkajous, lemurs, and allies of the

12

toxodonts belong to the Eocene ; now these animals characterize the southern
continental life, or, as is the case with toxodonts, have but recently become
extinct there, ^fhis mode of defining those faunae is not, however, exact,
since many modern types have found their way into them, especially in the case
of Africa.

How, then, is life significant of chronological station in the earth's strata?
Since very many forms of animals are so widely spread and at the same time
so distinctly limited in range on the earth's surface to-day, the same order
must have prevailed in past time, and have been of equal significance. That
this law of uniformity has prevailed in the past as in the present is amply
proven by the paleontology of a single zoological area taken by itself. The
apparition of types over the northern land-area has been nearly universal.
This fact has only been placed within our reach by modern investigations in
North America; for, until the sister continent of Europe-Asia was explored,
no one could be sure what degree of individual peculiarity her extinct life
might present. Now it is certain that the succession of Tertiary beds was
mutually similar, and that the contemporaneous deposits contained in a large
degree similar life, and that intermediate stages of the one can be properly
intercalated in the vacant interspaces of the other. The resemblances be-
tween the Lower Eocenes of New Mexico and Wyoming and that of France
are marked ; similarity between the Pliocenes of the respective continents is
evident. Descending in the scale, the parallels between the North American
and New Zealand Cretaceous are very apparent, and the faunae of the Caro-
linian and Wiirtembergian Trias were the same. The interruptions in the
record of life marked by the appearance of great land-areas near the close
of the Carboniferous and Cretaceous periods are universally observed in the
zoological areas of the northern hemisphere, or Arctogsea. The close of the
Cretaceous everywhere saw the end of ammonites, rudistes, and sauroptery-
gian and dinosaurian reptiles, in spite, in North America at least, of physical
continuity of deposits.

Was this succession of interruptions of life universal over the globe, and
do these trenchant lines justify the old assumption of repeated destructions
and rcTcreations of animal life'? The former question has already been
answered in the negative by the explanation of the characters of the existing
faunae of the southern hemisphere, where ancient types sfill remain in con-
siderable numbers. Moreover, some of (he later periods of both N(uth

13

America and Europe are characterized l)y a large predominance of forms of
the corresponding southern continent. It is, indeed, evident tiial inigrafioii
from tiie one continent lo the otlier has taken place, and is amply sufficient
to account for the abrupt changes in the life of each, without necessitating
the intervention of creative acts. If glacial periods be dependent on cosmic
movements, the increased obliquity of the earth's axis to the sun, at periods
25,000 years apart, due to the same causes as precession, would cause a coi--
responding alternation of cold periods in the opposite hemispheres. This is
well known as a most potent cause of migration and extinction, and the
known relations of the faunae would thus result from a greater or less alter-
nate invasion of the one hemisphere by the life of the other.

But within the great time-boundaries are distinct land-fonnjc whose
striking distinction may not thus be accounted for. Tims, the Miocene and
Loup Fork faunae of Western America are entirely distinct, but with corre-
sponding members. The alternate presence and absence of water-areas
adapted for the preservation of the remains of the animals will abundantly
account for such minor interruptions. Such changing topogra})hy is well
known as due to the slow vertical oscillations of the earth's crust.

The original question, the exactitude of the chronological significance of
structural types, has been momentarily held in abeyance. Is paleontology a
science so far exact as to furnish a chronological scale of terrestrial strata?
The admission that the known Tertiary faunae, for instance, are but fragments
of a continuous succession, would appear to invalidate any such claim. It
would indicate tliat tlie restriction of a given type to a given horizon is only
a matter of discovery, and that another accident may at any time give it a new
range. This objection has but little weight. Fragments though i:hcy be,
nearly-related formations, as the Tertiaries, are obviously the visible portions
of a serial succession of life. Like the bright lines in a spectrum, the order
is not disturbed by the temporary obliteration of a part of the colors, but the
visible portions indicate the relations of the component parts with ir.foUible
certainty. The more universal the physical interruption, the more far-reach-
ing tlie break in tlie succession of life in any one locality, the greater
is the significance of remains of animals as indication of relation in time. Tin;
change of faunae in Arctogaea at the close of the Cretaceous is a case in point.
A dinosaur, sauropterygian, ammonite, or rudist are as definite indicators of

14

the life that precederl the change as a tapir or civet-like carnivore is of the age
that foIlowe<l.

It has been stated that the life of the present period in the southern
hemisphere is not homogeneous. The same is true though in a less degree
of the northern. Thus, if we include India in the latter, the elephant is a
Pliocene form, and the true rhinoceros Upper Miocene. In the northern
hemisphere, the dogs are Miocene. In North America, the opossum, and
probably the raccoon, are Eocene; the wolves and foxes ai-e Miocene, and the
weasels Pliocene. Perhaps, the cats first appeared in our Pliocene. Com-
paratively few mammalian types mark, by their origin, the latest geologic
epochs. Siicli are the ruminants, as deer, and oxen, with the true horses,
which all commence in the Upper Pliocene of Arctogaea. Finally, man alone
signalizes the last or glacial period, and is to reacli his culmination in the
ages that intervene between that great time-boundary and one to come.

Thus, a certain proportion only of the life of a given epoch is character-
istic of it, that is, originates in it; the remaining members being legacies from
preceding ages. Hence, the latest forms of life embraced in an extinct fauna
are the true indicators of the chronological relations of that fauna.

P A K T I.

THE CLASSIFICATION AND DISTRIBUTION OF THE CRETA-
CEOUS DEPOSITS.

Messrs. Meek and Haydeu have classified the vast thickness of the Cre-
taceous formations of the West from observation of the section made by the
Missouri River from near its sources to the point where it enters the Carbon-
iferous strata of Eastern Kansas. This classification has been found by Dr.
Hayden to be applicable to the exposures of the rocks of this period along
the eastern base of the Rocky Mountains ; and I may add that an examina-
tion of the western flank of the same mountain-region in New Mexico, con-
ducted by myself, under direction of Lieut. Gr. M. Wheeler, United States
Engineers/ has extended the application of the same system.

The classification of Messrs. Meek and Hayden embraces five distinct
ejjochs, all of marine deposits. The lowest is a sandstone, which rests uncon-
tbrmably on Azoic, Carboniferous, Jurassic, or other beds, as the case may
be : it is succeeded by a series of usually dark-colored shales or clays, or
No. 2, which is overlaid by a gray, white, or yellow chalk or calcareous marl,
which forms the surface-rock of^a large area; this is No. 3. The superincum-
bent divisions, Nos. 4 and 5, consist of laminated shaly clays and sandy beds
respectively. Above this point, the character of the deposits becomes brackish
and lacustrine, constituting an approximation to the overlying Tertiary forma-
tions. This physical change has been regarded by the stratigraphers as tlie
boundary-line between the Cretaceous and Tertiary periods; and (he indica-
tions derived from the vegetable fossils are not inconsistent with such a view.
The vertebrate paleontology, on the other hand, shows the interruption in that
kind of animal life to have taken place much higher up in the series; hence,
in the present work, the Lignite, or Fort Union, epoch is included in the
Cretaceous formations as No. 6.

'See Appendix F F 3 of Anuual Kepoit of the Chief of Engineers, Wasbington, 1874.

16

I. — THE DAKOTA EPOCH.

The sandstones of this division are light-brown, buff, or white, moder-
ately soft or very hard, and varied with occasional conglomerates. They are
extensively developed on the Missouri, reaching 1,500 to 2,000 feet in depth.
They appear all along the eastern flank of the Rocky Mountains, and on the
west side of the Sierra Madre or San Juan. No vertebrate fossils have yet
been obtained from them.

II. — XnE BENTON EPOCH.

This formation embraces dark, lead-colored shales and clays, and is found
lying on the preceding in most regions where it occurs. It includes many
mollusks, some of which are identical with those of No. 3. Among these are
Ostrea covgcsta and Inocerair.us problematicus. Four vertebrate species have
been defined from the numerous remains discovered, viz : Lamna ? cuspidata,
a shark ; Pelecorapis varius, an ally of the flying-fish ; Ai)sopeUx sauriformis,
a related fish ; and Hijposaurus vehhii, a gavial-Iike crocodile.

III. — THE NIOBRARA EPOCH.

The exposures of this formation have a wide area between the Carbonif-
erous and Lower Cretaceous beds of the east of the plains and the Rocky
Mountains, and in Texas and Eastern New Mexico. Vertebrate fossils are
very abundant, and I have received them from Kansas, Colorado, Texas,^ and
New Mex^ico. Tlie following description is from the notes of my Kansas
expedition of 1871 :

" The geology of the regions marked l)y this formation is quite simple.
The following description of the section along the line of the Kansas Pacific
Railroad will probably ajjply to similar sections north and south of it. The
formations referal)le to the Cretaceous period on this line are the Dakota,
Benton, and Niobrara groups, or Nos. 1, 2, and 3. According to Leconte,^
at Salina, one hundred and eighty-five nfiles west of the State line of Missouri,
the rocks of the Dakota group constitute the bluffs, and continue to do so as

far as Fort Harker, thirty-three miles farther west. They are "a coarse,

\ — .. .

' From this region from Dr. A. R. Roessler.

2 Notes on tbe Geology of tlie Snrvoy for tbe Extension of the Union Pacific Road, Eastern Division,
from the Smoky Hill to the Rio Gramlr. I'.y .Tolin L. l.ecouto, M. D., Philadelphia, 1868.

17

lintu 11 saiidslonc, containing irregular concretions of oxide of iron," and nu-
merous mollusks of marine origin. Near Fort Harker, certain strata contain
large quantities of the remains (leaves, chiefly) of dicotyledonous and other
forms of land-vegetation. Near this point, according to the same authority,
the sandstone-beds are covered with clay and limestone. These he does not
identify, but portions of it from Bunker Hill, thirty-four miles west, have
been identified by Dr. Hayden as belonging to the Benton, or second, group.
The specimen consisted of a block of dark bluish-gray clay-rock, which bore
the remains of the ^&\\ Apsojjelix sauriformis, Cope. That the eastern bound-
ary of this l)cd is very sinuous is rendered probable by its occurrence at
Brookville, eighteen miles to the eastward of Fort Harker, on the railroad.
In sinking a well at this point, the same soft, bluish clay-rock was traversed;
and, at a depth of about thirty feet, the skeleton of a saurian of the crocodilian
order was encountered, the Hijposaurus vebUi, Cope.

"The boundary-line, or first appearance, of the beds of the Niobrara divi-
sion has not been pointed out; but, at Fort Hays, seventy miles west of Fort
Harker, its rocks form the bluffs and outcrops everywhere. From Fort
Hays to Fort Wallace, near the western boundary of the State, one hundred
and thirty-four miles beyond, the strata present a tolerably uniform appear-
ance. They consist of two portions: a lower, of bluish calcareo-argillaceous
character, often thin-l)edded ; and a superior, of yellow and whitish chalk,
much more heavily-bedded. Near Fort Hays, the best section may be seen
at a point eighteen miles north, on the Saline River. Here the bluffs rise
to a lieight of two hundred feet, the yellow strata constituting the upper
iialf. No fossils were observed in the blue bed ; but some moderate-sized Os-
trece, frequently broken, were not rare in the yellow. Half-way lietween this
point and the fort, my friend N. Daniels, of Hays, guided me to a denuded tract,
covered with the remains of huge shells described by Mr. Conrad, at tiie close
of this section, under the names of Haploscaplia grandis and H. eccentrka.
They may have affinities to the oysters; some of them were 27 inches in diam-
eter. They exhibited concentric ohtusc ridges on the interior side, and one
species a large crest behind the hinge. Fragments of fish-vertebne of the
Anogmius type were also found hereby Dr. J. H. Janeway, post-surgeon.
These were exposed in the yelhtw bed. Several miles east of the [)osl, Dr.
Janeway pointed out to me an iinineiise accumulation of Inoceramtis prohlc-
iii'itirus ill the l)Iue stratum. This si)ecies also occurred in abundance in the

18

bluffs west of the fort, which were composed of the blue bed, capped by a
thinner layer of the yellow. Large globular or compound globular argillace-
ous concretions coated with gypsum were abundant at this point.

"Along the Smoky Hill River, thirty miles east of Fort Wallace, the south
bank descends gradually, while the north bank is bluffy. This, with other
indications, points to a gentle dip of the strata to the northwest. The yellow
bed is thin or wanting on the north bank of the Smoky, and is not observable
on the north fork of that river for twenty miles northward, or to beyond
Sheridan station on the Kansas Pacific Railroad. Two isolated hills, 'The
Twin Buttes,' at the latter point, are composed of the blue beds, here very
shaly, to their summits. This is the general character of the rock along and
north of the railroad between this point and Fort Wallace.

" South of the river, the yellow strata are more distinctly developed. Butte
Creek Valley, fifteen to eighteen miles to the south, is margined by bluffs of
from twenty to one hundred and fifty feet in height on its southern side, while
the northern rises gradually into the prairie. These bluffs are of yellow chalk,
except from ten to forty feet of blue rock at the base, although many of the
canons are excavated in the yellow rock exclusively. The bluffs of the upper
portion of Butte Creek, Fox and Fossil Spring (five miles south) Canons are
of yellow chalk ; and the reports of several persons stated that those of Beaver
Creek, eight miles south of Fossil Spring, are exclusively of this material.
Those near the mouth of Beaver Creek, on the Smoky, are of considerable
height, and appear, at a distance, to be of the same yellow chalk.

"I found these two strata to be about equally fossiliferous, and am unable
to establish any paleontological difference between them. They pass into
each other by gradations in some places, and occasionally present slight
laminar alternations at their line of junction. I have specimens of Empo
semianceps, Cope, from both the blue and yellow beds ; and vertebrae of the
Platecarpus glandiferus, Cope, were found in both. The large fossil of Liodon
dyspelor, Cope, was found at the junction of the beds, and the caudal portion
was excavated from the blue stratum exclusively. Portions of it were brought
east in blocks of this material, and these have become yellow and yellowish
on many of the exposed surfaces. The matrix adherent to all the bones has
become yellow. A second incomplete specimen, undistinguishable from tliis
specie.^;, was taken from the yellow bed.

"As to mineral contents, the yellow stratum is remarkably uniform in its

19

cliaracter. Tlic Wiie shale, on the contrary, frequently contains numerous
concretions, and great abundance of thin layers of gypsum and crystals of the
same. Near Sheridan, concretions and septaria are abundant. In some
places, the latter are of great size, and, being imbedded in the stratum, have
suffered denudation of their contents, and the septa, standing out, form a huge
honey-comb. This region, and the neighborhood of Eagle Tail, Colorado, are
noted for the beauty of their gypsum-crystals, the first abundantly found in
the Cretaceous formation. These are hexagonal-radiate, each division being
a pinnate or feather-shaped lamina of twin rows of crystals. The clearness
of the mineral and the regular leaf and feather forms of the crystals give
them much beauty. The bones of vertebrate fossils preserved in this bed
are often much injured by the gypsum-formation, which covers their surface,
and often penetrates them in every direction.

"The yellow bed of the Niobrara group disappears to the southwest, west,
and northwest of Fort Wallace, beneath a sandy conglomerate of uncertain
age. In color, it is light, sometimes white ; and the component pebbles are
small and mostly of white quartz. The rock weathers irregularly into holes
and fissures, and the soil covering it is generally thin and poor. It is readily
detached in large masses, which roll down the bluffs. No traces of life were
observed in it ; but it is probably the eastern margin of the southern extension
of the Miocene Tertiary stratum. This is at least indicated by Dr. Hayden in
his Geological Preface to Leidy's Extinct Mammals of Dakota and Nebraska.

"Economically, the beds of the Niobrara formation possess little value
except when burned for use as a fertilizer. The yellow chalk is too soft in
many places for large buildings, but it will answer well for those of moderate
size. It is rather harder at Fort Hays, as I had occasion to observe at their
quarry. That quarried at Fort Wallace does not appear to harden by ex-
posure ; the walls of the hospital, noted by Leconte on his visit, remained in
1871 as soft as they were in 1867. A few worthless beds of bituminous
shale were observed in Eastern Colorado.

"The only traces of glacial action in the line explored were seen near
Topeka. South of the town are several largo, erratic masses of pink and
bloody quartz, whose surfaces are so polished as to appear as though vitrified.
They were transported, perhaps, from the Azoic anja near Lake Superior."

Tlic iijUowiiig species of Vertebrata have thus liir been detected in the
Niobiiirn ti)rmati()U :

20

AVES.

a. SAURUEiE.

Ichtliyornis dispar.
Icldhyornis celer.
h. Natatores.

Graculavus anceiJs.
Hesperornis regalis

REPTILIA.

a. DiNOSAURIA.

Hadrosaurus minor,
h. Ornithosauria.

Pierodactylus umbrosus,
Pterodactylus ingens.
Pterodactylus occidentalis.
Pterodactylus velox.

c. Sauropterygia.

JLlasmosaurus platyurus
Plesiosaurus gulo.
Polycotylus latipinnis.

d. Testudinata.

Protostega gigas.
Toxochelys latiremis.
Cynocercus incisus.

e. PYTHONOMORPnA.

Liodon proriger.
Liodon. dyspelor.
Liodon nepoiolicus.
Liodon micromus.
Platecarpus ictericus.
Platecarpiis coryphceus.
Platecarpus felix.
Platecarpus curtirostris.
Platecarpus crassartus.
Platecarpus simus.
I'latecarpus latifrons.
Plalecarpus gracilis.

21

Plalecarpus glandiferus.

Platecarpus mudgei.

Plnfecarpus tectulus.

Sironectes anguUferus.

CUdastes cineriarum.

Clidastcs jmmiluf).

CUdastes vymanii.

CUdastes velox.

CUdastes dispar.

CUdastes affinis.

CUdastes tortor.

CUdastes stenops.

CUdastes planifrons.

CUdastes rex - 37

PISCES.

a. ISOSPONDYLI.

Fortlieus molossus.
Porthcus thaumas.
For the us lestrio.
Portheus mudgei.
Portheus arcuatiis.
Portheus gladius.
Iclithyodectes ana ides.
Ichthyodectes ctenodon.
Iclithyodectes hamatiis.
Ichthyodectes prognathus.
Ichthyodectes inultidentatus.
Iclithyodectes perniciosus.
Xiphactinus audax,
Daptinus phlehotomus.
Sa u rocephalus Ian cifor mis.
Saurocephalus arapahovius.
Erisichlke nitida.
Pachy rh izodus can inus.
Pach i/rh izodus Intimentum.

22

i'drlii/rliizodus kingii.
Fackyrhizodus leptopsis
Pachyrhizodus sheareri.
Tetheodus j'ephredo.
Enchodus petrosus.
EncJwdus dolichus.
Enchodus anceps.
Encliodus calliodon.
Enchodus shumardti.
Phasganodus dims.
Phasganodus gladiolus.
Phasganodus carinatus.
Empo nepceolica.
■ Empo semianceps.
Empo contracta.
Empo merrillii.
Stratodus apicalis.
b. Elasmobkanchii.

Ptychodus janevaii.

Ptychodus poly gyrus.

Ptychodus occidcntnlis.

Ptychodus mortonii.

Ptychodus tvhijfpleyi.

Galeocerdo crassidens.

Galeocerdo linrtvellii.

Otodus divaricatus.

Oxyrhina exlenta.

Lamna cuspidata.

Lamna mudgei.

Lamna macrorhiza 48

Total number of species f- 9

Of the preceding forty-eight species of fishes, the greater part are phy-
sostomous ActinoiHeri ; and there is no species of a physoclystous flxmily in the
list. No trace of spines or scales of fishes of the latter character liavc Ix-cii
yet discovered in si lata of this period in (lie West, thongh one {Bcn/x

23

insculplus, Cope) lias been discovered by Dr. Lockwood in the greensand-
marl of New Jersey (No. 5).

In the second place, it is of importance 'to observe that the genera have
nearly all been obtained from the chalk of Europe. Porjheus is represented,
perhaps, by some specimens referred to Hypsodon ; one species of Ichthyo-
dcctes is figured Ijy Dixon from Sussex, and one of Cimolichthys, Erisichihe,
and Pachyrhizodus, each. Enchodus has long been known from Holland, &c.,
Apsopelix and Stratodus being, so far, the only ones not found in Europe.
This is of much interest in every aspect, and points to the synchronism, as
generally understood, between the chalk-formations of Kansas and of
England.

Species of Mollusca are not numerous in the beds of the Niobrara epoch.
They consist chiefly of Inocerami of two or more species. Through the kind
assistance of my friends, N. Daniels of Hays, and Dr. J. H. Janeway, post-
surgeon at Fort Hays, I was enabled to procure from the yellow chalk a
number of very complete specimens of the remarkable shells already men-
tioned. I submitted them to my colleague, T. A. Conrad, and add herewith
his account of them. He thinks they possess some resemblance to the Ru-
distes; but he doubts their being truly related to that division.

Fragments of these HaploscaphcB are common in the formation, and have
been described by authors as portions of huge Inocerami.

" HAPLOSCAPHA, Conrad.

" Shell subovate or subtriangular ; hinge long and straight, edentulous,

oblique ; curved, prominent ridges occupy the upper portion of the interior,

the ridges beginning and ending at a distance from the margins of the

shell ; a singular twisted callus composes the hinge, the back of which is

transversely ribbed.

" H. GRANDis, Conrad.

" Length greater than height ; hinge-line very long ; ridges concentric,
about twelve in numl)cr, extending into the cavity under the hinge.

" This shell. Professor Cope informs' me, has been found 27 inches in
diameter. The posterior side of the right valve is elongated and dilated, and
the form of the shell is not unlike that of Meleagrina. The su])stancc is
fibrous, or ralhcr coliinuinr, and niiicli resembles that of Cdpiiiid/n, ;is (igiircii
In- d'Orhii/ny, I'xccpt Unit (he fii)erf< are transverse. Tlic exU'rior is alwjiys

24

concealed l)y <i coating of rock and a crowded mass of Ostrca congesta, and,

in some specimens, they line the cavity of the shell; the submargin is thicli.

No nuiscular impression can be traced, nnless the ridged part indicates its

station.

" Subgenus Cucullifeka.

hinge.

Shell with an upright hood-shaped process on the posterior end of the
" H. ECCENTKiCA, Conrad.

"Ovato-triangular; hinge-line short, very thick; concentric ridges pro-
found, six in numlier ; hood strongly and irregularly plicated ; cavity profound.

" This shell, with the same structure of substance as the preceding, is very
unlike it in form, and is represented by one valve only, while a number of tlie
preceding species were found. In all specimens of the two forms, the right
valve only was obtained. Whether it is allied to the family Bndisk'S of
Lamarck is a question I leave for others to decide On the margin of one of
tlie valves are attached some small shells, resembling HippurUex, and the fibers
of which the shell is composed lie in broken masses on .some valves, and even
scattered like piles of pins.

" The hood of H. exce.ntr'tcus is 2i inches in height, and the height of the
valv(! 10 inches ; length, 9 inches.

"Accompanying these fossils were many specimens of Inoceramus pro-
hlematicm, and a fragment of an undetermined species of the same genus."

IV. PIERRE GROUP.

In Nebraska and Dakota and Middle Colorado, south of the divide
between the waters of tlie Arkansas and Platte Rivers; also, the lower bed
of greensand of New Jersey ; Weber River, Wyoming,' below the coal.
Besides the numerous remains of reptiles and fishes found in NewJersey,
this formation contains saurian (mosasauroid) remains in Colorado.

V. THE FOX HILLS GROUP.

Extended i;i Central Dakota ; on the Arkansas and tributaries ; in South-
ern Colorado; and as the second greensand bed in New Jersey."

' Hiiyili'ii's Animal Uepoit, 1S70, ]i. UiT.

- F(ir ii synopsis of tlie cxtiuct ItcplUia of this ciiocli aud the preceding, see tlie antliois " Extinct
liatratliin,"' ic, N. Anieiii^a, I'Inlailclpliiii, lHGO-70.

25

VI. — THE FORT UNION OR LIGNITE GROUP.

With this o[K)ch wo enter debatable grouiul, and begin to consider strata
deposited in brackisli or Cresh waters, which were more or less inclosed by
the elevation of parts of the Rocky Mountains and other western regions, and
which are, therefore, more interrupted in their outlines than tlu; marine for-
mations which underlie them. Dr. Hayden has recognized and located a
number of formations of this character, to some of which he has applied the
name of " transition-beds.'' That the period of their deposit was one of
transition from marine to lacustrine conditions is evident; and that a succes-
sion of conformities in position of beds may be traced from the low-cst to the
highest of them, and with the Tertiary strata above them at distinct locali-
ties, beginning at the south and extending to the north, is also proven by
Hayden and others. It appears impossible, therefore, to draw the line satis-
factorily without th(! aid of paleontology; but licre, whih; cnidence of inter-
ruption is clear, from I he relations of the plants and vertel)riife animals, it is
not identical in the two cases, but discrepant. I therefore append a synopsis
of the views expressed by authors, with a presentation of tiic evidence which
is accessible in my department.

Hayden has named the following as distinct epochs of transitional char-
acter, all of v,hich he originally referred to the Tertiary period. I give them
in the order of age which he has assigned to them : ' (1.) Placer Mountain ;
locality. New Mexico. (2.) Caiion City coals, Southern Central Colorado.
(3.) Fort Union or Lignite group, Dakota, Montana, and Wyoming. (4.)
The Bitter Creek series, end)racing the Bitter Creek coals, Wyoming. (5.)
Bear River group, Western Wyoming. To these may l)e added the Judith
River beds of Montana, which Dr. Hayden has placed, with reservation,
below the Fort Union series, leaving their final location to I'utu re discoveries.

No vertebrate remains having come undcu" the author's notice from the
Placer Mountain and Caiion City formations, no further notice can l)e here
taken of them beyond the statement that they are, as Meek iiulicates, of Cre-
taceous age, not llir removed from the horizon of the coals of WcIxt Kivcr,
Utah. The presence of iimnionitcs and liaculites above and below Ihcm h.id
alr(;a(ly indicated such a conclusion to Leconte," as it has in the case of the

' Gi;i)loj;ical Survey ut Colorado, 1869, p. 90.

- I'cjiorf nil tlic Oi'olosy of tbc Smoky Hill Paeitic Kailroad route, 18(i8, p. 6(i.

4 C

2G

Weber River l)e<ls to Dr. Hayden.' To near the same horizon are perhaps to
be referred the coal observed by Professor Marsh* on the south side of the
Uintah Mountains in Utah, which were overlaid by strata containing Ostrea
congesta. This may, indeed, be referred to a still older period, as that oyster
is characteristic of No. 3, according to Meek and Hayden. The Placer
Mountain and Canon City groups are nearer to No. 5, but the precise relation
to it has not yet been determined. I therefore proceed to the proper and
original Fort Union epoch as defined by Hayden.

This extended deposit is stated by Hayden^ to extend from the Missouri
Valley to Colorado, jiassing under Tertiary beds by the way. That this is
the case has been confirmed by the researches conducted in the northern and
eastern portions of Colorado during the season of 1873 by the writer.^ I
present comparative lists of the vertebrate species known from the Platte
and Missouri Valleys in the respective Territories.

COLORADO.

Compsemys victus.
Adocus lineolatus.
Plastomenus punctulatus.
Plasotmenus insignis.
Trionyx vagnns.

*

*

DAKOTA.

Co7n2>semys victus.

Adocus lineolatus.

Plastomenus punctulatus.
* * * *

Trionyx vagans.
Ischyrosaurus antiquus.
Plesiosaurus occiduus.

Bottosa urus perrugosus.
Polyonax mortuarius.
Cionodon arctatus.
Hadrosaurus 1 occidentalis.

*

*

%

*

*

*

«

«

*

Hadrosaurus

occidentalis.

■ The identity and correspondence of the species in the two columns indi-
cate that these remote localities contain the remains of the same fauna.
Further, the presence of Dinosauria and Sauropterygia demonstrates its
mesozoic character.

A number of vertebrate remains was collected by George W. Dawson,

' Annual Report, 1870, p. 168.

-Seo an interesting article by Prof. O. C. Marsh on tbe Geology of the Eastern Uintah Mountains
in the American Journal of Science and Arts for March, 1871.
-Aiiiiii:i1 Kcpoi-t (loldi-ado, ISIJiJ, p. ^d.
*:^vv l;i;lli.-lin cfllif I'liiled .Slates (icologieal Survey, No. 1, 1^74, )>. Id.

27

of Montreal, geologist of the British American Boundary Commission, near
the line of the northern boundary of the United States (latitude 49°),. within
the drainage-area of the Milk River. The formations are regarded by him as
belonging to the Lignite or Fort Union of Dr. Haydcn, and consist of green-
ish and greenish-brown arenaceous clays of various degrees of hardness, fre-
quently including small gravel-stones, and sometimes forming a hard cement
between them. The fossils were found near the base of the formation, and
"not more than one or two hundred feet above yellow arenaceous beds, which
I conceive represent Cretaceous No. 5, and which are rapidly followed in
descending by well-marked No. 4 with characteristic fossils." (Extracted
from letter of Mr. Dawson.)

The species are the following : Clastes, sp.; Compsemys ogmius; C. ? victus;
Plastomenus costatus;^ P.coakscem; Trionyx ?vagans; Trionyx Isp.; ? Hadro-
saurus .^sp.; Cionodon stenopsis.

The dinosaurian remains are quite abundant, and indicate several species,
but are mostly so fragmentary as to be unfit for determination. The diagnos-
tic genera of this list are Compsemys, Plasto?nenus, and Cionodon; the species
referred to Hadrosaurus being represented by caudal vertebrae only. The
first-named genus is characteristic of the Fort Union epoch only ; the frag-
ment referred to C. victus, the only species of the list previously known, is too
small for final specific reference. The Plastomenus coalescens is represented by
a moi-e perfect specimen than any other species referred to this genus from the
Fort Union beds, but is not sufficiently complete to render the reference to
this Eocene genus final. It is, in any case, not a member of any other known
genus. One species of Trionyx is represented by a hyosternal bone, and is
not definable ; while the fragment referred to T. vagans, thouglt closely
resembling that species, is not large enough for final determination. The
Cionodon, thougli based on incomplete remains, is quite sufficient for paleon-
tological purposes.

In conclusion, it may be stated that there are present two genera in this
collection which are diagnostic of tiie Fort Union epoch, but no species cer-
tainly so, though two species are probably identical with species of that epoch ;
also, that the presence o^ Dinosaur ia refers the fauna to the Mesozoic series;
and that there is no .satisfactory evidence of the co-existence of these reptiles

' Two R])pcitsaro |iriivisiiiii;i11y rpfiTrod to tlic Tertiary giiiiis I'liiilomniitu, Init :iri^ (oo frii^iiiriitary for
liiial ili'd'iiiiiiiatiuii.

28

with Tertiary forms ; that the species referred to Flastomenus constitute an
indication of affinity with corresponding Eocene forms. The presence of gar-
fishes of the genus Clasles in this formation is as yet peculiar to this and tlie •
Juditii l\iver locahties. As these gars have not heretofore been found in
Nortli America below the Eocene, they constitute the first case of apparent
commingling of Tertiary and Cretaceous animal life yet clearly determined.
Yet the evidence is far from being as weighty in indication of Tertiary rela-
tions as is the presence of the saurians in question as evidence of Mcsozoic
character; for the gars, though now living, are an ancient typCj their allies
having swarmed in the Jurassic seas, and it is therefore altogether reason-
able that they should be found in fresh-water deposits of Cretaceous and
Tertiary age. The rarity of the former deposits accounts for the late date of
their discovery there.

The longest known of these transitional faunae was discovered l)y Dr.
Haydcn near the mouth of Judith River, in Montana. As determined by Dr.
Leidy, it embraces the following Vertebrata: Hadrosaurtis mirabtlis ; Palceos-
cincus costahis; Tro'udon formosus ; Auhhjsodon liorrldus; Bottosaurus? hiimilis,
Trionyx foveatas ; Clastes occidentaUs {Lepidotus, Leidy) ; Lejndotus haydenii.

The first four species of this fist are Dinosauria, antl hence diagnostic
of the Mesozoic age of the formation. The Clastes indicates relation to the
MilK' River fauna.

That the Judith River ibraiation is Cretaceous would appear to have been
the suspicion of ]\Iessrs. Meek and Ilayden when they originally described the
deposit and its invertebrate fossils. Leidy suspected that the species "indicate
the existence of a formation like that of the Wealden in Europe."' Meek and
Hayden" remark : "We are inclined to think with Professor Leidy that there
may be, at tlie base o(" the Cretaceous system, a fresh-water formation lil^e the
Wealden. Inasmuch, however, as there are some outliers of fresh-water
Tertiary in these lowlands, we would suggest that it is barely possible these
remains may belong to that epoch." From the standpoint of the writer,
these beds would be at the top of the Cretaceous, and more or less related to
the Fort Union epoch. ]\Ir. Meek expresses himself^ cautiously with refer-
ence to the age of the Fort Union and Judith River formations, as follows:

' Proceedings of the Academy of Niitiiral Sciences. Philadelphia, 1856, p. 73.
- /,. i:, wm. i>. 114.

■'II::y(l('ii'M Aiiiiiuil Report for 187'2, p. 4G0.

29

" The occurrence of [I'os.sils specified] at the Jiiditii River localities would
certainly strongly favor the conclusion, not only that this Judith I'onnation,
the age of which has so long been in doubt, is also Cretaceons, but liiat even
the higher tVesii-water Lignite formation at Fort Clark and other Upper Mis-
souri localities may also be Upper Cretaceous instead of Lower Tertiary.
That the Judith River beds may be Cretaceous I am, in the light of all now
known oi" this rosion of the continent, rather inclined to believe. But it
would take very strong evidence to convince me that tiie higher Iresh-water
Lignite series of the Upper Mis.souri is more ancient than the Lower Eocene.
'J'hat they are not is certainly strongly indicated not only by the modern
afKnities of their molluscan remains, but also by the state of preservation
of the latter."

I presume it is now apimrent that the presence of the orders Sauropte-
rygia and Dinosauria establishes, conclusively, the Mesozoic and Cretaceous
character of this fauna. This reference was made by the writer in 1869,
and was, at that time, opposed to the views extant, both geological and pale-
ontological. The following list exhibits the state of opinion on tliis point at
that time and subsequently.

1856. Meek and Hayden, Proceedings of the Academy, Philadelphia, p. 63 ;

referred them to the Tertiary.
Meek and Hayden, /. c, p. 265 ; Lignite, referred to the Miocene.
Meek and Hayden, /. c, 113 ; referred to Lower Tertiary.
Leidy, I. c, p. 312 ; Thcspesius oocidentaUs {Hadrosaurus) ; referred to

the 3{am?nrJia, and regarded as dinosaurian.
Leidy, I. c, p. 89 ; Ischyrosaurus referred to the Mammcdia as a

sirenian.
1860. Hayden, Transactions of the American Philosophical Society, p. 123;

repeats former conclusions; Leidy refers llienpesius more decidedly

to the Sauria, and questions relations of Ischyi'osaurus.
1868. Hayden, American Journal of Science and Arts, p. 204; Lignites

regartled as Tertiary, from both vegetable and animal remains,

from the Missouri and the Laramie Plains.
Leconte, Exploration of the Smoky Hill Railroad Route, |). Go ; the

Middle Colorado beds are "older tiian those of the Mis.souri or

Great Lignite bed of llaydcn, which are prul)ai)ly Miocene," ttc.

30

1869. Cope, Transactions of the American Philosophical Society, pp. 40, 9S,
243 ; supposed mammalian i-emains proven to be reptilian, and the
formation referred to the Cretaceous.

1871. Newberry, in Hayden's Annual Report, pp. 95-96 ; Lignite flora
regarded as Miocene.

1874. Dawson, Report of Progress of the British North American Boundary

Commission : on tlie Tertiary Lignite Formation, p. 20 ; Milk River
beds regarded as lowest American Tertiary.
Cope, Bulletin of the United States Geological Survey of the Territo-
ries, No. 2 (April) : Fort Union beds of Colorado referred to the
Cretaceous.

1875. Cope, Proceedings of the Academy of Natural Sciences, Philadelphia

(January) ; Milk River beds regarded as Cretaceous.

From the above, it appears that both paleontologists and stratigraphers,
excepting the writer, have maintained the Tertiary age of the beds of the
Fort Union epoch.

Whether the Bitter Creek and Bear River groups of Hayden present
much ditlerence of horizon remains to be determined. For the present, they
are retained as distinct.

VII. — THE BITTEE CREEK SERIES,

mentioned by the writer as a distinct group, in the Proceedings of the Amer-
ican Philosophical Society, 1872 (pubUshed on August 12), is apparently
regarded by Mr. Meek also as representing a distinct epoch.' He says, " The
invertebrate fossils yet known from this formation are, in their specific rela-
tions, with possibly two or three exceptions, new to science, and diflferent
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." On this account, and because of the great stratigraph-
ical diflferences exhibited by the Bear River and Evanston coal-strata, I have
followed Hayden in regarding the Bear River group, on the west side of the
Bridger basin, as representing a distinct series of rocks, with present knowl-
edge. On this account I omit, as heretofore, allusion to determinations

I Ilajdcn'M Animal Rciioi't for 1872, pp. 4S9-4G1 ; piiblihliod April, 1873.

31

of age of the latter ioniiation as irrelevaut. in discussing the age of the Bitter
Creek epoch.'

My own observations on tlic relations of these rocks, made during the
summer of 1872, have been, in a measure, anticipated by the detailed reports
of Messrs. Meek and Bannister," which, with the older observations of Dr.
Hayden and Mr. Emmons (of King's survey), leave little to be added. How-
ever, as none of these gentlemen paid especial attention to the vertebrate
paleontology, the bearing of this department in relation to the stratigraphy
remains to be explained.

As Dr. Hayden remarks, the Union Pacific Railroad, at Black Butte
station, passes through a inonoclinal valley ; the rocks on both sides having a
gentle dip to the southeast. This dip continues to the ea'stward to near
Creston, where the beds pass under the newer Tertiary strata. Following
the railroad westward from Black Butte, the same dip continues to near Salt
Wells, where we cross an anticlinal axis, the dip of the strata being gentle to
the northwest. There are minor variations in the dij), but the general result
is as stated. They disappear five miles east of Rock Spring station, beneatii
the later beds of the Green River Tertiary, which, at this point, presents a
line of strike extending northeast and southwest, across the railroad, in the
form of a range of bluffs, of considerable elevation. They are composed of
lighter-colored and softer material than the Bitter Creek strata. The latter
consist of alternating beds of hard and soft sandstone, with argillaceous and
carbonaceous strata. The upper part of the series contains eleven coal-strata.
At Rock Spring, I was informed that the upper was ten feet in thickness,
and the next, four feet. Returning eastward, the heavier-bedded sandstone is
low in the series at Point of Rocks, in consequence of the southeast dip, and
tiie upper beds are softer and abound in fossil shells. At Black Butte station,
the heavy sandstone-bed disappears from view toward the east, and the eleven
coal-strata appear above it. About twenty feet above the sandstone, between
two of the thinner beds of coal, the bones of the Agathaumas sylvestris were
found, iml)edded in leaves and sticks of dicotyledonous plants, cemented
together by sand and clay. Where the heavy sandstone-bed disappears below
the level of the track of the railroad, in the course of its eastern dip, a thin

' Tliis cmirsi! li;is liefii iiiisiiiulorstood by Mr. Meek and otlieis :us iiii|ilyiii^ a design to i<;iio.c tlioso
(Ibtcruiiiiatioiis. Uolh Mr. liiiiiiions and Mr. Mi'iOi arc clear in tbi; i-xprrssioii of tlmir conilu.sioiis a.s to
the ago of (ho ISoar River i'|iiii;!i.

'Si:;; ll:!\d.-:i'.i Ammti.iI Keport for l^li. pp. l.'iT, '<!>.

32

bed of coal, just above it, soon follows; then a bed of shells containing oys-
ters, more and less numerous at different points, may be traced lor some
distance before it also disni^pears. Near the latter point, a bed of Melanians
and other fresh-water shells is seen a few feet above them.

A section, carried for eight miles south of Black Butte station, exhibits
the relation of the Bitter Creek series to the superincumbent Tertiarics very
instructively. The whole series rise slightly to the southward, and more dis-
tinctly to the westward, so as to form an escarpment as the eastern border of
an open valley, which extends south from the railroad, just west of the station.
The heavy bed of sand-rock is here, as elsewhere, the landmark and strati-
graphical base-line. Moving south from the railroad, we keep along the strike
of the lower coal-beds. Just above the sand-rock, the softer stratum thickens,
and six miles from the station is covered with the debris of immense numbers
of Lcptesthes crassatelUformis. Passing over the edges of the strata, toward
the southeast, I connted eight beds of coal, separated by various short inter-
vals, the eighth being the heaviest, and five or six feet thick. Above this one,
three thin beds of lignite were crossed in succession, each accompanied with
an abundance of leaves of chiefly dicotyledonous plants. Then came the
ninth bed of coal, and then, in order, three more beds of lignite, with abun-
dant leaves. During this time the ascent became less steep, and a number of
short, level tracts were passed before reaching the upper bed of lignite.
Beyond this, I passed another short Hat, which was marked by a number of
worn banks of the light-ash color that distinguishes the material of the bluffs
of the Grreen River Tertiary which overlie the coal-series near liock Sjirings.
I had not ridden a (juarter of a mile before reaching a low line, from which
one of my men picked up a jaw of a small mammalian, allied to the Bridger
Hyopsodus or to Hyracotlierium of the Eocene of France and Switzerland, and
a number of PaludinaASkG shells. I had thus reached the summit of the
Bitter Creek formation, wliich did not appear to be much more than 350 feet
al>ove its base at the railroad. In full view, a mile or two to the south, rose
the first of the benches which constitute the horizons of the Green River
formation. Between this and the first mammal-producing bed, just descril)ed,
rose three banks, one beyond the other, measuring, altogether, 120 ieet ; perhaps
the lowest was 10 feet above the first bank, and this one not more elevated
above the last lignite and leaf-bed. In all of these I found l)()nes of (ircen
River Verhbrala exceedingly abundant, but all dislocated and scaltcn'd, ^o as

33

to be rarely in juxtaposition. These consisted of the following species :
Fishes : Clastes 1 glaher ; Reptiles : Emys megaulax ; E. pachylomus ; E. eu-
tlinetus; Trionyxscutumantiqunni ; Alligator heterodon; Mammals: Orohijypus
vasacciensis ; and fragments too imperfect for determination.

In the third bank, in immediate juxtaposition with the remains just
enumerated, I found another thin bed of lignite, but this time without any
visible leaves. In a fourth line of low blutis, a little beyond, I found that
remarkable mammal, Meialophodon armatus, with its dentition nearly complete,
in connection with fragments of other mammals and reptiles.

Behind these rises the first line of white bluff's already described, which
extend away to the east ; to the west, they soon terminate in a high escarp-
ment in north and south line with that of the Bitter Creek beds, already men-
tioned as bounding a north and south valley. This and the suj)erjacent strata
which we pass over in going south appear to be conformable to those of the
Bitter Creek series beneath them. I say ''appear," for slight differences of
dip are not readily measured by the eye; yet I suspect that the conformability
is very close, if not exact, and similar to that mentioned by Meek and Bannis-
ter as exhibited by the beds of the Washakie group which lie upon the coal-
series east of Creston. The white bluffs add perhaps one hundred feet to
the elevation. On their summit is a thin bed of buflT clay and sand-rock sim-
ilar to the upper strata of the Bitter Creek series, and containing numerous
shells and some scattered teeth and scales of fishes. I called Mr. Meek's
attention to the specimens of these shells, which I sent him, and his reply
was that most were of identical species with those of the coal-series (Creta-
ceous), and that they presented no general peculiarity.

At a short distance to the southward, another line of white bluffs extends
across the line of travel. This is not more elevated than the preceding one.
I only found remains of tortoises on it. Several miles to the south we reach
another bench, whose bluflfy face rises four or five hundred feet in buttress-
like masses, interrupted at regular in*;ervals by narrow terraces. Tliis line is
distinguished f'or its brilliantly-colored strata extending in horizontal bands
along the escarpment. They are brilliant cherry-red, white, true purple witli
a bloom shade, yellow, and pea-green, forming one of th(> most beautiful dis-
plays I ever beheld. The lower portions-are bright-red, wiiicli color predom-
inates toward the west, where the bluffs descend to a lower elevation. 1
fi)und on them remains of a turtle {Emys cvtlnHtus, Cojje) and some borings

34

of a ? worm in a hard layer. On top of these are clay and slate rocks of a
muddy-yellow color, with their various ledges rising to perhaps two hundred
feet. Continuing now to the southeastward, along the old stage-road, we cross
South Bitter Creek at the old Laclede station. Some miles south and cast
of this point, a band of buff sandstones form a bluff of fifty or more feet in
elevation. Below it lie more white or ashen beds, which contain i-emains of
mammals and turtles, rather decayed. A short distance beyond these, and
forty miles from Black Butte station, we reach tiie base of the enormous pile
of sediment which I have called the Mammoth Buttes. These form a horse-
shoe-shaped mass, the concavity presenting south and eastwardly, the summit
narrow, serrate, and most elevated to the east, and descending and widening
toward the south. I estimated the height of the eastern end to be at least one
thousand feet above the plain surrounding it. Numerous mammalian remains^
demonstrated that this mass is a part of the Bridger Eocene; although, as Mr.
Emmons, of King's survey, informs me, no continuous connection with the
principal area west of Green River can be traced. The total thickness of
the Green River and Bridger formations on this section cannot be far from
two thousand five hundred feet, at a very rough estimate.

The point of transition from the Cretaceous to the Tertiary deposits, as
indicated by the vertebrate remains, is then in the interval betvi'een the last
plant-bed at the summit of the buff mud-rocks and the mammal-bone deposit
in the lowest of the ash-gray beds. Below this line, the formation must be
accounted as Cretaceous, on account of the presence of the dinosaurian Aga-
thau?nas sylvestris, and those above it, as I have already pointed out, Eocene,"
on account of the types of Mammalia contained in them.

The authorities on the Bitter Creek formation have presented views
more or less at variance with those entertained by the writer, or of such a
dubious character as to fall very far short of the requirements of evidence.
Dr. Hayden has regarded them as Tertiary and as transitional from Creta-
ceous to Tertiary. Mr. King, in his very fall article on the Greon River Basin,
definitely refers the lower part of the series to the Cretaceous, in the follov/-
ing language:^ "We have then here the uppermost members of the Cretaceous
series laid down in the period of the oceanic sway, and quite freely charged

'See tlio Monster of MaiDuioth Buttes, Penn Monthly Magazine for .ingnst, 1873.
^Ou Bathmodon, an extinct genus of Ungulates, February 10, 1872, Hayduu's Annual Keport (1872),
p. 431 ; Annual Eeport for 1872, p. 645.

3 Exploration of tlic Fortieth Parallel, p. 458.

35

willi the fossil relics of marine life; tlien an uninlernipted passage of con-
formable beds tlirough the brackish period up, till the whole Green River /
basin became a single sheet of fresh water.'' He regards the line of the Lu
npper bed of oysters as the summit of the Cretaceous, and the superimposed
beds as Tertiary, in the following language (page 453): "while the fresh-
water species which are found in connection with the uppermost coal-beds
seem to belong to the early Tertiary period." He thus places the line some
distance within what I have regarded as the Cretaceous boundary.

Mr. Lesquereux, as is known, regards these beds as Tertiary, not only
on account of their vegetable fossils, but also on account of the stratigraphic
relations of the formation. His conclusion to this effect is consistent throuirh-
out, and is a fact of the highest importance in this connection.

Mr. Meek has fully discussed the age of this series in his interesting
article in Hayden's Annual Report for 1872, the general tenor of whicli is
indicated by the passage I have quoted from the opening of his remarks, in the
beginning of the present notice of the Bitter Creek beds. His opinions may
be cited as follows: In the Annual Report for 1870, he determined the same
beds visible at Hallville as Tertiary; in that of 1871, three species of oysters
from other parts of the Bitter Creek beds are placed in the Cretaceous list,
each one with question as to the identification of species, a point, in the case
of oysters, of first importance in the determination of the age of the deposit.
The remarks in this report, as well as those in Mr. King's report, refer either
to the much lower Weber River coal or to the different area of the Bear Rivcr
group, and are consequently noticed under that head.

In a paper on the age of these beds, published August 12, 1872, the writer
asserted the Cretaceous age of the series. On this, Dr. Bannister, tjic com-
panion of Mr. Meek, writes^ "that Mr. Meek, and, I believe, Mr. Emmons also,
had considered that these beds might be Cretaceous; but lliis was rather on
account of the change in the fossil fauna from pur(;ly fresh-water, as in the
characteristic Tertiary of this region, to brackish-water marine, and tiie spec-
ific affinities of a few of the fossils to California Cretaceous species, than from
any very positive evidence. As far a.3 I know, the only evidence of this kind
is in the identification by Professor Cope of the saurian remains found by us
at Black Butte."

It only remains to observe that the strata and coal of the Bitter Creek

' AoQual Ueport for 1873, p. 534.

36

group of the Cretaceous are either wanting on the western and southern
borders of the Green River basin, or are concealed by the superincumbent
Tertiaries. Instead of these, a comparatively thin bed of apparently unfos-
siliferous quartzite or sandstone lies at a high angle against the bases of the
Uintah^ and Ham's Fork Mountains, respectively, on beds of Jurassic age,
which are probably Cretaceous No. 1 (Dakota). Tlie beds observed by Pro-
fessor Marsh, on the south side of the Uintah Mountains, on Brush Creek,
belong neither to the Dakota nor Bitter Creek epochs, but perhaps to No. 3,
if, as Professor Marsh asserts, the oyster found in a superjacent stratum is
Ostrea congesta, Con. It is, in any case, of no later date than the Canyon City
or Weber River coals. Hence, the assumption of some writers that this dis-
covery determined the age of the Bitter Creek series to be Cretaceous is
without foundation in fact.

VIII. — THE BEAR KIVER GROUP,

of Hayden, occupies, according to him, a distinct basin, to the west of an
anticlinal axis, which separates it from that of Green River. It is buried
under Tertiary beds, the age of which has been a question of interest, and
will be hereafter considered. In order to determine the relations of the two
basins, a section was carried across the rim of the eastern, starting from the
Fontanelle Creek, eighty miles north of the Union Pacific Railroad, and con-
tinuing toward the upper waters of Ham's Fork of Green River to the
westward. My notes are as follows :

The beds of the Green River epoch dip gently from the point where
my last notes left them, near to Rock Spring station, toward the northwest,
all the way to Green River. The upper strata become slaty in character,
and descend to the water-level at the river, where they form a high bluff. In
these slates occur the fish-beds discovered by Dr. Hayden, as well as the
insect-beds noticed by Messrs. Denton and Richardson. They are worn into
towers and other picturesque forms at Green River City (see Hayden's
Annual Report for 1870). Passing north from the railroad, up the valley of
Green River, the slates display a gentle dip to the north, and eighteen miles
beyond have disappeared from view. On both sides of the river huge mesas
of the Bridger formation come into view; those ou the east extending to the

' See Hayden's Annual Report for 1870 ; Marsli, American Journal of Science and Arts, March,
1871.

37

Big Sandy River, and those on the west to Ham's Fork. A.t Slate Creek,
further to the north twenty miles, a yellowish-brown sandstone rises into view,
and continues to increase in importance toward the north. At the mouth of
Fontauelle Creek, it rises on the east side of the river to a heigh'; of perliaps
250 feet, but sinks toward the north and east, from near the mouth of La-
barge Creek, fifteen miles up the river. North of Labarge, a similar bed of
sandstone rises again, ajid is immediately overlaid by white shales, resembling
those of the Green River epoch, which have here a great thickness. Oppo-
site the mouth of the Labarge, their lower strata are bright-red ; but, on the
west side of the river, the sandstone only is visible. All the beds rise to the
north ; the red beds forming the summits of the cliffs in that direction.

In passing up Fontanelle Creek to the westward, the heavy beds of buff
sandstone gradually descend, and the white shales come into view. I exam-
ined the former for lignite and coal, but found none. There are several thin
beds of a tough, carbonaceous material in the white shales (which I take to
be of the Green River epoch). In the lower strata in this locality, as well as
on the east side of Green River above the mouth of Labarge Creek, are nu-
merous remains of fishes similar to those of Green River City, with insects
and their larvae, shells like P«^?a and Cyrena, and millions of Ct/prLs. The
larvae are dipterous, some nearly an inch long, and others minute and in pro-
digious numbers. With them are found stems of plants, but no leaves.
These beds rise with a very gentle dip, and, twenty miles from the mouth of
the creek, terminate against stecplj^-inclined strata of earlier age. At this
point the lower beds exhibit the bright-red colors that are so often seen in the
lower part of the formation at other points. The uplifted beds form a ridge
of high hills, having a north by east and south by west trend, across which
the Fontanelle cuts its way in a deep cafion. This range is monoclinal ; the
strata dipping 45° east, and their outcrop on the summit and western face.
The first bed which forms the surface of the incline is rather thin, and is
composed of a reddish quartzite without fossils, no doubt of Cretaceous age.
lielow it is a stratum of highly fossiliferous bluisli limestone of Jurassic age,
coniaining Pentacrin us asteriscus, M. and H. ; Trigonia, &c. Below this, a
reddish sandstone presented a similar thickness, which may represent the
Trias, which rests on a bluish-shale formation. We have now reached the
base of the western side of the hills ; from their summit, wc have had a Ijcau-
tiful and inicres)in<r view ol" i!;('()l(»i,nc;d stniclurc Tiir viillcv. of thrci- or lour

38

miles in width, is bounded on the west side by a range of low mountains,
whose summits are well-timbered. The valley is excavated at an acute angle
to the strike of the strata, so that, as far as the eye can reach to north and
south, successive hog-bacUs issue, en echelon, from the western side, and run
diagonally, striking the eastern side many miles to the southward. At the
cafion of the Fontanelle, six of these hog-backs occupy the valley, and the
number varies as we proceed down the valley. The structure changes from
the same cause, as we explore in either direction. The dip of all these hog-
back strata is, to the west and sligiitly north, less steep at the eastern side,
but reaching 45°, and a still higher angle at the middle and west side of the
valley. There appears to be an anticlinal near the base of the eastern range,
which has been deeply excavated ; from its western slope (in the valley), the
upper beds, seen in the eastern range, have been carried away, leaving only
probable Triassic and Carboniferous strata exposed. In one of the latter, I
found a well-marked horizon of carbonaceous shales, extending as far as I
explored them. Toward the western side of the valley, the descending strata
are sandstones, but, whether identical with that of the eastern hills of Creta-
ceous age, I could not ascertain. Lower down the valley (to the south),
similar beds form a high, vertical wall of very light color, the scenery resem-
bling that of the Garden of the Gods in Colorado. I suspect that the exist-
ence of more than one fold can be demonstrated in these hog-backs and
mountains.

The result, which bears on the history of the Bear River group, is, that,
on this side of the Green River basin, the Bitter Creek epoch is either want-
ing, or represented by a thin layer of red quartzite (or, perhaps, Cretaceous
No. 1), and that no coal of Cretaceous age exists along its western rim.
After following the valley to Ham's Fork River, and proceeding a short dis-
tance along it, toward the southeast, I crossed a thin bed of coal in the
upturned edges of the same beds crossed in the valley above. The discovery
of the extension of the fish and insect beds sixty miles north of the principal
localities is a point of interest in Tertiary geology.

The Ham's Fork Mountains form the divide between the waters of Green
and Bear Rivers, respectively, and is passed by the Union Pacific Railroad at
and west of Aspen station, as is described by Dr. Hay den (Annual Report
for 1870, p. 149). He here points out that the distinctness of the two basins
was marked during the Tertiary [jcriod. and lieuce names the deposits of the

39

western area the Walisatch group, regarding it, at tiie same time, as synchro-
nous with those of the Green River epoch. The writer has attained the
same opinion on paleontological grounds, and has hence employed the same
name for botli areas, viz, the Green River epoch.'

As ah-cady stated," the upper or red-banded Tertiary beds of this locality
yielded the following species : Perissodaetyle bones, two species ; Oroh/])pus
vasacciensis ; Crocodilus, sp. ; Alligator heterodon ; Trionyx scutumantiquum ;
Emys testudineus ; E. gravis ; Clastes glabcr ; Unio, two species.

The lower sandstone-beds yielded the following mammals: BathmodoH
radians; B. semicinctus; B.latipes; Orohippus index ;^ Pkenacodus prinmvus.

West of the contact of Bear River with the Tertiary bluffs, the strata
consist of sandstone and conglomerates, and dip at about 30° to the northeast.
Five hundred feet vertically lielow tlie Bathmodon bed, a stratum of impure
limestone crops out, forming the slope and apex of a portion of the bluff. In
this I found the following vertebrates: Reptiles : Trionyx scutumanliquum ;
Etwjs ? euthnetus ; Fishes: Rhineastcs calvus ; Clastes glaber.

In comparing this list with that given for the lower beds of the Green
River epoch, where they overlie the Bitter Creek coal, such resemblance may
be observed as is sufficient to identify the two series.

This is the nearest to a determination of the age of the Evanston coal-
bed, which Hayden regards as the most important west of the Missouri
River that I have been able to reach. From the limestone just described to
the coal-bed two miles to the west, the strata are very similar in character,
and apparently conformable, so that they appear to belong to the same series.
Dr. Hayden confesses his inability to correlate them with those of Bear River
City and Weber River, but discovered remains of plants which were identi-
fied with some of those known to occur in the Fort Union beds on the Lara-
mie Plains and the Upper Missouri If this be the case to a sufficient
extent, the Evanston coal must be referred to that division of the Cretaceous
period. This conclusion is, however, only provisional, and Dr. Bannister's
remarks* are much to tlie point. He says, of the upper beds northeast of
Evanston (the ones 1 describe above): "There seems to have been a consider-
able disturbance besides the mere tilting of the beds; and, from the altered

'Proceedings of the Acaileuiy of Nafnial Sciences, 1872, p. 279.

'Proceedings of the Aiiicricaii Philosophical Society, 1872, p. 47.3.

^ Cope, Paleontological Uiillctin, No. 17, 1873 ; also, The Extinct Vcrtcbrata of Now Mosico.

■•Ilaydcii's Annual Report for 1872, p. 541.

40

direction of the strike/ we were led to suspect considerable lateral displace-
ment with faulting, which might very possibly cause the appearance of the
same beds both here and at the coal-mines, although at first sight these would
appear much higher in geological position. * * I do not know the
grounds of Professor Cope's reference of the coal at this point to the Creta-
ceous, while he admits the Tertiary age at least of some of the overlying
sandstones ; but, as we found no break nor line of demarkation in the whole
2,000 feet or more which we examined, and found our fossils in coal-bearing
beds immediately above and conformable to the main coal, the facts, so far as
they are known to me, do not seem sufficient for such identification." This
point offers, therefore, a more complete continuity in stratification and min-
eral character, from the Cretaceous to. Tertiary deposits, than any other which
I have had the opportunity of examining.

CONCLUSION.

Having traced the transition series of the coal-bearing formations ofgthe
Rocky Mountain region from the lowest marine to the highest fresh-water
epochs, it remains to indicate conclusions. I have alluded but cursorily to
the opinions of Mr. Lesquereux and Dr. Newberry as based upon the study of
the extinct flora. The former has, as is well known, pronounced this whole
series of formations to be of Tertiary age, and some of the beds as high as
Miocene. The material on which this determination is based is abundant,
and it must be accepted as demonstrated beyond all doubt. I regard the
evidence derived from the mollusks in the lower beds and the vertebrates in
the higher as equally conclusive that the beds are of Cretaceous age. There
is, then, no alternative but to accept the result that a Tertiary Jlora was
contemporaneous with a Cretaceovs fauna^ establishing an uninterrvpted succes-
sion of life across what is generally regarded as one of the greatest breaks in
geologic time. The appearance of mammalia and sudden disappearance of
the large mesozoic types of reptiles may be regarded as evidence of migration
and not of creation? It is to be remembered that the smaller types of lizards
and tortoises continue, like the crocodiles, from Mesozoic to Tertiary time

'Hayden's AuDual Report for 1872, p. 541.

2 The circumstance of the discovery of a Mesozoic Dinosaur, Agatlmumas syh-estris, with the cavities
of, and between, his bones stnffed full of leaves of Eocene plants (Lesquereux), would prove tliis proposi-
tion to.be true, had no other fossils of either kind ever been discovered elsewhere!

= Griticism of this conclusion bv Professor Dawsou is noticed in the Report U. S. Geol. Snrv. Terrs.
1873 ,p. 1 14.

41

without extraordinar}- modification of structure. It is the Dinosauria which
disappeared from tiic laud, driven out or killed In- the more active and intel-
ligent mammal. Herbivorous reptiles like Agathaumas and Cionodon would
have little chance of successful competition with beasts like the well-armed
Bathinodon and Metalophodon.

It then appears that the Transition series of Ha}'den is such not only iu
name but in fact, and that paleontology confirms in a highly satisfactory
manner his conclusion, ''already shown many times, that there is no real
physical break in the deposition of the sediments between the well-marked
Cretaceous and Tertiary groups."'

1 Annual Keport 1870, p. ICC. Also Geol. Survey Colorado, 18C9, p. 197, Dr. Hayden observes:
" There is no proof, so far as I have observed, in .ill the western country of true non-conformity between
the Cretaceous and Lower Tertiary beds, and no evidence of any cliange in sediments or any catastro-
phe sufficient to account for the sudden and apparently complete destruction of organic life at the close
of the Cretaceous period."
6

P^RT II.

DESCRIPTIONS OF VERTEBRATE FOSSILS.

DESCRIPTION OF THE FOSSILIFEROUS LOCALITIES AND ANCIENT LIFE OF KANSAS.

The vast level tract of onr territory occupied by the Niol)rara Cretace-
ous, and lying between Missouri and the Rocky Mountains, represents a con-
dition of the earth's surface which has preceded, in most instances, the mount-
ainous or hilly type so prevalent elsewhere, and may l)e called, in so far, in-
completely developed. It does not present the variety of conditions, either
of surface for the support of a very varied life, or of opportunities for access
to its interior treasures, so beneficial to a high civilization. It is, in fact, the
old bed of seas and lakes, which has been so gradually elevated as to have
suffered little disturbance. Consistently with its level surface, its soils have
not l)een carried away by rain and flood, but rather cover it with a deep and
wide-spread mantle. This is the great source of its wealth in nature's crea-
tions of vegetable and animal life, and from it will be drawn the wealth of its
future inhabitants. On this account, its products have a character of uni-
formity ; l)ut, viewed from the standpoint of the political philosopher, so long
as peace and steam bind the natural sections of our country together, so long
will the plains be one important element in a varied economy of continental
extent. But they are not entirely uninterrupted. The natural drainage has
worn channels, and the streams flow below the general level. The ancient
sea and lake deposits have neither been pressed into very hard rock beneath
piles of later sediment, nor have they been roasted and crystallized by internal
heat. Although limestone-rock, they easily yield to the action of water; and
so the side-drainage into the creeks and rivers has removed their high banks
to from many rods to many miles from their original positions. In many cases,
these banks or blufishave retained their original steepness, and have increased
in elevation as the breaking-down of the rock encroached on higher land. In
other cases, the rain-channels have cut in without removing the intervening
rocks at once, and formed deep gorges or canons, which sometimes extend to

43

great distances. They frequently communicate in every direction, forniiiig
curious labyrinths; and, wlien the intervening masses are cut away at various
levels, or left standing, like monuments, we have the characteristic peculiari-
ties of "bad lands,'' or mauvaiscs terres.

In portions of Kansas, tracts of tliis kind are scattered over the country
along the margins of the river and creek valleys and ravines. The upper
stratum of the rock is a yellow chalk, the lower bluish, and the brilliancy of
the color increases the picturesque effect. From elevated points, the plains
appear to be dotted with ruined villages and towns whose avenues are lined
with painted walls of fortifications, churches, and towers, while side-alleys
pass beneath natural bridges or expand into small pockets and caverns,
smoothed by the action of the wind carrying hard mineral particles. But
this is the least interesting of the peculiarities presented by these rocks. On
the level surfaces, denuded of .soil, lie huge oyster-like shells, some opened
and others with both valves together, like remnants of a half-finished meal of
some titanic race, who had been frightened from the board never to return.
These shells are not as much thickened as many fossil oysters, but contained
an animal which would have served as a meal for a large party of men. One
ot them measured twenty-six inches across.

If the explorer searches the bottoms of the rain-washes and ravines, he
will doubtless come upon the fragment of a touin or jaw, and will generally
find a line of such pieces leading to an elevated position on the bank or bluff
where lies the skeleton of some monster of the ancient sea. He may find
the vertebral column running far into the limestone that locks him in his last
prison ; or a paddle extended on the slope, as though entreating aid ; or a pair
of jaws lined with horrid teeth, which grin despair on enemies they are help-
less to resist; or he may find a conic mound, on wiiose apex glisten in the sun
the bleached bones of one whose last office has been to preserve from destruc-
tion the friendly soil on which he reposed. Sometimes a pile of huge remains
will ])e discovered, which the dissolution of the rock has deposited on the
lower level; the force of rain and wash having been insufficient to carry them
away.

But the reader inquires, What is the nature of liicse creatures thus left
stranded a tiiousand miles from cither ocean ? How came they in the lime-
stones of Kansas, and were they denizens of land or sea ? It may bf replied
that our knowledge of this chapter uf ancient iiistory i.s only aliout live years

44

old, and has been brought to light by geological explorations set on foot by
Dr. Turner, Professor Mudge, Professor Marsh, W. E. Webb, and the writer.
Careful examinations of the remains discovered show that they are nearly all
to be referred to the reptiles and fishes. We find tliat they lived in the period
called Cretaceous, at the time when the chalk of England and the greensand-
marl of New Jersey were being deposited, and when many other huge rep-
tiles and fishes peopled both sea and land in those quarters of the globe. Tlie
thirty-seven species of reptiles found in Kansas up to the present time varied
from ten to eighty feet in length, and represented six orders, the same that
occur in the other regions mentioned. One only of the number was terres-
trial in their habits, and four were fliers; the remainder were inhabitants of
the salt ocean. When they swam over what are now the plains, the coast-
line extended from Arkansas to near Fort Riley, on the Kansas River, and
passing a little eastward traversed Minnesota to the British possessions, near
the head of Lake Superior. The extent of sea to the westward was vast, and
geology has not yet laid down its boundary; it was probably a shore now sub-
merged beneath the waters of the North Pacific Ocean.

Far out on the expanse of this ancient sea might have been seen a huge,
snake-like form, which rose above the surface and stood erect, with taper-
ing throat and arrow-shaped head, or swayed about, describing a circle of
twenty feet radius above the water. Then plunging into the depths, naught
would be visible but the foam caused by the disappearing mass of life.
Should several have appeared together, we can easily imagine tall, flexible
forms rising to the height of the masts of a fishing-fleet, or hke snakes,
twisting and knotting themselves together. This extraordinary neck — for
such it was — rose from a body of elephantine proportions, and a tail of the
serpent-pattern balanced it behind. The limbs were probably two pair3 of
paddles like those of Plesiosaurus, from which this diver chiefly diflfered in
the arrangement of the bones of the breast. In the best-known species,
twenty-two feet represent the neck in a total length of fifty feet. This is the
Elasmosaurus platyurus, Cope, a carnivorous sea-reptile, no doubt adapted for
deeper waters than many of the others. Like the snake-bird of Florida, it
probably often swam many feet below the surface, raising the head to the
distant air for a breath, then withdrawing it, and exploring the depths forty
feet below, without altering the position of its body. From the localities in
wliich the bones have been found in Kansas, it nnist have wandered far from

45

land ; and that many kinds of fishes formed its food is shown by the teeth
and scales found in the position of its stomach.

A second species of somewhat similar character and habits differed very
mnch in some points of structure. The neck was drawn out to a wonderful
degree of attenuation, while the tail was relatively very stout, more so, indeed,
than in tlie Elasmosaurus, as though to balance the anterior regions while
occupied in various actions; e. g., while capturing its food. This was a pow-
erful swimmer, its paddles measuring four feet in length, with an expanse,
therefore, of about eleven feet. It is known as Polycotylus latipinnis, Cope.

The two species just described formed a small representation, in our great
interior sea, of an order which swarmed, at the same time or near it, over the
gulfs and bays of old Europe. There they abounded twenty to one. Per-
haps one reason for this was the almost entire absence of the real rulers of
the waters of ancient America, viz, the Pythonomorphs. These sea-serpents —
for such they were — embrace more than half the species found in the lime-
stone-rocks in Kansas, and abound in those of New Jersey and Alaba-na.
Only four have been seen as yet in Europe.

Researches into their structure have shown that they were of wonder-
ful elongation of form, especially of tail; that their heads were large, flat,
and conic, with eyes directed partly upward ; that they were furnished with
two pairs of paddles like the flippers of a wiiale, attached by short wide
peduncles to the body. With these flippers and the eel-Hke strokes of their
flattened tail, they swam, some with less, others with greater speed. They
were furnished, like snakes, with four rows of formidable teeth on the roof of
the mouth. Though these were not designed for mastication, and, without
paws for grasping, could have been little used for cutting, as weapons for
seizing their prey they were very formidable. And here we have to consider
a peculiarity of these creatures, in which they are unique among animals.
Swallowing their prey entire like snakes, they were without that wonderful
expansibility of throat due in the latter to an arrangement of levers support-
ing the lower jaw. Instead of this, each half of that jaw was articulated or
jointed at a point nearly midway between the ear and the chin. This was of
the ball-and-socket type,and enabled the jaw to make an angle outward, and thus
widen by much the space inclosed between it and its fellow. The arrangement
may be easily imitated by directing the arms forward, with the elbows turned
outward, and the hands placed near together. The ends of these bones were

46

in the Pijthonomorpha as independent as in the serpents, being only bound by
flexible ligaments. By turning the elbows outward and bending them, the
space between the arms becomes diamond-shaped and represents exactly the
expansion seen in these reptiles, to permit the passage of a large fish or other
body. The arms, too, will represent the size of jaws attained by some of the
smaller species. The outward movement of the basal half of the jaw neces-
sarily twists in the same direction the column-like bone to which it is sus-
pended. The pecuhar shape of the joint by which the last bone is attached
to the skull depends on the degree of twist to be permitted, and therefore to
the degree of expansion of wdiich the jaws were capable. As this diflfers
much in the different species, they are readily distinguished by the column or
" quadrate " bone when found. These are some curious consequences of this
structure, and they are here explained as an instance of the mode of recon-
struction of extinct animals from slight materials. The habit of swallowing
large bodies between the branches of the under jaw necessitates the prolonga-
tion forward of the mouth of the gullet ; hence the throat in the Pythono-
morpha must have been loose and almost as baggy as a pelican's. Next, the
same habit must have compelled the forward position of the glottis or open-
ing of the windpipe, which is always in front of the gullet. Hence these
creatures must have uttered no other sound than a hiss, as do animals of the
present day which have a similar structure ; as, for instance, the snakes.
Thirdly, the tongue must have been long and forked, and for this reason : its
position was still anterior to the glottis, so that there was no space for it
except it were inclosed in a sheath beneath the windpipe when at rest, or
thrown out beyond the jaws when in motion. Such is the arrangement in
the nearest living forms, and it is always in these cases cylindric and forked.
The giants of the Ptjthonomorpha of Kansas have been called Liodon
proriger, Cope, and Liodon dyspelor, Cope. The first must have been abundant,
and its length could not have been far from seventy-five feet; certainly not less.
Its physiognomy was rendered peculiar by a long projecting muzzle, remind-
ing one of that of the blunt-nosed sturgeon of our coast ; but the resemblance
was destroyed by the correspondingly massive end of the branches of the
lower jaw. Though clumsy in appearance, such an arrangement must bave
been effective as a ram, and dangerous to his enemies in case of collision.
The writer once found the wreck of an individual of this species strewn
around a sunny knoll beside a Ijluff, and his conic snout pointing to ihc

47

heavens formed a fitting monument, as at once bis favorite weapon, and the
mark distinguishing all his race. The Liodon di/spelor was probably the
longest of known reptiles, and probably equal to tbe great finner-whales of
modern oceans. The circumstances attending the discovery of one of these
will always be a pleasant recollection to the writer. A part of the face, with
teeth, was observed projecting from the side of a blufi' by a companion in
exploration, Lieut. James H. Whitten, United States Army, and we at once
proceeded to follow up the indication with knives and jjicks. Soon the lower
jaws were uncovered, with their glistening teeth, and then the vertebrae and
ribs. Our delight was at its height when the bones of the pelvis and part of
the hind limb were laid bare, for they had never been seen before in the
species, and scarcely in the order. While lying on the bottom of the Creta-
ceous sea, the carcass had been drag'ged hither and thither by the sharks and
other rapacious animals, and the parts of the skeleton were displaced and
gathered into a small area. The massive tail stretched away into the blutf,
and, after much laborious excavation, we left a portion of it to more persever-
ing explorers. The discovery of a related species {Platecarpus coryphceus,
Cope) was made by the writer under circumstances of difficulty peculiar to
the plains. After examining the bluffs for half a day without result, a few
bone-fragments were found in a wash above their base. Others led the way to
a ledge forty or fifty feet from both summit and foot, where, stretched along in
the yellow chalk, lay the projecting portions of the whole monster. A con-
siderable number of vertebra3 were found preserved by the protective embrace
of the roots of a small bush, and, when they were secured, the pick and knife
were brought into requisition to remove the remainder. About this time, one
of tbe gales, so common in that region, sprang up, and, striking the bluff
fairly, reflected itself upv^ard. So soon as the pick pulverized the rock, the
limestone-dust was carried into eyes, nose, and every available opening in the
clothing. I was speedily blinded, and my aid disappeared in the canon, and
was seen no more while the work lasted. A handkerchief tied over the face,
and pierced by minute holes opposite the eyes, kept me from total blindness,
though dirt in abundance penetrated the mask. But a fine relic of creative
genius was extricated from its ancient bed, and one that leads its genus in size
and explains its structure.

On another occa^on, riding along a spur of a yellow chalk-blulf, some
vertebrae lying at its foot met my eye. An examination showed that the series

48

entered the rock, and, on passing round to the opposite side, the jaws and
muzzle were seen projecting from it, as though laid bare for the convenience
of the geologist. The spur was small and of soft material, and we speedily
removed it in blocks, to the level of the reptile, and took out the remains as
they lay across the base from side to side.

A genus related to the last is Clidastes. They did not reach such a size
as some of the Liodons, and were of elegant and flexible build. To prevent
their contortions from dislocating the vertebral column, these had an addi-
tional pair of articulations at each end, while their muscular strength is
attested by the elegant striae and other sculptures which appear on all their
bones. Ten species of this genus occur in the Kansas strata, the largest
(Clidastes cineriarum, Cope) reaching forty feet in length. A smaller species,
of elegant proportions, has been called C. tortor, Cope. Its slenderness of body
was remarkable, and the large head was long and lance-shaped. Its lithe
movements brought many a tish to its knife-shaped teeth, which are more
efficient and numerous than in any of its relatives. It was found coiled up
beneath a ledge of rock, with its skull lying undisturbed in the center. A
species distinguished for its small size and elegance, is C. pumilus. Marsh.
This little fellow was only twelve feet in length, and was probal)ly unable to
avoid occasionally furnishing a meal for some of the rapacious fishes which
abounded in the same ocean.

The flying saurians are pretty well known from the descriptions of
European authors. Our Mesozoic periods had been thought to have lacked
these singular forms until Professor Marsh and the writer discovered remains
of species in the Kansas chalk. Though these are not numerous, their size was
formidable. One of them (Pterodacti/lus occidentaUs, Marsh) spread eighteen
feet between the tips of its wings, while the P. umhrosus, Cope, covered
nearly twenty-five feet with his expanse. These strange creatures flapped their
leathery wings over the waves, and, often plunging, seized many an unsuspect-
ing fisli ; or, soaring at a safe distance, viewed the sports and combats of the
more powerful saurians of tlie sea. At night-fiill, we may imagine them
trooping to the shore, and suspending themselves to the cliffs by the claw-
bearing fingers of their wing-limbs.

Tortoises were the boatmen of the Cretaceous waters of tlic eastern
coast, but none had been known from the deposits of Kansas until very
recently. The largest species on record (Protostega gigas, Cope) is strange

49

cnougli to excite the attention of naturalists. It is well known that the house
or boat oi llie tortoise or turtle is t'ornied hy the expansion of llie iisu;il
bones of the skeleton till they meet and unite, and thus l)ecome continuous.
'I^hus the lower shell is iornied of united ribs ol" tiie breast and of the
breast-bone, with bone deposited in the skin. In the same way. the roof
is formed by the union of the ribs with bone deposited in the skin. In
the very young tortoise, the ribs are separate as in otiier animals ; as they
grow older, they begin to expand at the upper side of the upper end, and
with increased age the expansion extends throughout the length. The
ribs first come in contact, where the process commences, and in the kuid-
tortoise they are united to the end. In the sea-turtle, the union ceases a little
aijove the ends. The fragments of the Protostega were seen by one of the
men projecting from a ledge of a low blutr. Their thinness and the distance
to which they were traced excited my curiosity, and I straightway attaciced
the bank with the pick. After several square feet of rock had been removed,
we cleared up the tloor, and found ourselves well repaid. Many long, slender
pieces of two inches in width lay upon the ledge. They were evidently ribs,
with the usual heads, but behind each head was a plate like the flattened
l)owl of a huge spoon, placed crosswise. Beneath these stretched two broad
plates, two feet in width, and no tliicker than binder's board. Tlie edges
were fingered, and the surface hard and smooth. All this was quite new
among fully-grown animals, and we at once determined that more ground must
be explored for further light. After picking away the bank, and carving the
soft rock, new masses of strange forms were disclosed. Some bones of a
large paddle were recognized, and a leg-bone. The shoulder-blade of a huge
tortoise came next, and further examination showed that we had stumbled on
'the burial-place of the largest species of sea-turtle yet known. The single
bones of the paddle were eight inches long, giving tlie spread of the expanded
flippers as consideral)ly over fifteen feet. But the ril)s were those of an ordinary
turtle just hatched, and the great plates represented the l)oiiy deposit in tlie
skin, which, commencing independently in modern turtles, unites with each
otiier at an early day. But it was incredible that the largest of known turtles
should be but just iiatched, and for this and other reasons it has been con-
cluded that this "ancient mariner" is (»ne of those forms not uncommon in
old days, whose incompleteness in some respects points to the truth of the
belief that animals have assumed their modern perfections l)y a process of

growth from more simple beginnin>rs
7 c

50

The Cretaceous ocean of tlie West was no less remarkable tbr'its fishes
tliau tor its reptiles. Sharks do not seem to have been so common as in the
old Atlantic, but it swarmed with large predaceous forms related to the
salmon and saury.

VcrtebrjB and other fragments of these species project from the worn
limestone in many places. I will call attention to perhaps the most formida-
ble as well as the most abundant of these. It is the one whose bones most
frequently crowned knobs of shale, which had been left standing amid sur-
rounding destruction. The density and hardness of the bones shed the rain
off on either side, so that the radiating gutters and ravines finally isolated the
rock mass from that surrounding. The head was some inches longer tlian
tliat of a fully-grown grizzly bear, and the jaws were deeper in proportion to
their length. The muzzle was shorter and deeper than that of a bull-dog.
The teeth were all sharp cylindric fangs, smooth and glistening, and of irreg-
ular size. At certain points in each jaw they projected three inches above
the gum, and were sunk one inch into deep pits, being thus as long as the
fangs of a tiger, but more slender. Two pairs of such tangs crossed each
other on each side of the end of the snout. This fish is known as Forthcus
molossus, Cope. Besides the smaller fishes, the reptiles no doubt supplied the
demands of its appetite.

The ocean in which flourished this abundant and vigorous life was at last
completely inclosed on the west by elevations of sea-bottom, so that it only
communicated with the Atlantic and Pacific at the Gulf of Mexico and th(;
Arctic Sea. The continued elevation of both eastern and western sliores
contracted its area, and when ridges of the sea-bottom reached the surface,
Ibrming long, low bars, parts of the water-area were inclosed, and connection
with salt-water prevented. Thus were the living beings imprisoned and sul)-
jected to many new risks to life. The stronger could more readily capture
the weaker, while the fishes would gradually perish through the constant
freshening of the water. Willi the death of any considerable class, the l)al-
ance of food-supply woukl be lost, and many larger species would disappear
from the scene. The most omnivorous and enduring would longest resist the
approach of starvation, but would finally yield to inexorable fiite ; the last one-
caught by the shifting bottom among shallow pools, from which his exhausted
energies could not extricate him.

REP'J'ILIA.

LITERATURE OF THE SUBJECT.

1834. Harlan, in Transactions of the American Philosophical Society, v. IV,
p. 405, pi. XX, figs. 3-8. Mosasaurus missuriensis. Snout de-
scribed.

1845. Gohlfuss, ill Nova Acta Acad. K. L. C. Nat. Cur., XXI, p. 179, pis. vi,
vii, viii, ix. Mosasaurus 7naximiliani described (with malar arch)
(possibly remaining part of Harlan's specimen).

1856. Leidy, in Proceedings of the Academy of Natural Sciences, Philadel-
phia, p. 8t^. Ischijrosaurus antiquus described.
Leidy, /. sup. c, p. 73 ; six species of reptiles from the Fort Union beds

of Judith River described.
Leidy, I. sup. c, p. 311. Hadrosaurus occidentalis and two tortoises
from the same beds described.

1-8G0. Leidy, Transactions of the American Philosophical Society, p. 145. All
the species of the Fort Union epoch redescribed and figured.

1868. Cope, in Leconte's Notes on the Geology of the Extension of the Union

Pacific Railroad, Eastern Division, on the Smoky Hill, p. Q)'6.
Elasmosnurus platyurus described.

1869. Co|)e, in Proceedings of the Boston Society of Natural History, p. 250,

January {separata). The ordcv Pi/tkononiorpha defined; the num-
ber of cervical vertebrae, structure of the posterior regions of the
cranium, of the lower jaw, scapular arch, and tore-limb first deter-
mined ; genus Platecarpus defined.

Cope, in Proceedings of the American Philosophical Society, June.
Articulation of splenial bone in Pythonomorpha described ; genus
Polycotylus described.

(Jope, Transactions of the American Phihisophical Society. Part I of
Synopsis of Extinct Batrachia, Reptilia, and Aves of North Amer-
ica (to p. 104), issued in August. Elasmosaurus and Polycotylus
described and Reptilia of Fort Union epoch determined.

1870. Cope, in Transactions of the American Philosophical Society, same

work, part II (to p. 235), issued in April. Order Pythonomorpha
further described, its genera further defined, and Liodon jrronger
described.

52

Cope, Proceedings of the Aiiiericaii Philosophical Society, p. 574. Ij'io-
don dyspelor, Platecarpus nmdgei, P. ictericus, and Clidastes chicrin-
ruin described.

1871. Marsh, in American Journal of Science and Arts, p. 472 (separata Jnne).

VoW\s 0^ Pythonomorpha first described, and the hind limbs for
the first time stated to exist , and the species Clidastes dispar, tjc/ox,
vymanii, and pumilus, and Ptei'odactylus owenii described.

Cope, in letter to J. P. Le-sley, in Proceedings of the American Philo-
sophical Society, p. 168 (separata October). Hind liml) of a Py-
thonomorph first described, and Platecarpius latispmus and P. cras-
sartus described.

Cope, I. c, p. 172 (separata October). Protostega described, with hind
limbs and pelvis of Liodon dyspelor. Testudlnata first recorded.

Cope, /. c, p. 264, December. Catalogue o{ Pythonomorpha found in
the Cretaceous strata of Kansas. Pterygoid bones first deter-
mined, and hind limbs described ; genera determined, and six new
species described.

1872. Cope, /. c, p. 308 (separata January). Cynocercus incisus and Hyposaurus

vebbii described.

Marsh, American Journal of Science and Arts {separata March 9).
Pterodactylus ingcns and P. %-elox described. Name of P. owenii,
changed to P. occidentalis. Scuta o^ Pytlwnomorpha described.

Cope, Proceedings of the American Pliilosophical Society (separata
March 11). Pterodactylus umbrosus and P. harpyia described.

Cope, /. c, p. 403 (separata March). Protostega gigas fully described,
and southern and eastern species named.

Marsh, American Journal of Science and Arts (separata May). Deter-
mination of position of phalanges of fore-limb, and of two addi-
tional hones connected with the opistliotic in Pythonoiuorpha; cor-
rection of position of the quadrate and confirmation of the previous
locations of the malar arch and stapes ; description of six new spe-
cies, and better definition of the genera Platecarpus and Liodon.

Co|)c, in Proceedings of the Academy of Natural Sciences, Philadel-
phia (separata June 6th). Descri[)tion of Plesiosaurus gulo.

Coj)e, in /. sup. c. (separata June 8th). Review of Professor Marsh's
paper of May, 1872; corrections of questions of nomenclature and
disc(iv(M'v.

53

Cope, in Proceedings of the American Pliilosopliicnl Society (.sejfarafa
August 12). Genus Agathnumas described.

lS7;->. Cope, Proceedings of the Academy of Natural Sciences, Philadelphia,
p. 10. Toxoclielys latiremis described.
Leidy, Iveport of the United States Geological Survey of the Territo-
ries, by F. V. Hayden. Clidastes affinis and Plesiosaurus occiduus
described and other known species figured.

1874. Cope, in Hayden's liulletin of the United States Geological Survey of the
Territories, No. 2 (issued April). Species of Cionodoii, Polyonax,
Bottosaurus, Triony.i; PlastonieHuy, Adocus, C/idasfcs, and ,Liodon
described; structure of Z?/'«os«wr/a of the Fort Union epoch eluci-
dated.

DINOSAUEIA.

AGATHAUMAS, Cope.

Proceedings of tbo American Pliilosopbical Society, 18T'J, p. 482.

The characters of this genus are derived from the typical species A. syl-
vcstrh, which is represented by dorsal and lumbar vertebrae and an entire
sacrum, with tlie ilia, one nearly entire, ribs, and a number of other bones the
character of which have not yet been positively ascertained. One of these
resembles the proximal part of the pubis; others, portions of the sternum, &c.

On eight (and, perhaps, nine) vertebra?, anterior to the sacrum, there is
no indication of the capitular articular facet for the ril). This fticet is found,
as in Crocodllia, at or near the base of the elongate diapophyses. The centra
are slightly concave posteriorly, and still less so on the anterior face, with
gently convex margins. The neural canal is very small, and the neural arch
short, and quite distinct from the centrum, having scarcely any suture. The
neural arch has a sul)cubical i'orm, partly truncated above by the anterior
zygapophyses. In like manner, the base of the combined neural spine and dia-
pophyses are truncated below by the square-cut posterior zygapophyses. The
diapophyses are long, and directed upward ; they are triangular in section.

There are eight (and, perhai)s, iiine) sacral vertebra?, which exhibit a con-
siderable diminution in the diameters of the centra. The diapophyses and
neural arches are shared by two centra, the anterior part of a centrum bear-
ing the larger portion of both. The diapophyses are united distally in pairs;
each pair inclosing a large foramen. 1'he anterior is the most massive

54

rest on Iheiliiiiu; the posterior pair llie most expanded; tlie superior mar-
gins oi' its posterior edge form an open V, with the apex iorward on the
neural arch of the fifth vertebra. On the last sacrals, the diapophyses rise to
the neural arch again. The exits of the sacral spinal nerves are behind the
middles of the centra, and continue into grooves of the sides in all but the
last vertebrjE. The reduced and rather elongate form of the last sacral verte-
l)ra induces me to believe that this animal did not possess such large and
short caudal vcrtebrte as are found in the genus Hadrosaurus, and that the tail
was a less massive organ.

The iUinii is much more elongate than the corresponding element in
Hadrosaurus, Cetiosaurus, or Megalosaurus. Its ujiper edge is turned and
thickened inward above the anterior margin of the acetabulum, and here the
middle of the conjoined diapophyses of the second and third sacral vertebra;
was applied when in place. In front of this point, the ilium is produced in
a straight line and a stout flattened form with obtuse end. Posterior to it, its
inner face is concave to receive the second transverse rest of the sacrum, and
the superior mai-gin is produced horizontally toward the median line like the
corresponding bone in u bird. The posterior j)ait of the bone is the widest;
for it is expanded into a thin plate and produced to a considerable length.
From one of the margins (my sketch, made on the ground, represents it as
the upper), a cylindric rod is pi'oduced still farther backward. This it is
believed is only the shaft of a displaced rib. The base of the ischium is co-
ossified with the ilium, and is separated behind its base from the iliac portion
of the acetabulum. There is no facet nor suture for the pubis at the front
of the acetabulum.

The ribs are compressed. There are no bones certainly referable to the
limbs.

The form of the ilia distinguishes this genus from those known hereto-
fore.

Agathaumas sylvestrls, Cope.

Proceedings of the American Pliilosopbical Society, 1872, p. 4"i'J.

The last nine dorsal vertebras have rather short centra ; the most poste-
rior the shortest. They are higher than wide; the sides are concave, the in-
ferior face somewhat flattened. The neural arch is keeled behind from the
canal to between the posterior zygapophyses, and a similar keel extends from
the base of the neural sjiinc to between the anterior zyga])oj)hyses. The

55

neural spine is elevated uud compressed; the diapophysis is convex above
and concave along the two inferior faces, most so on tin' posterior. The
articular face of the first sacral vertebra is wider llian deep. The eight sac-
ral vertebrae are flattened below, in all except th(! first, by a plane which is
separated from the sides by a longitutlinal angle. The neural sphies of the
anterior five sacral vertebra; are mere tuberosities. A large sutural surface
for attachment of a transverse process is seen on the posterior Ihird of the
eighth sacral vertebra, which descends nearly as low as the plane of the in-
ferior surface. On the 1 tenth sacral, there is no such process, but its neural
arch and that nf" the '. nintli support transverse processes. I^hese are mon;
likr those of the dorsals in having lliree stmng l)a.'<nl supportiny; ribs, the
anterior and posterior extending for some distance along the arch.

Either naturally, or in consequence of distortion, the plate ol" the ilium
is at a strong angle to the vertical axis of the acetabulum, and at the posterior
p,nt of it its plate pre.sents a free margin on the outside as well as the insi(h"

of the femoral articulation.

Measurements.

M.

l.,eu^tli of tbe niiib posterior dorsal vertcbrte 0. 8-^0

Lciiylli of the nine sacral vertebra; (36} iuches) 0.93U

Ltiigth of right ilium (two pieces, 0.84+0."<i2), (41 inches) I.OIJO

Length of eighth dorsal from the sacrnm 0.090

Length of the base of the ueurapoijhjsis 0. 085

l)e]ith of the articular face 0. 153

Widlh of tlie articuhir face 0. 123

Lengtli of the second from sacrum 0.070

Depth of t lie ailiiular face 0. 155

Width of the articular face 0. 137

Elevation of the neural canal 0. O^")

Width of the neural canal (). 028

Elevation to the face of the zygapophjses 0. 104

Elevation to the ba.se of the neural spine 0. 150

Length of t he diapophysis from the lower base 0. 200

Length of the diapophysis from the capitular arti( ulati< n 0. 125

Antero-posterior width above 0. O.'O

Antero-posterior width of the base of the neural spine (i. 070

Antero-posterior width at the zygapnphysis. 0.078

Length of the neural spine (fragment) 0.200

Width of the centrum of tlie tirst sacral 0. KiO

Depth of the centrum of tlie tirst sacral (to the nenrapopbysis) 0. 14.")

Length of the centrum of the tirst sacral 0. 100

Leugrli of the centrum of tlie seventh sacral 0. lOO

Depth of tin; centrum of the seventli sacral (behind) 0.085

Wid'.h of the centrum of the seventh sacral (behind) 0. 100

Expanse of the second sacral transverse sujiport (22 inches) 0. 560

Length of the ilium anterior to the acetabulum 0. 470

Length of the acetabulum 0.200

Length of the ilium posterior to the acetabulum 0.390

Width nf the ilium at the anterior extremity 0. 140

56

M.

Width of the ilium at the frout of tho aoetabuhiiu 0. 210

Width of tho ilium at tho posterior expansion 0.200

Thickness above the acetabulum 0. 0(50

Width of the acetabulum 0. 105

Width of the basis of the ischium 0. 0H5

Wulthof theshaft of arib 0.062

Other bones, not yet deteniiined, will be iiicliided in the description in
the final report.

Tiiis species was no doubt equal in dimensions to the largest known ter-
restrial saurian s (jr niaramals. /

HADROSAURUS, Leidy.

Hadrosaukus mirabilis, Leidy.

Numerous bones of Dinosauria, from the Milk River, British America,
submitted to me by Dr. G. M. Dawson, of the British North American
Boundary Commission, embrace specimens appropriate to this species, but
not certainly referable to it. The locality is nearer to the bad lands of Judith
River, from which the H. mirabUi-s was procured, tlian to any other exposure
of the Fort Union beds.

There are portions of femora, humeri, and ilia, but the only piece suffi-
ciently characteristic for description is a median caudal vertebra. The
inferior ridges and facets for chevrons are strongly developed ; the latter
appearing at both ends. The anterior articular face is plane ; the posterior,
concave. The centrum is deeper than wide, slightly narrowed below ; no
traces of diapophyses. Length, 0"".060 ; depth of articular face, O^.UTT ;
width of articular face, 0"'.071 ; length of base of neural arch, 0™.038. The
reference of this vertebra is uncertain, but it seems too small tijr the H. occi-
dentalis, although in the opisthocoeliau character it is similar.

Hadrosaurus occiDENTALis, l^eidy.

Referred by Professor Leidy to a distinct genus under the name of
Thespesius, on account of the slightly opisthocoeliau character of the large
caudal vertebra?. Teeth unknown.

Fra<(ments of a large Dinosaur, from Colorado, were found associated
with species of tortoises identical with those found in Dakota, in tlie horizon
which contains the H. occidentaUs (see under head of Cionodon ardatm), and
may possibly belong to it. I have no identical parts in the two for com-
parison.

57

Char, specif. — Tlie largest fraginent of a long bone is jjrobably fioin I he
proximal end of llie til)ia; it inehidcs the curved border of the inner .si<le, and
the inner posterior tul^erosity, with five inches of the inner side of the sliaH.
The su|)erficial layer is marked with numerous closely-i)laced longitudinal
grooves, which are replaced at intervals by a few coarser and deeper ones,
which interrupt the angle with the articular surface, giving it a lobate margin.
There was probably a prominent cnemial crest. Another fragment exhil)its
one flat i)laue, and a concave posterior face. It comes I'roni near the extremity
of the humerus or the femur ; it was found near the fragment of the tibia. The
sacral vertebra is probably that of an animal not fully grown, as it was not co-
ossified with those adjacent. The articular extremities are expanded, and pre-
sent distinct faces for articulation for the large diapophyses. The one e-xtremity
is more expanded and less thickened, the other more thickened and less
dilated; on this rests the greater part of the base of the neural arch. Just
at the extremity of this base, the large sacral nervous foramen issues, which
is continued in a wide groove downward between the transverse expansions.
Inferior surface convex. As compared with the fourth sacral vertebra of
Agatltaumas sylvestris, Cope, which it nearly resembles in size, it is to be
observed that the anterior extremity is less expanded transversely as compared
with the posterior ; that the bases of support for the anterior diapophyses are
not produced downward so far; that the sides of the centrum are nearly ver-
tical, and not sloping obliquely toward the middle line; and tluit there is no
inferior plane separated from the lateral by a longitudinal angle, as in A.
sylvestris. It differs in like manner from the third and second sacral verte-
brae, and still more from tlie first of the latter saurian.

Memxirements.

M.

Leuf^tli of tbo centrum of the fourth sacral vertebra 0. 092

Trausverse diameter :

III front 0.103

At middle 0.072

Posteriorly 0.121

Vertical diameter, posteriorly 0. 01)2

Diameter of tho bead of the tibia, anturo-posteriorly 0. 200

CIONODON, Cope.

Bulletin of the lluitcd States Geological Survey of tbc Territories, No. 1, 1874, p. 2.

Remains of species of Dinosauria were obtained at two localities in
Colorado, not many miles apart; the greater number at one of them, from
8c

58

which also all the ciocodiliau aud turtle I'eiiiaius were derived. Thosie from
the other deposit consist of poi'tious of limb-bones, apparently of a single
individual of gigantic size. The more abundant fragments are referable to
three species. A fragment of a limb-bone is very similar to jiortions from
the other locality, and associated is a sacral vertebra of appropriate size and
characters. All of these were, therefore, referred provisionally to a single
species under the name of Agathaumas milo, l)ut are here described under
Hadromurus occidentalis. The remaining specimens fall into two series. In
the one, the bones are occupied by a heavy mineral, and the surfaces covered
by a white layer, which is marked by irregular ridges, as lliougli produced by
deposit along tbe lines of small adherent foreign bodies. In the other set,
the l)ones are lighter, more spongy, and not covered with the white layer;
some of them are stained by the sesquioxide of iron. Both present vertebrje
and limb-bones, which are related appropriately as to size and structure; that
is, the larger limb-bones have the same mineral character as the larger verte-
brae, and the smaller as the smaller. These limb-bones represent correspond-
ing parts in the two, and, diifering widely, confirm the belief in the existence
of two species derived from the different types of vertebrae. In these fossils,
then, I sec evidence for the existence of two species of two genera, which I
name — the larger, Pohjonax mortuarms ; the smaller, Clonodon arctntus.
Both genera present a solid, cancellous filling of femora, tibiae, and other long
bones, and hence differ from such genera as Hadrosawus, Hypsibeiiia, L(Blai>s,
and others. Cionodon differs in dentition from all Dinosauria where that part
of the structure is known; but it remains to compare Polyonax \\\i\\ Tro'udon
and Palceoscincus of Leidy, which are known from the teeth only, while no
portions of dentition are preserved with the specimens at my disposal.

Ckar. gen. — Established primarily on a portion of the right maxillary
bone, with numerous teeth in place. The posterior portion exhibits a suture,
|)robal)ly for union with the palatine bone, while the rest of the interior margin
is free. It is removed some distance from the tooth-line in con.sequence
of the horizontal expanse of the bone, while the outer face is vertical.

The teeth are rod-like; the upper portion subcyliudric in section, with
the inner tiice flattened from apex to base, whiic (he lowci' half is flattened by
an abrupt excavation to the middle, for the accomnioilalinn of the crown of
the successional tooth. The inner face of I lie tooth, from apex to base, is
shielded liy n plate of enamel, which is somewhat elevated ;il the iiiiirgins, and

59

supports a keel in tlic middle, thus giving rise to two shallow loiigitudiual
troiiijhis. Tlie reiiKiiiuler of the tooth is covered with a layer of some dense
substance, possibly cenientuin.
which overlaps the vanishing
ninrgins of the enamel. The
outer inferior excavation of the
shaft presents a median longitud-
inal groove to accommodate the
keel of the closely-appressed
crown of the successional loolli.
The apex of the tooth being ob-
tusely wedge-shaped, the func-
tional tooth is pushed forward
and transversely toward the inner
side of the jaw. The tooth slides

downward in a closely-fitting ver- ^'^o- l.-Diagr-im of the umsillary dentitiou of Cionodon

arctatiig : o, griucliiig-fncK ; ft, superior or radical view of
tical groove of the outer alveolar the maxillary boue. natural size.

wall. The inner wall is oblirpie, its section forming with that <jf the outer a V ;
it is furrowed with grooves similar and opposite to those of the outer wall, but
entirely disconnected from them. The base of the shank of the functional
tooth, on being displaced by the successional, slides downward and inward
along the groove of the inner side; each lateral movemeni being accompanied
by a corresponding protrusion. At the most, three teeth form a transverse
line ; namely, one new apex external, one half-worn crown median, and the
stump or basis of a shank on the inner. The new crowns are, however, pro-
truded successively in series of three in the longitudinal direction also. Thus,
will 11 an apex is freshly protruded, the shank in front of it is a little uunr
prominent, and the third stands beyond llie alveolar border. As each shank
increases somewhat in diameter downwird in the ('. (itrfafxs, the section
increases in size with pndrusion : hence, befon- the appearance oi a new
crown outside of it, there are but two functional teeth in a cross-row. Thus,
in the outer longitudinal row, only every third tooth is worn by functional use
at one time ; in the middle series, all are in use ; while, in the inner, every t bird
one is simultaneously thrown out in the form of a minute stump of the shank,
if not entirely ground u|).

The dorsal vertebrae, are opisthocoelian, the anteiinr more cnmpressed

60

than the posterior ; capitular articular faces, if existing, are slightly marked.
The zygapophyses are but little j)rominent lieyond the arch. A caudal verte-
bra is plano-concave, with rather depressed centrum, a little longer than broad.
The condyles of the femur have a short arc and chord ; the head of the tibia
displays a large cnemial crest, but is not emarginate behind.

The type of dentition exhibited by this genus is, perhaps, the most com-
plex known among reptiles, and is well adapted for the comminution of vege-
table food. While the mechanical effect is quite similar to that obtained by
the structure of the molars of ruminating mammals, the mode of construction
is entirely altered by the materials at hand. Thus, the peculiarly simple form
and rapid replacement of the reptilian dentition is, by a system of complica-
tion by repetition of parts, made to subserve an end identical with that
secured by deep plication of the crown of the more specialized molar of the
mammal.

Cionodon is evidently allied to Hadrosaurus, but displays greater dental
complication. In that genus, according to Leidy, the successional crowns
appear on the froni side of the shank of the tooth, not behind, and below the
base of the enamel-area, so that the tooth is distinguished into crown and
shaft. It also follows, from this arrangement, that the successional tooth does
not appear until its predecessor has been worn to the root, in which case
there can be only one functional tooth in a transverse section, instead of two

or three.

Cionodon akctatus. Cope.

Bulletin, /. c, i>. 2. ' '

Char, specif. — The enamel-plate of the tooth extends from apex to near
the base of the shaft. Its margins are thickened and without serration, while
the surface generally is nearly smooth. The dense layer over the remainder
of the tooth is much roughened by a great number of shorl, serrate, and
somewhat irregular longitudinal ridges.

31easurements.

M.

Width of the alveolar groove - 0. 0120

Length of a triad of teeth on the alveolus 0. 0140

Lengt h of an iiuworu tooth 0. 0^)50

Diameter of the surface of attrition of a tooth of the middle row :

Longitudinal 0.0063

Transverse 0.0072

Width of the maxillary bone 0.0350

Depth of the maxillary hone at the inner margin 0.0140

61

What I suppose to bo ilio poslrrior end dl' I lie inaxilliiry hoiic nxliihits
the grooves to near its apex as well as a consideral)lc surface of articulation
for the malar.

Two dorsal vertebreB are preserved, whose neural arches are coossified,
with a trace ol" the suture remaining. Both articular faces exhibit a transverse
fossa for ligamentous or bursary attachment. Round these, on the convex
face, there are transverse rugosities, while oliliquc-ridgcd lines descend on
each side from the floor of the neural canal. The centra are shorter than
deep, and subquadrate in a horizontal section. The sides are concave ; the
anterior one compressed with lenticular vertical section with the angle below.
The more posterior is less compressed, and the surface is smooth ; in the an-
terior, it is thrown into weak longitudinal ridges near the edges of the articular
extremities. There are large nutritious foramina on the sides. The neura-
pophyses are excavated vertically on their posterior edges. Neural canal on
the anterior dorsal, a broad vertical oval. A caudal vertebra is rather elongate
and depressed ; as it has no diapophysis, it is not from the anterior part of the
series. There is no prominent lateral angle, but the two inferior angles con-
necting the chevron -facets are well marked. Neurapophysis only measuring
half the length of the centrum. The articular faces exhibit the same trans-
verse tbssa as is seen in the dorsals. The anterior is plane, the posterior

reniformly concave.

Measnrements.

Anterior ilorsal : M.

Length of tbo centrum 0.074

Elevation of the articnlar face 0.0?:?

Width of tbo articular face 0. OTII

Vertical diameter of the neural canal 0. 027

Elevation of the anterior zygapophysea 0. l'i'2

Middle dorsal :

Leugth of the centrum 0. 0G8

Elevation of the articular face - 0. 08.'>

Width of the articular face 0.080

Middle cautlal :

Length of the centrum 0. 0(')2

Elevation of the articular face (at canal) 0. 047

Width of the articular face 0. OCH

Width between the inferior angles n. H-'l

Width of the uenral canal 0. Ui:!

The femur is only represented by the distal end, witii the condyles per-
fectly preserved. Tiie latter form a single trochlear surface, whost; borders
form arcs of circles. It is slightly hour-glass-shaped, chiefly by excavation
of the posterior lace, which is, however, shallow, the deep fossa; seen in

62

Hadrosdurus and other genera l)eing absent. The area of the ailieiilnr car-
tilage is clearly marked out, and the dense snrface ol' the shaft is marked with
delicate striae, which terminate at the edge of the former. One side of the
end of the bone is nearly plane, the other is longitiidinally excavated ; some
shallow grooves furrow the angle with the trochlear face. The section of the
shaft, three inches from the end, is a wide, transverse parallelogram. This
bone looks no little like the distal end of a metapodial bone, but there are
Yarious reasons why it is more probably femur or humerus. The presence ot
the tibia, especially, determines it to be the former element.

The head and distal end of the tibia, with six inches of the shaft, are pre-
served. The former relates with the end of the femur, resembling it both in
size, simplicity of contour, and details of surface. The form is crescentoid ;
one horn being the cneinial crest, the other posterior, and replaced by a short
truncation. The inner (convex) flrce is rendered angular by a median tuber-
osity, and all round this margin shallow grooves cut the solid angle at irreg-
ular distances. The articular face displays the smooth area, and the shaft the
delicate strise seen in tlic femur. The distal end is nnsymmetrically lenticu-
lar in section, one side being more convex ; the aiticular face; is rugose, show-
ing a fixed ligamentous articulation with the astragalus. The convex iiice of
the shnft is coarsely striate-grooved near the extremity ; on the other side, the
intervening ridges are represented by exostoses or rugosities. The flatter side
becomes the more convex on the lower ])art of the shnft.

Miitr.uvcinnti:!.

Ti-.iDsverso iliameter: il.

Of the coudyles of the Hiiiiar - (I. OH-i

Of tbo shaft of tho femur 0. Of.:!

Diiimeter, fore and aft :

or the iiiidclU) of the condyles 0. (J54

Of the side of the condyles U. 001)

Of the shaft 0.0?8

Diameter of the head of the tibia:

Greatest (1.102

Fore and aft 0. 091!

Transverse 0. UOO

Diameter of the shaft of tibia ((iroxinially) :

Transverse 0. OoO

Foreandaft 0.045

Diameter of the distal end of the tibia :

Transversely 0. 1 15

FortJand afc 0. 0(iO

Remarks. — If the bones above described as pertaining to the hind linil)
are really such, they are smaller as compared with the dorsal vertebra* than
in Hadro^diiyns fau/hri. \\\h\ iiidifiite an aniniid as lar<fe as a imrse.

G3

ClONODON STENOPSI8, Cop<'.

'i'liis 1 )iii()saiir was (liscovcicd by CTeoigc; M. Dawson, ol" Muiit real, geolo-
gist of the 15rilish Nortli American Boundary Commission, in (lie Fort Union
beds of the IMilk River region. It is represented by fragments oC maxillary
hones, with a tew contained pieces of teeth. Probably, several ol'llie numer-
ous i)ones of reptiles of this order, obtained by Mr. Dawson, pertain to the
same genus and species; but there are as yet no means by which to distin-
guish them from other species of Hadrosaurirhc in the collection.

The maxillaries exhil)it the vertical grooves characteristic of the genus,
and the teeth are of the rod-like form of those of the C. arctalus. Cope, and
the roots are similarly compressed. An important dift'erence in the teeth is
seen in their lack of the carina on the enamel-face of the base of the crown at
least ; the apices being in (sach case broken away. Accordingly, the root
exhibits no corresponding groove on its inner side, as is the case in C. arcta-
tus. The form of the maxillary lione is also characteristic. In C. arctatus,
this piece bears a longitudinal protuberance on its inner side so as to have
given the face great proportionate width. In C. stenopsh, tiiis j)rotuberance
is much less pronounced; the inner face, instead of being nearly horizontal
above, is curved abruptly downward, and a shallow horizontal face of no
great width replaces the wide oblique cornice which extends from the alveolar
border in C. arctatus. The remains indicate a species of the size of C. arc-
tatus.

Measurements.

M.

Depth of the maxillary bono on n dental gioovi' 0. 0350

Diameter of the maxillary bone external to alveoli 0.0200

Diameter of a tooth 0. 0045

POLYONAX, Cope.

Char. gen. — A species considerably larger than tin; last, n;presented
by vertebra3 and numerous fragments of linil)-bones. The most characteris-
tic of the former are two, probably from the posterior dorsal region, whicli
are somewhat distorted by pressure. The more anterior is shorter than the
other, and exhibits both articular taces slightly concave; the one more so than
the other. They are higher than wide, and the border is scolloped above tor
the capitular articulation for the rib. There are num(n-ous nutritious fora-
mina, and some ligamentous pits on the articular surtiuies. The inferior
face is rounded. In the longer vertebra, both faces are more strongly con-

64

cave, and at each end of the lower side there is an obtuse hypapophysial
tuberosity. The sides of the centra of both vertebrae are concave. The
neural canals are relatively small, and the neurapophyses co-ossified. A third
vertel)ra, without arches, is similar in specific gravity, though without the
white surtace-layer of the others. It is appropriate in size and form to this
species, and is peculiar in its flat form, resembling the anterior dorsals of Ha-
drosmirus. In this respect, it is related to the shorter vertebra of the two
above described as the latter is to the longer. The surface of the posterior
articular face is damaged. It was not concave, and is now slightly convex.
The anterior is preserved, and is concave.

POLYONAX MORTUARIUS, Cope.

The articular faces are deeper than wide in the vertebrae ; the sides are
smooth ; the lower face is narrowed and probably keeled.

Measiiremtnts.

M.

Anterior dorsal :

Length of the ceutrnm 0. 048

Elevation to the neural canal 0. 094

Width 0.004

Median dorsal:

Length of tlic ccutriun 0. 0^<7

Elevation to the neural canal 0. 117

Width 0.08:i

Posterior dorsal :

Length of the- centrum 0. 00:i

Elevation 0.104

Width 0.08:!

Diameter of tlie neural canal 0. 010

The measurement of the neural canal is made near the base of the neura-
pophyses, and is probably a little affected by pressure.

The linil)-bones cinl)raco portions of tibia, fibula, and some others not
yet determined. The portion of tibia is Irom the base of the cnemial crest,
so tliat one extremity is trilobate, the other transverse-oval. The former
outline indicates two posterior tuberosities. The bone is solid, and the
superficial layer for 3""" or less, is so dense and glistening as to resemble
ccmcntum. Portions referred to fibulfB have a subcrescentic section, with
narrowed width in one direction. Two fragments of shafts of long bones I
cannot determine, l)iit they may lielong to the pelvis. Tliey belong to opposite
sides; each is oval in section, and the diameter regularly contracts to one
end. One side is slightly convex in both directions ; the other is less con-

65

vex transversely and gently convex longitiulinally. A peculiarity consists of
a central cavity present in both at the fractured large end, which is bordered
by a layer of dense bone like the outside.

Measurements.

M.

Transverse diameter of the tibia-fiagincnt below the cuemial crest 0. 125

Autero-postcrior diameter of the tibiafragmeut at the base of the crest 0.095

Width of the fragment of the iibiila 0.073

Thickness of the fragment of the fibula 0.035 ^

Length of the fragment of an unknown pelvic bone 0. 145

Proximal diameterof the fragment of the unknown bone 0.088

Distal diameter of the fragment of the unknown bone 0.065

The above measurements indicate a much larger animal than the Ciono-
don arctatus, and are not very different in size from the Lcalaps aquilunguis.

ORNITHOSAURIA.

As compared with the European Cretaceous, the corresponding beds in
North America have yielded but few species. These are of the largest size
to which species of the order are known to attain. To Professor Owen and
H. von Meyer, science is indebted for the first explanations of the structure
of these remarkable reptiles ; while Prof H. G. Seeley, of Cambridge, En-
gland, has added greatly to the work commenced by the elder authors, and
also extended our knowledge of the genera and species.

PTERODACTYLUS, Cuv.

Pterodactylus umbeosus. Cope.

Represented, by, the distal portion (ten inches) of the wing-finger meta-
carjjal ; the proximal portion (eight inches) of the first phalange of the same
digit, with a lateral carpal, and two phalanges of claw-bearing digits. I'he
distal condyles of the first-named bone are separated by the usual deep groove
above and below, and wind spirally to their terminations on the inferior face.
The narrow base which supports the inner condyle is bounded posteriorly by
an acute edge ; directly outside of the base of this ridge is a deep groove or
foramen, which is bounded next the external condyle by another ridge, which
rises to the base of the inner condyle on the trochlear side. The transverse
diameter of the condyles is O^.OiS, or 17 lines.

The proximal end of the first plialange is perfect, but flattened by press-
ure. It presents the two usual cotyloid cavities well separated by an ele-
9 c

m

vated ridge. Anteriorly, it presents an elevated crest for muscular insertion.
This terminates abruptly, and is followed distally by a deep notch. Distal to
this is another prominence of the bone, also probably an insertion. Antero-
posterior diameter (flattened), 24 lines.

The lateral carpal is short and wide ; both its articulations are simple
and concave. Both outlines are keeled; one very strongly at one end, and at
the other presenting beyond the articular surface for the distal carpal, a wide
prolonged process for muscular insertion. Length of carpal without proc-
ess, 13 lines; process, 4 lines; diameter, widest extremity, 11 lines. This
indicates a very stout carpus. The phalange is penultimate, and is remark-
able tor its small size, perhaps indicating an external rudimental digit. It
is only supposed to belong to the anterior Hmb from its having been found
with the preceding bones. It is slender, and has a convex distal articulation,
divided by a trochlear groove, and the concave proximal one in like mannei
divided by a trochlear carina. Length, 9 lines ; proximal depth, 3 lines.

'1 his species is the largest Pterodactyle as yet known from our continent;
the end of the wing-metacarpal exceeding in diameter that of the species
described by Professor Marsh from the same region by more than four milli-
meters.

From near Butte Creek, Kansas, from the yellow chalk.

Pterodactylus occidentalis. Marsh.

Established on wing-metacarpals and phalanges of three individuals.
The articular extremities indicate a species from one-half to two-thirds the
size of the last-named. Those of the metacarpal are very prominent above
as well as below, and there is no distinct ridge in the trochlear groove
between them. The inner condyle does not stand on a base with an acute
posterior ridge, but overhangs a rather obtusely-edged support. There is no
second ridge on the outer (trochlear) side of it. The same condyle termi-
nates abruptly posteriorly on the superior face of the shaft. Width of con-
dyle in No. 1, 11 lines; in No. 2, 13 lines; vertical diameter, inner condyle.
No. 1, 11 lines; transverse diameter of the shaft above, 8 lines.

The proximal articular surfaces of the proximal wing-phalanges are
deeply concave: the inner protected by an elevated margin behind; that of
the outer much lower. They are separated chiefly by a deep emargination
l)ut on their short adjacent portions by a low ridge. The process for liga-

07

mentoiis insertion is well developed. The distal extremity is slightly
widened, and its articular surface is wedge-shaped with very convex base.
Its surface is slightly concave in both directions, and without median ridge.
The margin of the shaft terminates in a short tuberosity bearing articular
surface. Transverse diameter, 16 lines; length of shaft preserved, but in-
complete, 9 inches 1 line.

This is possibly the species originally described by Marsh as P. owenii,
a name which could not be used on account of its pre-occupation for another
species from England. It was described by the writer in 1872 under the
name of P. harpyia; but, a fire occurring in the establishment printing the
paper, its publication was delayed until two days atler Professor Marsh had
republished his species as P. occidentalis.

The type-specimens of the two species described were procured from
the bluifs of Butte Creek south of Fort Wallace, Western Kansas.

Order I.— CROCODILIA.

Only two species of this order are known from the Cretaceous formations
west of the Missouri River : the one from the lead-colored stratum of the
Benton epoch, or No. 2, in Kansas; the other from the Fort Union group, or
No. 6, of Colorado. The latter is a short-headed species, allied to, and near
the size of, the alligator of Louisiana. The former is a Hyposaurus, of eight
or ten feet in length, found in digging a well at Brookville, and presented to me
by my friend Dr. William E. Webb, of Topeka. The individual discovered
was not fully grown, but indicates a smaller and stouter crocodile than the H.
rogersii, Owen, of the New Jersey greensand. This genus belongs to the
group witili sul)biconcave vertebrae, and had a long, subcylindric snout.

HYPOSAURUS, Owen.

Hyposaurus vebbif. Cope.

An anterior cervical vertebra presents the following characteristics. It
is that one in which the parapophysis occupies a position opposite the lower
third of the vertical diameter. Its centrum is stout in form; the articular faces
but little concave; the posterior a little more so than the anterior. The ante-
rior is almost regularly hexagonal; the posterior subround, a little deeper than
wide. The inferior surface po.'sses.ses a strong, obtuse, median carina, wiiich

68

disappears in front of the posterior margin. Anteriorly, it terminates in a
short, obtuse hypapophysis. The suture of the neural arch is very coarse.
Surface of the bone smooth.

Measurements.

M.

Length of the centrum - - 0. 037

Diameter of the centrum anteriorly :

Vertical 0.032

Horlzoutal 0.031

Diameter of the ceutrum posteriorly :

Vertical 0.033

Horizontal 0.031

Length ot the surface of the parapophysis 0.015

As compared with the H. rogersii of the New Jersey Cretaceous, this
vertebra is shorter and stouter, and the extremities less concave ; the suture
for the neural spine is much coarser.

This crocodile was discovered in a bluish stratum, encountered in digging
a well in Brookville, Kansas.^ This point is considerably east of the, expos-
ure of Cretaceous rocks seen near Forts Hays and Wallace. It was the first
of the Crocodilia found between the Tertiaries of the Rocky Mountains and
the Cretaceous of New Jersey.

It was given me by my friend Dr. William E. Webb, of Topcka, to
whom science is also indebted for the Polycotylus latijr'mnis. I have dedicated
this species to him.

BOTTOSAURUS, Agass.

Cope, Proceedings of the American Philosophical Society for 1871, 48.
BoTTOSAURUS PERRUGOSUS, Cope.

Represented by numerous fragments, with vertebrae and portions of skull,
which accompanied the dinosaurian and turtle remains from Eastern Colorado,
already alluded to.

A portion of the left dentary-bone, containing alveoli for ten teeth, shows
that this species iS' not a gavial. The dental series passes in a curve from the
iimer to the outer sides of the bone; one or two alveoli behind being probably
bounded on the inner side by the splenial only, as in B. macrorkynchus, when
that bone is in place. The dentary is compressed at this point ; in front, it
is depressed. TheTe is a slight difference in the sizes of the alveoli, but not
such as is usual in Tertiary crocodiles. Tlie external face of the bone exhib-

' This stratum is similar to that iu which Dr. Hayden found the fish Apaopclix aaiiriformis a'i
Hunker Hill.

69

its deep pits in longitudinal lines. The angle of the mandible is depressed.
The cotylus of articulation is partially concealed on the bulerside by the ele-
vation of the surangular, whose upper border is parallel with the inferior
margin of the ramus for two inches to where it is broken off. Tiic outer fiice
of this region is marked by irregular coarse ridges, more or less inosculating,
separated by deep pits. The lower posterior half of the angular bone is
smooth.

A posterior dorsal or lumbar vertebra has a depressed cordate articular
cup. The zygapophyses are large and widely spread, and strengthened by
obtuse ridges running from the base of the neural spine to the posterior
margin of the anterior, and the posterior outer angle of the posterior. One
pit at basis of neural spine in front; two before. Ball prominent: sides of

centrum concave.

Measurements.

ii.

Lergth of a fragment of the ramus 0. lUO

Width iufiout 0.034

Depth Uchiud 0.0:52

Length of tight alveoli — 0. 0(ii)

Diameter of the largest alveolus 0.012

Diameter of the smallest 0. 007

Width of the base of the angle of the ramus 0.04a

Depth at surangular 0. O^il

Length of the centrum of the vertebra 0.045

Width of the articular cup 0.031

Vertical diameter of the articular cup 0.025

Vertical diameter of the neural arch 0. Oil

Expanse of the anterior zyg.apophyses 0. OiJG

The specimen is adult, and indicates an animal about the size of the alli-
gator of the Southern States. Its reference to the present genus is provisional
only.

SAUROPTERYGIA. —

As compared with the European and New Zealand Cretaceous beds,
those of our country have yielded ])ut few remains of reptiles of this order.
Four only are embraced in the present work, three of which are from No. 3,
and one from No. 6.

The structure of the scapular arch in Pulycotylus is yet unknown ; in
Elasmosaurus, it is quite distinct from that of Pleslosuurus, so much so as to
have induced me to regard it as type of a distinct family, the Elasmosauridce,
characterised by the absence of di.stinct mesosternal bone. Professor Seeley
has since more fully defined this group, and has discovered several species of

70

it. ill tlie Eiiglisli formations, some ot" which he reil'is to tlirce additional
genera, under the names of Cijmhomurus, Erclmusaurus, and Murcenosaurus.
Allied species, including a Polycotylus, have also been described by Mr. Hector,
from corresponding strata in New Zealand.

POLYCOTYLUS, Cope.

This genus is establislied on a series of vertebrae, witli portions of pelvic
arch and posterior extremity, discovered in the upper Cretaceous of Kansas
by W. E. Webb, superintendent of the land-office in Topeka, Kansas. The
point at which the remains were found is about five miles west of Fort
Wallace, on the plains near Smoky Hill River, Kansas, in a yellow Cretaceous
limestone.

There are wholes or portions of twenty-one vertebrse, of which but two
retain their neural arches, and six are represented by neural arches only. Four
centra may be referred to the caudal series, the remainder to the dorsal ; only
two indicate the characters of the cervical vertebrae. All of these verteljrse,
except the distal caudals, are remarkable for their short antero-posterior diame-
ter, and deeply concave, articular faces. This concavity is not, however, of
an open, conic form, as in Ichthyosaurus, but is flattened at the fundus, thus
exhibiting a small, slightly disciform area. The usual pair of venous foramina
appears on the under side of the centrum. The neural arch is continuous
■with the latter, and exhibits no trace of connecting suture. The diapophyses
arise from the neural arch in all the dorsals; they are compressed and verti-
cal in section. The arch is, of course, narrow antero-posteriorly, and presents
a pair of moderately prominent zygapophyses in each direction; the posterior,
as usual, articulating downward, the anterior upward. On some of the verte-
brae they become closely approximated. The neural spines arc narrow antero-
posteriorly, and much stouter transversely than in Elasmosaurus ; they are
strongly grooved at the base, both anteriorly, and most so posteriorly.

The caudal vertebrae are anteriorly quite as large as the dorsals. Two
anterior caudals present, on the latero-inferior part of the posterior margin,
a pair of widely-se^jarated articular surfiices for chevron-bones. A portion of
one of the latter remains ; it is narrow and subcylindric at the base. The
diapophyses are situated on the upper part of tlie centrum, and are continuous
with it, and without trace of suture. There are two distal cervicals, which
are much smaller than the preceding. They are solidly coossified, and have

71

been Ijiokeii i'nnn one anterior to them, with which they have been also anchy-
losed. Processes in the position of the diapophyses have disappeared, while
a strong infero-lateral process projects from the middle of each, similar in
position to the parapophyses (or whatever they may be) of the Elasmosnurus.
These processes are decurved, and much thickened and rugose; they may be
described as more or less elongate-conic. The neural canal of these vertebrae
is well-marked, though small. The coossitication of cervical vertebra? is a
remarkable character, and very unusual. It does not seem probable that these
specimens re])resent a diseased condition, since they are symmetrical, and the
inferior surfaces and foramina are unaffected. The rugosity is much that of
a ligamentous articulation. Their size indicates a remarkably slender neck,
even more so than in Pleswsaurus.

That the portions of an extremity alluded to belong to the posterior is
rendered probable by the presence of part of an ilium, and by the fact that
the portions of the vertebral column secured are chiefly median and posterior.
The fragments consist of the extremity of the femur, the tibia, several tarsal
bones, and numerous phalanges. The whole limb is of great size compared
with the vertebral column, and indicates powerful natatory capacity in its
possessor. What the relative length of the femur may be cannot be ascer-
tained, as the proximal portion is wanting ; but, if it were like the tibia, it was
characterized by stoutness rather than by length. The portion remaining is
flattened, and presents distally two distinct articular faces for ulna and radius,
instead of the uniformly convex outline characteristic of most of the species
of Plesiosaurus. The tibia is pentagonal, broader than long, and emarginate
externally. The fibula is not preserved. One of the tarsal bones is a flat,
unequally hexagonal disk, of less thickness than the tibia and the tarsals,
which appear with three faces of broad, plane articulations, and the outer edge
rounded in section. Another tarsal or metatarsal is a parallelopipedon, except
that one extremity presents two faces meeting at a right angle. Another is
similar, but oblirpie, i. e., rhombic in section; one of the longitudinal angles
is also prolonged.

Of the phalanges, there are individuals from three series. Portions of
ilat ixMies, perhaps belonging to the pelvic arch, indicate, as do all tlie other
pieces, that tlie bony structure in Polycotylus is more nuissive than in Elas-
mosaurus, if the only known species has not attained such huge dimensions as

72

some of the latter. These fragments do not throw much light on the
structure of the pelvic arch.

The structure of the bones is, like that in the order generally, of the
coarsest description. There are no medullary cavities, but the cancellous cells
are large, and extend everywhere in the direction of the axis- of each bone

The characters which separate this genus from riesiosaurus may be
derived from the preceding, as follows :

First. The deeply biconcave and very short vertebral centra.

Second. The tibia broader than long, resembling that of Ichthyosaurus.

Third. The coalescence and depression of the cervicals.

Fourth. The continuity of the neural arches.

Fifth. The continuity of the diapophyses of the caudals with the centra.

The only genus with which this one may be here compared is the Thau-
matosaurus of Meyer. This is known by but a few fragments, and of these
but few are present in the Kansas animal. The character on which I rely at
present to distinguish them is the much less concavity of the dorsal vertebras
in Thauniutosaurus. This is, however, not entirely satisfactory. T. oillthicus,
Meyer, is from the Lower Oolite of -South Germany.

The bones of the specimens of Pohjcotylus are thoroughly mineralized,
and the adherent matrix is a light-yellow chalky limestone, similar to that
which yielded the fine fragments of the Liodon proriger.

POLYCOTYLUS LATIPINNIS, Cope.

The anterior dorsal vertebrae have the centra slightly compressed or ver-
tically oval, while the posterior are more rounded. The anterior caudals
appear to have been round, or nearly so ; they are somewhat distorted by
pressure. The sides of the centrum are slightly concave in the longitudinal
direction ; below, there is no carina, but at least two venous foramina. There
is another large foramen on the side of the centrum, usually not far from the
neural arch ; there are usually other smaller foramina below tliis. The bases
of the diapophyses are longitudinally grooved behind, and separate a concavity
of the arch in front of them from one behind. In the most nearly median, the
most elevated diapophyses stand about equally on the neurapophysis and the
neural spine above it. The diapophyses are vertically compressed, and the cos-
tal articulation of the only one preserved is in the same plane. The margins of
the external surfaces are not coarsely striate, as in many Sauropterygla. The

73

venous foramina of the distal co-ossified cervicals are in \nnvs, and of large
size. In the proximal caudals, the diapophyses are aijove the middle of the
sides of the centra. In one, the basis of a chevron is preserved. It is
cylindric and striate. The zygapophysis on the hinder aspect of a dorsal has
a disciform articular surface directed outward and downward ; tlie prominence
of its upper face is continuous with the lateral ridge of the neural spine.
The anterior up-looking surface is equally small and little divergent.

Measurements.

iDChus.

Length of two coossifieil cervicals - • 2.5

Width of the anterior in front 1.7

Depth of the anterior in front 0.9

Vertical diameter of the centrum of the dorsal 3. 42

Transverse diameter of the centrum of the dorsal 2.7

Antero-posterior diameter of the centrum of the dorsal (below) 1.K5

Vertical diameter of the centrum of the dorsal (posterior) 2.98

Transverse diameter of the centrum of the dorsal 2.9

Transverse diameter of the neural canal 0.86

Lougitudiu.il diameter of the base of the neural spine 1.22

Longitudinal diameter of the laso of the diapophysis 1.2

Length between the extremities of the zygapophyses (dorsal) 2.26

Depth of the cup of the vertebra) 0.63

Length of the centrum of the anterior caudal I.7.!

Distance between the bases of the chevron-bone (caudaf) 2. TiS

It may be observed that the anterior caudals have a nearly round articu-
lar extremity ; one of tliem is a little wider than high/ but they are too much
distorted to furnish reliable measurements.

The portion of tlie ilium preserved is an extremity. It is flat on one side
and convex on the other. The shaft is solid. The articular extremity is
oblique in one direction and truncate in the other, which is at right angles to
a short, recurved margin, which has been an insertion or articulation ; the flat
surface is rugose distally. Long diameter of extremity, 2.75 inches ; of shaft,
1.9 inches. The articular faces of the ext-cemity of the femur are at an open
angle with each other, and are strongly concave in transverse section. The
femur is here very flat, with narrow margins ; it becomes stouter with dimin-
ishing width. Distally, the surface is marked by grooves and small foramina.
What may be tibia is the basal frustum of a wedge ; t.he articular faces arc
broad; the outer margin narrowed ; the faces slightly concave. The inner
margin is shorter than the outer, and the distal part of it presents a broad
articular face. Some of the tarsal bones have been already described. There
are thirteen metatarsals and phalanges. They arc of stout proportions and
10 c

74

•am considerably constricted medially. Those oi' one scries are cqiiare in

section ; those of another, transverse ; those of a third, transverse, with one

edge thinned or acuminate in section. Some of each form are more elongate

than others.

Measurements.

lucbes.

Width of the femur at the extremity (restored) H.

Depth of the femur at the extremity (median) 1.3

Width of the femur four inches from the extremity C.

Thickness of the femur four inches from the extremity 1.95

Width of the tibia 3.88

Length of the tibia externally '^-G

Width of the tarsi (tibial) 2.46

Thickness of the tarsi (tibial) 1.52

Length of the parallelepiped phalange 1.56

Width of the parallelopiptd phalange 1.2

Thickness of the parallelepiped phalange 1.2

Thickness of the depressed phalange 1-

Width of the depressed phalange 1-4

Length of t ho depressed phalange 1 • 'J

These powerful extremital pieces indicate a body to be propelled of not
less than usual proportions. If this be the case, the number of dorsal verte-
brjE is considerably greater than in the species of this order in general, and
approaching more the Ichthyosauri. I 'do not intend to suggest any affinity
between the latter and the present genus, as none exists. What the extent
of the cervical vertebrfe may have been is uncertain. The caudals have prob-
ably been numerous, though not probably so extended as in Efasmomurus.

The size of the species may be approximately estimated from the pro-
portions furnished by Owen (Reptiles of the Liassic Formation) in Plesio-
saurus rostratus. The skeleton of this species measures 11 feet 8 inches in
length, and the dorsal vertebrfE are of less vertical and equal transverse
diameter compared with those of the present saurian. V/c may, therefore,
suppose that the latter exceeded the former in dimensions.

William E. Webb, of Topeka, discovered the specimens from which this
species was first described, and liberally forwarded them to me for examina-
tion and description Other specimens have been discovered since that time
by various other persons. I have received numerous fragments of an individ-
ual of about the size of the one above described, which were found by Prof
B. F. Mudge, at a point near the mouth of the north branch of the Smoky

Hill River.

These consist ol' a f(;w vertebra;, portions of pelvic and scapular arches,
and three proximal bones ol' the limbs. Which of these is fcnuir and which

7o

humerus, I am unable to delcrmiiic, owing to their close resemblance. The
vertebra; do not (liUcr i'rom tliose of the specimen just described. The limb-
bones are stout and expanded, and thinned distally ; this thinning is remark-
able, and indicates a much flattened metapodial region. The head is slightly
expanded, the articular face being turned obliquely to the inner face of the
shaft ; the surface is pitted for attachment of the jirticular cartilage ; two-
fiflhs the length from the proximal end is an extensive and exceedingly rugose
surface, as wide as the shaft, for the insertion of the adductor-muscles.

Measurements.

M.

Diameter of the ceutrum of tho lumbar vertebra 0.08

Leugtb of tbe (?) humerus 0. 4.'>

Diameter of tbe be.ad 0. liS

Diameter of tbe sbaft 0. 098

Diameter of the distal eud (transverse), restored in part 0. 18

Should the humerus have been related to the fore-limb, as in Flesio-
saurus dilichodirus, Conyb., the latter would have had a length of 4 feet
3 inches ; as the proportions of the radius and phalanges are shorter, the
limb was probably relatively shorter. If related to the total length, as in the
same Plesio.saur the humerus would indicate a length of \lh feet. The cer-
vical vertebrae become attenuated, as compared with the dorsals, to a greater
degree in Polycotylux than in Flesiosaurus.

ELASMOSAURUS, Cope.

This genus has been more completely preserved to us than any oilier
American representative of the order. In the interpretation, however, con-
siderable care is necessary, as the form appears, at first sight, to reverse, to
a remarkable degree, the usual proportions of known reptiles. The scapular
arch, in the absence of the episternum, presents the same number of ele-
ments as the pelvic, and is not without resemblance to the latter, as it
exists in some species of the order. The fortunate preservation of the series
of cervical vertebraj shows this to have been, in the typical species, three
times the lenglii of the body; much exceeding in this disproportion that
knowu to exist in other species of the order.

The neural arches are everywhere continuous with the centra, witiiout
sign of suture, and are externally plane. The nenial canal is exceedingly
small for the size of tlic vertebra;, especially on the lumbar and caudal verte-
br;e.

76

The dorsal vertebra arc remarkable from tlie fact that the diapophyscs
disappear on the anterior part of the series, and gradually diminish in length
from behind forward to the point of disappearance. On the median and
posterior parts of the series, they are very elongate, and rise for a short dis-
tance from the basis of the neural arch. Anteriorly, they descend and shorten,
and finally remain only as the slightly-elevated borders of rib-pits. Through-
out the whole of the anterior portion of the column to the cervicals, the neural
spines are of great elevation, and of such antero-posterior extent as to be
nearly continuous.

The cervical vertebrm are not only more numerous, but become anteri-
orly much smaller and more attenuated than in its allies of the same family.
They are remarkably compressed, the centra much longer than deep, and
deeper than wide, and with smooth concave sides.

The ribs of the anterior cervico-dorsal region are inserted directly in the
vertically-oval pits of the centrum. Immediately at the point where these
cease, thin traverse processes appear to arise from the lower edges of the rib-
pits. They form a continuous series with the ribs, and soon rise from the
plane of the lower face of the centrum, and are directed obliquely downward.
At the end of the cervical series, they are directed nearly vertically down-
ward. The number of these vertebrae is very great; the anterior diminishing
to a very small size ; the whole measuring a little more than half the total
length. Most of the cervicals possess two venous foramina below, the dorsals
two, and most of the caudals one.

The resemblance of the caudals to the usual type of Plesiosaurus is seen
in the fact that each bears near its posterior articular aspect, on the inferior
face, a pair of articular surfaces for chevron-bones. Similar vertebrae had
been described by Leidy as the caudals of a genus he called Discosaurus.
The study of the present genus shows that they are really of the caudals of the
allied genus Chnoliasaurus, the support caudals of the latter being the cervicals.

The ribs are simple-headed ; the abdominal ribs seen in Plesiosaurus are
possibly wanting, as none were found by the discoverer of the fossil, after a
careful search.

The end of the muzzle, with symphysis mandibuli, was preserved. This
is flat, the symphysis co-osified and rather short, tlie premaxillary grooved at
the intervals between the dental alveoli. The teeth are deeply implante<I,
with small pulp-cavity, are cylindric, and furnished with nearly straight elon-

77

gate conic crowns, which arc minutely but sharply striate to the tip; the
ridges straight, continuous. There are no indications of nostrils, so that these
were probably posterior, and near the orbits, as in Plesiosaurus.

The pelvic arch is more extended than the scaj)ular, and strongly re-
sembles the pelvic arch of the Fksiosauridcc. The scapular arch is peculiar ;
the claviculi are broad, flat bones, resembling the pubes of certain tortoises,
while the coracoids are much like the coracoids of Plesiosaurus. The clav-
icles have a greater transverse extent than the latter, and have a very exten-
sive line of union medially, and a narrow posterior prolongation, which meets
a similar anterior one of the coracoids, separating the intervening foramina.
They appear to form about one-third of the walls of the glenoid cavity, and
have a constricted base, as in some riesiosauria, applied to the extremity of
the coracoid. The form of the glenoid cavity cannot be readily ascertained
from the absence of the scapula. What we have of it would suggest the ex-
istence of a fore-limb, of comparatively little power, though no remains of such
have been found. The acetabulum is smaller than the glenoid cavity ; this
point, with the obvious source of propulsive power in the tail, indicates that
the hind limbs were smaller than the fore. There is no trace of sacrum, nor
of any modified diapophyses for support of an ilium.

The ischia are flat, subtriangular bones, with a long median line of junc-
tion, and communicating anteriorly with the posterior prolongation of the
pubic plate. Their postero-exterior margins project well backward. The
pubes are broad plates, whose anterior margins diverge from each other.
They are broader than the ischia, and form a broad, shallow basin for the
support of the viscera. The suture defining these elements is obliterated ;
they are continuous, and form a weak, inferior keel on the median line. A
simple curved ilium has been preserved, for which there appears to be a
smooth articular surface on the pubis, to which it was attached.

The acetabular portions of these elements are flattened and furnished
with convex articular surfiiccs. The supposed ilia are short, curved bones,
resembling that of Plesiosaurus latispinus, Ow., or of some of the other spe-
cies of that family. The shank is flattened cylindric ; the distal extremity
dilated, rounded, and flattened ; the proximal extremity subtruncate, or trun-
cate in two or three unequal planes, and witli a median pit. It fits well
when applied to a concavity on the articular surface of the pul)is. Tiie ver-

78

tebra? above the pelvic arch were furnished with elongate, subcylindric dia-
pophyses.

Dr. Turner, the discoverer of the original specimen, having made a sec-
ond careful search and renewed excavations at the original locality, failed to
find any bones which can be assigned to humerus, ulna, radius, carpus, or
phalanges, or similar elements of the hind limbs. The pelvic and scapular
arches were further completed, and an additional number of ribs obtained.
The glenoid cavities are rather angular, and both were filled with solid
argillaceous matrix. The acetabula are not cuplikc, but merely exposures
of the narrow plane extremities of the pubes and ischia ; they were covered
with thin layers of gypsum ; the pieces of the ilia were found imbedded in
the mass of matrix which occupied the pelvic arch.

This genus is well distinguished from Plesiosaurus by the peculiarity of
the scapular arch. The mesosternum appears to be wanting, and the clav-
icles and coracoids form a breastplate. If the claviculus was ever united
with the scapula, as in Plesiosaurus, no evidence of it can be seen in the speci-
men ; it is also broader and more extended anteriorly.

Differences from other Sauropterygia. — The only genus which it is nec-
essary to compare with the present one is Cimoliasaurus. The following
may be noted as generic distinctions : the series of cervicals rapidly diminishes
in Cimoliasaurus in absolute size and in relative length of the vertebrae, which
are not compressed. In the present genus, they maintain a similar and in-
creased length for a considerable distance, diminish in length very gradually,
and are much compressed. The diapophyses erf the dorsal vertebrae, as they
descend in Cimoliasaurus, continue well developed until they attain the infe-
rior planes of tlie centra, and have there a downward direction. In Elasmo-
saurus they cease while yet on the middle of the centrum, and are replaced
by pits throughout the remainder of the length.

The neufal canal is everywhere markedly larger in Cimoliasaurus.

The American genera of Elasmosauridce may be compared as follows:

Posterior cervical vertebrae without diapophyses :

Cervicals long, compressed; neck very elongate Elasmosaurus.

Posterior cervical vertebrae with diapophyses :

Cervicals cpiadrate, short, depressed, transverse, rapidly

diminishing in size, hence the neck short Cimoliasaurus.

79

Professor Owen figures and describes (Reptiles of the Cretaceous, Palc-
ontographical Society) a vertel)ra whicli very closely resembles the cervical
6i Elasmosaurus. He considers it to be the cervical of a peculiar Plesiosau-
nis, which he calls 1\ constrictus, remarking, at the same time, its remark-
ably inferior plcurapophyses.

Elasmosaurus platyurus, Cope.

This, after Mosasaurus, the most elongate of the sea-sanrians yet discov-
ered, is represented by a more than usually complete skeleton in the museum
of the Academy of Natural Sciences of Phila'delphia. It was found by Dr.
Theophilus H. Turner, the physician of the garrison at Fort Wallace, a point
situated near the boundary-line between Kansas and Colorado, and a short
distance north from the Smoky Hill Fork of the Kansas River. Portions of
two vertebrae, presented by him to Dr. Leconte when on his geological tour
in the interest of the United States Pacific Railroad Company, were brought
by the latter gentleman to Philadelphia, and indicated to the writer the exist-
ence of an unknown plesiosauroid reptile. Subsequent correspondence with
Dr. Turner resulted in his employing a number of men, who engaged in exca-
vations, and succeeded in obtaining a large part of the monster. Its vertebrae
were found to be almost continuous, except a vacancy of some four feet in
the anterior dorsal region. They formed a curved line, a considerable part
of whose convexity was visible on the escarpment of a bluif of clay-shale rock,
with seams and crystals of gypsum. The bones were all coated with a thin
layer of gypsum, and, in some places, their dense layer had been destroyed by
conversion into sulphate of lime.

The scapular arch was tpiind in large part adhering to the bodies and
neural spines of a series of the anterior dorsal vertebras, and was detached
from it at the academy. The pelvic arch had been slightly crushed, and the
lumbo-sacral vertebrae forced into contact with the ischia, where they remain.
A broken extremity of the sup|)osed ilium was forced into the matrix which
supports the ischia. Many of the dorsal and caudal vertebrae were sent, and
remain in continuous masses, so that the succession is readily traced, and the
true relations of the extremities preserved.

In removing the matrix from beneath the vertebrae, scales and teeth of
some six species of Physoe/ysf and Physostomus fishes were found, including
an Enchodus and a Pha,sgamrfits ; the latter indicating a new species, which I

80

have called P. car'matus. These aiiiinals had doubtless been the food of the
Elasmosaurus. '

The end of the muzzle was broken from a part or whole of the cranium,
which has not been rediscovered, though Dr. Turner has made careful search.
It was found in front of the vertebrae, here regarded as cervical, at some dis-
tance from them.

The whole skeleton has been under considerable pressure, so that most
of the ribs have been pressed flat on the vertebrae ; the long parapophyses of
the cervicals have most of them been fractured at their bases and compressed,
those of opposite sides thus approaching more nearly in the form of chevron-
bones than they otherwise would have done. The proximal cervicals are
ol>liquely flattened by the pressure ; the other cervicals have the bodies natu-
rally flat, with the articular surfaces much less so than the median portion.
Some of the caudals are obliquely distorted. •

DescrijJtion. — Vertebrce. — The neck may be safely assumed as a point of
departure, as it consists of above sixty mostly continuous vertebrae, which
graduate to an atlas of very slender proportions. Most of them preserve more
or less developed parapophyses. At the posterior extremity of this series
sixteen are perfectly continuous, and in this portion a great gradation in form
is apparent. The anterior are narrow, compressed, and similar to the more
distal cervicals in the elevated position of the lateral angle; the anterior are
subquadrate, thick, and with lower lateral rib, and stronger (f) pleurapophy-
sis. In these respects, the latter resemble the dorsals which follow toward
what I beHeve to be the tail. Four anterior dorsals are in one mass (figured
in Plate 3) ; in this series, the lateral angle first approaching is finally lost in
the margin of the rib-pit, the posterior thus resembling other dorsals.

In a series of four anterior dorsals, which, like the preceding, are in
their original continuous mass, those of one extremity have centra rounded in
section, with inferior rib-pits ; those of the other have quadrate centra and
elevated diapophyses; the former have the character of the first dorsals, the
latter of the median dorsals. The posterior dorsals and anterior caudals form,
in like manner, a continuous series of eleven vertebrae, fractured in four places.
In them, the diapophyses steadily descend, reaching the inferior plane in the
last; thus, with the reduction of the venous foramina to one at the seventh,
indicating the point of transition from dorsal to caudal series. The zygapo-
physes preserve the usual arrangement, but are much compressed, so that the

81

posterior or dowii-looking arc conHneul, and scarcfly sciuuatcd In an cmar-
gi nation.

The neural spines, at their bases, have a slight posterior ol)liquity ; and tlie
superior portions lean strongly in the anterior direction. The inferior liml)s
of the cervical pleurapophyscs appear to be entirely wanting. TIk; articular
faces for the chevron-bones are seen at the extremity of the interior rib of the
caudal.

Of the cervicals, there are both axis and atlas. Of the caudal series,
probably the distal half, at least, is lost. A single vertebra near the middle
does not relate to either of those anterior or posterior to it. There ai'c, therc-
ibre, at least four lost from that region also.

There is a considerable interruption immediately anterior to the last
dorsal vertebra. Three large vertebrae, with long diapophyses belonging
here, were embedded in the hard matrix which protected the pelvic arch.
These are far from relating immediately to the vertebrae preserved before and
behind them. I estimate the number missing as follows: Seven of the four-
teen dorsals preserved have more or less elongate diapophyses. In Flexio-
auri, vertebrae of this character are much more numerous; in P. homalospon-
dijlus, Owen gives seventeen. If we add ten to the series in the present
species, it will give the abdominal space between the adjacent margins of
the oo. pubis and coracoidea an extent equal to the length of the pelvic arch.
This is rehitively shorter than in the Plesiosauri. Dr. Turner found that a
space of " three or tour" feet intervened between the two portions of the
skeleton, which was otherwise continuous. I think ten an average number
to represent safely the missing dorsals.

From the cervical proximal regions, probably three vertebrae are missing
from two interruptions. The remainder of the cervical series exiiibits three
interruptions. Most of the proximals have been broken medially, leaving
the articulations solid, an advantage in determining their continuity. Three
vertebra' and one-half are thus found missing in this region.

The whole number of vertebrae preserved and Inst, with the relative
lengths of each, may be stated as follows :
11 c

82

I'li-sciit.

Lost .

Total

No.

Lenglli.

No.

Length.

l<,-ngtb.s.

CraDium . ..

TtichiK.

Inches.
24.
22.3
37.0
CO.

Inches.
24.

Ccrvicals . . -. ..

(W-1

M

21

257..'-)
5r>. 10
00.4

3i
10
30

27U.t^

Dorso-lumbars

'.12. 10
120. 4

Total

lo:u

43i

;->!(;. 30

This gives the totnl length to the animal of 43 feet 2 inches, wiiich, in-
creased by the amount taken up by intervertebral cartilages, will give roundly
about forty-five feet. Of this, twenty-two must be reckoned to the neck.

. The cervical iicrtehrce are assumed to commence where the rib-pits cease
anil the continuous lateral processes commence. This point is ascertained
with difficulty on the specimen. It is, however, perhaps the same point where
tile hmgitudinal lateral ridge leaves the upper margin of the rib-pit; and it
was to the series of vertebrae which pass this point that the scapular bones,
the clavicle, and the coracoid were found attached. On the anterior dorsals,
the inferior margin of the rib-pit is most prominent, and is finally produced
into the flat, thin process which is directed obliquely downward on the cervi-
cals. Both these and the posterior ribs are crushed on the centra, and project
obliquely below them ; their mode of attachment is thus rendered rather ob-
scure. A similar structure exists in the posterior cervicals of Cimoliasaurus,
while, on the anterior dorsals, short, tiiick diapophyses support the ribs, The
proximal cervicals are remarkable for their compressed and elongate form.
They are, for a considerable distance, longer than any dorsals. The lateral
longitudinal ridge rises successively nearer to the neural arch and disappears.
The articular .surfaces are vertically oval, flattened above and below. The
inferior faces are slightly grooved in line with the venous foramina. These
vertebrae diminish in length, and, in front of the posterior third of the series,
materially in depth. They diminish to terminal ones of very small size. In
most, the decurved (?)parapophyses are broken near the base; but the l)asal
pnition of various lengths generally adheres. The}' are as wide as a rib and
scarcely half as thick. They have much greater antero-posterior extent on
tlie terminal than the proximal cervical centra, having a base five-sixths the
length of the latter. The zygapophyses have relatively a larger size on these

83

than iiii\ (jIIk r \( r1il)i;i'. In siicli. Ilic (•ciilniiii is Ics-; ctJiiipressed I lnuiali
wifli concave sides, and witli a section rallirr ([luidratc.

Tli(! coi-vicid vcit('l)rii', IVoiii Ilic sixty-sixtli to flic Inirty-niiitli, arc all
longer than the dorsals : thcv eoninicnce I'onr inches in Icnsith, increase 1o
five, and diminish (o tonr auain.

^f^■allnrcmentll.

Iix-lirs. l.illf.

I.iiigtli i>tsixt.v-lliii.I i-iivical 4 !».■-'

De]>lli oftlie articiiliir (ace of tbo ciMitniin :> (•.

Wiiltli of tilt! articular face of the (U'lilniin •! 1". -

Total elevation of the uinth cervical - '.'.

Anterior posterior diameter of tbe third cervical '■i -•

Transverse diameter of the tbiril cervical 2 11.

Length of the head of rili 1 !'• '

Width of the head of rib 1 X

Widthof the shaft of rib 1 It'.-''

Many of the r/hs jireserved have been pressed upon the vertebra- and
crushed.

The first dorsal is that vertel)ra which first presents a distinct articula-
tion for a rib. The diapophyses are never much elevated above the cen-
trum, and are longest on the thirteeiitii (inserting seven supposed to be lost).
Their form is stout and much depressed, and distally expanded. They dimin-
ish gradually, and, on the third, are represented by a longitudinal angulation;
the superior angle is first distinct on the first, and bounds the articular sur-
tiice last on the third. They give the transver.se section of" the posterior eer-
vicals a pentagonal form; that of the anterior dorsals is nearly circular. The
latter are strongly constricted medially, and the articular faces are slightly
concave. The external surfiice near the included angle is coansely ridged, in
conformity with coarse cellular texture of the spongy bone. The venous
foramina gradually become more widely separated, approaching each other
again on the posterior cervicals. On the dorsals, they occupy the bottom of a
more or less pronounced concavity. These concavities, on the posterior dor-
sals, are bounded externally by a strong obtuse longitudinal angulation, giving
a fpiadrate outline to the section of the centrum in this part of the series.

The posterior cervicals are not readily distinguished from the anterior
doisals. In the latter, the ril)s appear to be present, of reduced length, judg-
ing from the smaller size of the remaining licads. The artieidar pits continue
to descend till their lower marginal ridge is.*he inferior lateral angle of tiie
vertebra. On such vcrlebrar^, the inferior surface is flat.

84

Measurements.

]u<lii9. LiucM.

Auttio-posteiiur diaiiiuter of (ho (?) twelfth dorsal :i 7. 'J

Transverse diameter of tiie articular surface v -1 !"■

Vertical diameter of the articular surface 4 2.5

Neural caual aud spine (the latter broken) 5 H. 5

Length of the diapophysis of the twelfth dorsal 4

Width of the diapophysis at the middle 1 10.

Autero-jiosterior diameter of the (?) eleventh dorsal :! 4.5

Transverse posterior diameter of the articular face ."> 3.

Vertical posterior diameter of the articular face :i 10.

Trausvcrse posterior diameter of the neural canal 10.2

Transverse posterior diameter of the articular face of the third dorsal 5 2.5

E leva, ion of the centrum, arch and spine of the second dorsal 1 19.

Elevation of the upper edge of the zygapophysis of the second dorsal (i

Length of the zygapophysis at the upper edge of the second dorsal 1 10.2

Length of the centrum of the last cervical 4

Width of the centrum of the articular face of the cervical 5 3.

Elevation of the neural arch and spine of the cervical 7 9.

Antcro-posterior width of the neural spine of the cervical at the zygapophysis 3 7.

The caudal vertebrce have slightly concave articular surfaces, which are
not bounded by groove or ridge. The neural arches have flat sides; and there
is no longitudinal ridge above the diapophyses. The neural spines are ele-
vated ; the margins of those of the adjacent vertebrae close together. The
diapophysis is very short aud wide, terminating in a large oval concavity for
the pleurapophyses. Each limb of the chevron-bone is attached to an articu-
lar surface on the lower posterior face of the vertebra at the extremity of a
strong inferior ridge. These inferior ridges are rather close together, and
distinguish the vertebra3 from those of Cimoliasaurus magnus, where they are
wanting. They are absent on the anterior seven of the caudal series. The
diapophysis is nearer the anterior than the posterior face of the vertebra.
The venous foramen is single and median on all but the last six caudals.

Me(xs^lrements.

Inches. Lines.

Antero-posterior diameter of the fourth caudal 2 4.

Transverse posterior diameter of the fourth caudal 3 10.5

Total elevation of the fourth caudal 8

Vertical diameter of the centrum of the fourth caudal 3 1.5

Antcro-posterior diameter of the diapophysial iiit 1 9.2

Length of the ninth caudal 1 ".5

Transverse diameter of the articular face 1 •'■

Vertical diameter of the articular face 1 2. 7

Heads of fourteen ribs tire preserved, and a great number of shafts.
The heads are simple, with cloiigate-oval articular face. They are oblique
in the narrow direction, and fretpiently in their length also; the margins are
somewhat everted. The extremities of the diapophyses of the larger dorsal
v(>rtol)ia? are tntnsverse, some flattened, the others more oval; the more ante-

85

rior are siibtriangular; aud the rib-pits on the first dorsals an; siibround or
vertically oval. Thus, the heads ol" the ril)s also vary. The shafts arc all
flat, probably partly from pressure. They are frequently curve<l in llic direc-
tion of the compression, which suggests a vertical head. They, however, arc
probably more or less distorted, and the plane of compression changed. No
wcU-dctined distal extremity of a rib can be made out: nor have anythin<T like
ai)dominal ribs been preserved.

The scapular arch is remarkable for its large clavicles (or procoracoids).
As preserved, the latter are quite convex downward, i)olli antero-postcriorly
and transversely; while the coracoids are equally concave in both directions.
The clavicles have a remarkable external tlat projection, which is .separated
iioni the glenoid cavity by a deep sinus. The glenoid cavity is bounded by
an elevated ridge, which sends a branch along the claviculo-coracoid suture
to the precoracoid foramen. This foramen is relatively of small size, and is
longitudinally oval ; the two are separated by an isthmus composed equally
of ]irocesses of clavicle and coracoid. The coracoids are very thin, except
in a transverse portion, which extends across behind the precoracoid foram-
ina; a strong elevated rib extends across the [)osterior iiice at this point.
The outer margin of the coracoid is thickened, rounded, and slightly concave.

Mea&uremGnts.

lucbcH, Lines.

Greatest antero-posterior length of the scapular arcli 33 <;

Greatest autero-posterior leugth of the clavicly 14 ii

Greatest autero-postcrior length of the glenoid cavity f> <)

Greatest autero-posterior length of the precoracoid cavity 7 :i

Transverse extent of the elaviculi t>7

Transverse extent of the coracoidia K;

From the acetahulum to the foramen 7

The form of the posterior margin of the coracoidca is unknown, and tlicy
are much broken on the inner margin.

The greater part of the pelvic arch appears to be preserved. From the

ol)liqiiity of the median suture, and from the ibrm ol' the pubes, as they arc

preserved on a large nodule of indurated clay, it is evident that they have

formed a boat-shaped supi)ort to the abdominal viscera, with an obtuse keel

on the median line below.

McnsuremcnU.

Inches. LllH-H.

Greatest aDtero-posterior leugth of the pubis and ischium 25

Greatest antcro- posterior leugth of the pubis 13 (',

Antero-posterior median length to the notch of the ischia 7

Leugth of the coracoids behind the notch 4 f,

(ireatcst wiillh nf the pubPR v;7 (i

Greatest wiillh of lhi> ischi.i 21

8(^

1

The anterior and lateral portions ot the piibes are very thin, as are also

the median posterior portions of the ischiadic plates. The pubic bones are

thickest on tiie posterior margin ; they present a downward-projecting median

convexity near the anterior end. Depth of" the articular face, 2 inches 8

lines.

»
The superior surface of this arch was brought to light by the exertions

of my friends B. Waterhouse Hawkins and William M. Gabb, who removed a
large mass of matrix, which fortunately protected and accompanied it. This
presents a transverse tliickening extending across it, and continuous with the
posterior margin of tlie clavicles. A median longitudinal thickening extends
from this to the anterior emarginatiwi, embracing in its angle with the trans-
verse a shalbw concavity. The posterior projection, which is continuous
with the median part of the ischia, is strongly detlexed behind the transverse
rib, and is continuous with the basinlike concavity formed l)v the united
pubes. The glenoid surface of the pubes is a sigmoid, while that of the
ischia is regularly convex. The articulation of the ilium nas been exclusively
with the former.

Of the pleura])ophysial portion of the two arches nothing appears to be
preserved except two lateral symmetrical long bones. One was found em-
bedded in the mass carrying the pelvic arch, and they articulate well with
the pnbes ; but the articular extremity is too short to articulate with ischia at
the same time. Though they resemble the inferior view of the ])rocoracoids,
they represent the ilia oi' Plexioaaurus. The head is subdiscoid, rather Hat,
slightly j)rojecting eccentrically with a ligamentous pit. The articular surlacc
is very oblique to the axis of the shaft, and is separated from the surtiice by
a marked angle all round. Nothing like a trochanteric ridge is appai-ent in

this bone.

Measurements.

iDcbes. Lines.

Length iu tbo middle of the curve 9 9

Diameter jit the head 3 :S

Diameter, distally, on the; curve (i

Dianictor, distally, straight 4

The shaft is flattened cylindric : much flattened nearest the proximal
extremity. The latter is very oblique to the shaft, and slightly convex neai-
the proximal margin.

The end of the muzzle preserved includes also the symphysis and parts
of the rami of the mandible. '^I'lic ])!nts have been crushed together, ami \\\v.

ST

ends of" the teeth l)rokeii oil'. The alveoli ot the Iwo jaws iiifliiic ;i1 a iiai row
angle to each other; hence the teeth, which iiltciiiatc, cross each other near
the middles of the crowns. The purls preserved appear lo l)eloiiir 1o the
premaxillary bone, though no suture can be found, and the bony walls are so
thin as to render their ol)literation a probability. There is a keeled ridge
along the middle line above, which is not continued to-fhe margin of the l)one.
The form of the muzzle is narrow; tlie sides subparallel near the tip. which
is elongate rounded. The mandil)ular synipiiysis, iiowever, is not very elon-
gate, as the rami are given otf at three inches from the tip. The latter ap-
|)ear to iiav(; l)een quite slender from tln^ various small sections or pieces sent
with the nuizzle The premaxillary border of 4 inches 7 lines exhibits eight
teeth, or their alveoli, of which tlie median two are close together, and not sep-
arated by any mandibulars. The sections of the teeth are round or oval, and
their sizes are irregular, jirobably on account of dilfering age and degrtu; ol
protrusion. The diameters at alveolar margin vary from 6 lines to 3. Their
form is slender conic, or, with the root, slender fusiform, and the pulp-cavity
is small and median, sometimes cylindric, and sometimes narrowed. The
surface, from a short distance above the alveolar margin to the tip, is marked
with acute, threadlike ridges, which are sometimes interrupted, and .sometimes
furnished with short branchlets. They are more or less undulate, and do not
unite, but simply cease as the tip of the tooth is approached. The latter is
smooth without lateral cutting-edges. The width of the mandible at the
commencement of the rami is 3 inches 0.05 line; of the muzzle of the
seventh tooth, 3 inches 7.5 lines; at the third tooth, 2 inches 4.2 lines.

General liemarks. — The tail is a powerful swinnning-organ, more or
less compressed in life: hence the specific name, which means tlat-tailed.

The dang(!r of injury to which such an excessively elongate neck has been
exposed wouhl render the recovery of a perfect sj)ecimen like the present an
unusual accident. The neural spines of the dorsal region are so elevated and
(dosely jdaced as to allow of little or no vertical motion of the column down-
ward ; while, those of the cervical and caudal region being narrower, the
elevation ol tiie head is (pate possible, and an upward ilexure easy.

The iiabits of this species, like that of its known allies, were rapacious,
as evinced by the numenius caniiielike teeth, and the tish-remains taken
from beneath its vertebra?. The general Ibrui of tliis reptile was that of a
serpent, \sitli a rejalivi-ly ■shorter, more rolmst, and nmre posteriorly-plnccil

88

body than is characteristic of true serpents, and with two pairs ot" hnibs, or
paddles. It progressed by the strokes of its paddles, assisted by its jiowerlul
and oarlike tail. The body was steadied by the elevated keel of the niediaa
dorsal line, formed by the broad, high, neural spines. The siiakelike neck
was raised high in the air, or depressed at the will of f lie animal, now arched
swanlike, preparatory to a plunge after a fish, now stretched in repose on the
water or deflexed in exploring the depths below.

Comparisons. — In Cimoliasnurus magnus, the dorsals with elevated diapo-
physes have considerably larger centra than those in which they are situated
lower down. In E. platyurus, these vertebrae are oi" relatively equal length.
The cervical pleurapophyses in C. magnus are anteriorly considerably stouter
and less flattened.

In comparing this species with the Cimoliasaurus grandu, Leidy, from
Arkansas, we observe, first, the generic character of the strong inferior dia-
pophyses in the latter. That species marks itself also as a pre-eminently
short-necked form, as these anterior dorsals are even shorter than in C. ivag-
71US, being nearly twice as wide as long. The depth of the articular faces is
also relatively greater than in the E. platyurus.

Localities. — This species has been found in various parts of Kansas, lie-
sides that from which the specimen above described was procured. Prof B. F.
Mudge obtained vertebrae from a point thirty miles east of Fort Wallace, which
probably belong to this animal.

PLESIOSAURUS, Conybeare.

Two American species have been provisionally referred to this genus :
the P. lockwoodii, Cope, from No. 3 of New Jersey; and the P. gulo. Cope,
from Kansas. This determination is only temporary, since the structure
of the sternum, in which the type-characters of the sauropterygian families
are to be observed, are unknown. The two species agree with Plesiosaurus,
and difl!er from Elasinusaurus and Cimolianaurus in the non-coossification
of the arches and centra of the vertebra?.

Plesiosaurus gulo, Cope.

The typical specimen consists of eleven cervical, thirteen dorsal, luid
seven or eight other vertebrse, with portions of scapular and pelvic arch and
ril)s.

80

The cervicals are longer than wide, and coiisiderahly compressed in form
anteriorly, but depressed posteriorly. This is partly due to pressure, l)ut not
wholly; and it is likely that the i)osterior centra are abont as transverse as
in Cimoliamttrus magnus, Leidy ; while the anterior are relatively several
times as long. In (he length, ihe latter resemble the English Plesiosauri,
in which the centra arc also (•()nii)ressed. The compressed anterior centra
exhibit a ridge on the side above the middle. A more massive ridge extends
between the articular extremities at the lower part ol' the side, and presents
a pit for the parapophysis. The pit for the neural spine is of nearly simihir
size. Where the cervicals begin to be depressed, two foramina appear near
together on the inferior face, and the articular extremities display an open
obtuse emargination below. They are also eniarginate for the neural canal
above, so as to have a form approaching a transverse figure 8. In the large;
posterior cervicals, the sides are contracted both below and at the sides. In
all the cervicals, the articular fiices are a little concave; in the larger, with some
median convexities.

In none of the dorsals preserved are the diapophyses seen to issue from the
centra; hence the tbrmer are probably not posterior in position. The centra
soon become smaller than those of the posterior cervicals, and are sul)round
in section, with a well-marked emargination for the neural canal. The sides
are gently concave, and are without angulation, but are marked near the
articular extremity with short, sharp, and regular undivided ridges, eight in
a half-inch. The articular faces are slightly concave and without ridges.
There are the two inferior foramina, and one on the lower part of each side.
The articular flice for the neural arch is an oval i)it extendinjr the lenirth
of the centrum, and interrupted by some transverse ridges near the middle.
The vertebrae diminish in size posteriorly. Two centra, probably .sacral,
resemble the dorsals, but present an extensive vertical articular surface on
each side. This facet has raised edges, and terminates al)ove in the longitudinal
surface for the neural arch, having thus a T-shape ; it narrows beloAv to an
obtuse point, and. \w doubt, support(!d a free diapopliysis.

The fragments of tiie pelvic and scapular arches indicate that thov are

capacious. The clavicle incloses a large t()ranuM), and is ihickened on the

inner edge. The glenoid surface of the coracoid was wide and subrhombic.

Some oF tiie other bones are cpiitc thin. The median suture of tiu; isciiiiini

12 c

90

is relatively aljuut as lung as in the P^nglisli Flesiosauri. and the adjacent part
of the bone has a similar form.

MeasuremenU.

M.

Lcii^lli of tljc iiiiturior ceivicul 0. 06'.i

Depth <it' tlie articular face of the anterior cervical ' . . 0. 050

Width of the articular face of the anterior cervical 0.050

Length of the posterior cervical 0.070

Depth of the articular face of the po^sterior cervical 0.052

Width of the articular fare of the posterior cervical 0.090

Distance between parapophvsial pits 0.048

I.tngth of the anterior dorsal vertebra 0. OiiO

Depth of the articular face of the anterior dorsal 0.06a

Width of the articular face of the anterior dorsal 0. 072

Width of the neural canal on the centrum ii. 017

Long dianieti'r of the proxinial end of the clavicle o. 114

This saurian is readily distinguished from the Elasinomurus plati/urus,
Cope, by the relatively shorter cervical vertebra? and the regular acute ridges
on the exterior surfaces near the margin of the articular faces, as well as the
less contracted form ul' all tlie vertebral centra. As the neural arches and
the cervical parapophyses are not coossified witii the centra, the species is
referred to the genus Plesiosaurus.

The bones of this reptile and those of a smaller species, probaldy a
CUdastes, were found in close proximity, near Sheridan, Kansas, by Joseph
Savage, of Leavenworth. According to this gentleman, the vertebral column
of the CUif/ftden was found immediately below that of the plesiosauroid, and
ill a reversed position, as though it Iiad been swallowed by the latter, and larger
reptile. The largest vertelM-ag of the CTtddsfe.s were about three-quarters the
length and oiic-fourtli (he diameter of those oi' the plesiosauroid, and the
animal must liave furnished a large, or at least a long, mouthful for its captor.
The bones of the GUdmtes were not in good condition, and resembled those
of C. rineildiuiii. Cope, (hough smaller.

TESTUDIXATA.

Thisoidci is l)ii( sparingly represented in the marine tormation of No. 3,
and more abundantly in the fresh-water beds of No. (J. ()i the former, there
are three species, two of them with natatory limbs of the character now
known among sea-turtles. Of the latter, all are Triontjchidce and Emiididm;
the land-tortoises not iippeariiio- amonsr them, ticcording to pre.sent information.

01

COMPSEMYS, Leidy.

This genus presents the characters of Etnys in its \v(!ll-dev(!loped
niarifinal bones, united to the costals hy suture. The surfaces ot" the
carapace possess a dense layer, wliich is sculptured in two of the known
species. One of these, the C. rictus, Leidy, has been found to have had a
wide range in the West (hiring th(! P^ort Union epoch; while a second has
been lound in corresponding strata near the northern boundary of Dakota.

CoMPSEMYS OGMIUS, Cope.

Ki'presented in the collecliuns of the British American Boundary Coui-
Miissi(ui by portions of the carapace and plastron. These are massive, and indi-
cate a s[)ecies of large size. As in other species of the genus, the external
surface is a dense layer of cement or allied substance, which is sculptured
\\itli shallow pits.

A portion of the costal bone is concave, and increases rapidly in thickness
in one direction. The suture is coarse, but neither gomphosial nor squamosal.
A portion of the plastron is thinner, not curved, and displays a very coarse
median suture, in part squamosal in character. The sculpture consists of shallow
pits, not wider than the low, smooth ridges which separate them. There are
deep superticinl grooves, marking the boundaries of dermal areas; a feature in
which this tortoise differs much from the P. coalescens, and resembles the
species of Compaemys. Should nuirginal lioues be found to exist in the F.
ogm'ms, its rci'crence to that genus will l)e further established.

From six miles west of first branch of Milk River, near latitude 49°.

ADOCUS, Cope.

Pi-oceeiliiigs of tbe Academy <j( Natural Sciences, Philadelpbia, 1868, p. 235; I'roceediugs of tUo
American Philosopliical Society, 1870, November.

This genus possesses a large intergular plate. This I have verified on A.
heutus and A. si/ntheticus. Having also perfect xiphisternal bones of these two
species, I can show that there is no sutural attachment lor the pelvic bones.
Tiie co-existence of these two characters has been hitherto found to be uni-
versal, and the present deviation from it is a point of much interest. Instead
of sutural surfaces, there is an obtuse ridge corresponding to the pubis, and a
knob answering to the extremity of the ischium, both more prominent than is
usual ill genera of Kmydidm.

92

This exceptional combination of characters points to the propriety of sep-
arating Adocus as the type of a family equally distinct from the Emydidce iiin\
the HydraspididcB^ to be called the Adocidce.

Further characters of the genus have been already pointed out in the later
essays above quoted. They are: the free lobes of the plastron narrowed and
shortened, furnishing extensive posterior and anterior entrances to the carapace;
a series of intermarginal scuta on the bridge; costal capitula reduced or
wanting.

No recent or even Tertiary form of the Testudinata has yet been discovered
which possesses the remarkable combination found in this genus; and I think
it must be regarded as a generalized group, and as such of much interest to
the student of paleontology.

Adocus (?) lineolatus, sp. nov.

Estai)lished on a number of fragments from ditlerent exposures oi' the Lig-
nite beds, primarily on a vertebral and sternal bone, from the Dinosaurian
locality in Colorado. As the diagnostic portions of this specimen are
wanting, it is referred to this genus provisionally, and because the structure
and sculpture of the parts resemble most nearly known species of it from the
Cretaceous greensand of New Jersey.

The sternal bone is Hat, and presents the median and tran^fverse sutures
forming the usual right angle, and of a rather coarse character of a median
serrate keel, with pits on each side, for the reception of corresponding pits.
The vertebral bone is rather thick, and is shallow ]y emarginate in front. The
sculpture consists of delicate, obscure, parallel lines, which are more or less
interrupted, and occasionally joined, so as to inclose, faintly marked areolae.

Measurements.

M.

Width of the vertebral bone iu front 0. 013.')

Width of the vertebral bone (greatest) 0. 0:>80

Thickness of the vertebral bone 0. 0070

TIjickness of the sternal bone 0. OOf^O

From Lignite of Colorado, and mouth of Big Horn River, Montana.

PLASTOMENUS, Cope.

This genus has been discovered to embrace tortoises having characters
of both Trwni/.r and Eniijs. The carapace is like that of the former, in the

03

iibsPiKc (if iirticiilated inaiginal bones, and the presence of a superticial cement
layer, wliieli is sculptured in various patterns. 'I'lie plastron resembles
that of some emydoid genera, but presents cej'tain fontanelles indicating an
incomplete grade of ossification. The species known to possess the typical
structure are Ibund in the Eocenes of Wyoming and New Me-xico; and those
here referred to it are all IVoni IIk; Fori Union or Transition beds of the Cre-
taceous. In none of them is the sternum so well-preserved as to exhibit the
characters which should finally refer them to the genus Plastomenus. This
i.s due to the fact that the\', as well as other vertebrate remains from this
horizon, are always much broken or dislocated.

Plastomenus coalesceks, Cope.

This species is represented by large fragments of carapace and plastron
(>r a single individual. These indicate a large animal of adult age. Tiie
fragments are thick, and the sutures separating the component elements have
disapjieared. Dermal sutnral grooves are also wanting. Tlie portions of the
plastron preserved are emydoid in character, being most thickened in the
lateral portions, especially in the inguinal region. The borders of the
carapace are free and obtuse; at some points, somewhat thinned out. The
ribs proper, in the portions preserved, terminate in a short, free extremity^
shorter than in nio.st species of Trionyx. There is no indication of the
existence of marginal bones.

The surfiice of plastron aixl carapace is covered with a dense layer,
which is thrown into rather coarse, inosculating folds These lurni an open,
reticulate pattern towards the middle portion of the cara])ace, and become
obscure near the Ijorders They are well-marked on the plastron, and arc
more or less longitudinal. The appearance is that of a Trionyx.

The costal axis scarcely projects on the inner fiice of the carapace. Tiie
anterior border of the carapace is a free, thickened margin, divided by a hor-
izontal groove. The presence of nuchal l)one cannot be ascertained.

Measurements.

M.

Tliickuess of the carapaco at tUu iiiidiUf of the leii^jtli of a costal boue n. (114

Thickness of the carapace at the atiterior margin li. 008

Thickness of the carapace at tlio lateral margin 0. 00«

Thickness of the plastron neiir the briiljje 0. 015

Thickness of the plastron ninrc centrally (i. OO'J

I'onr arenhn in 0"M0.

94 .

This species is Iburid in .1 jjreeinsh-brown areniiceoiis rla\' deposit near the
Milk River in British America, belonging to the Transition series, probably
tiie Fort Union or Lignite epoch. Collected by George M. Dawson, oC
Montreal, geologist of the British North American Boundary Commission,
near Woody Monnt.

Plastomenus costatus, Cope.

Represented, in the collections made by Mr. Dawson, by small portions
of plastron and carapace, which display distinct osseous, but no dermal scutal
sutures. These specimens were discovered together, and are believed to
belong to the same individual. The bones are thinner than corresponding
ones of the two other species of tortoises described, from the same locality,
(excepting at the costal enlargement, which is remarkably prominent and \\ell-
detined on the under .side of the carapace. The dense or cement layer of
the carapace is thrown into very delicate, but prominent ridges, which run
parallel to the axis of the carapace, and occasionally inosculate, or are crossed
by a similar ridge rumiing at right angles to them. The sculpture of the plas-
tron is similar, but more obtuse and obscure The superior edge of the free
border of one of the lobes of the plastron projects beyond the inferior, and
is not, as is usual, less prominent than the inferior.

Measuremoits.

M.

'I'bickness of a costal .at the margin 0. 005

Thickness of a costal at the rib 0.009

Thickness of the free edge of the jilastron 0. 009

'I'he costal bone of this species is much like that of a 7rioin/x, but the
character of the plastron refers it to Plastomenus

Collected in the bad lands of the Fort Union Cretaceous, south of
Woody Mount, near latitude 49°, British America. Associated with this
species were the F. coalescem, Comjjuemi/s ogmius, fragments of perhaps
Compsemys vidus and Trioni/x vagans, with Dmosauria, Cionodon stenopsis,
etc. (See chapter I.)

I Plastomknus punctulatus. Cope.

Established on a costal bone found in as.sociation with the preceding
species, and referred tj the genus Plastomenus provisionally, and with a
])ossibility that it will be found not to pertain to it when fully known.

Tliat ucims liii!^ so far only been louiid in tlic Moccnc I'ornuilKtn. 'llir lionc
is rather thin, and sufficiently curved to indicate a convex carapace of moder-
ate thickness. The surface is marked with (^osely-packed shallow pits w itli-
ont material variation of l()rm on the proximal half of the hone. The result
is an ()i)Solete sculpture (piite similar to that seen in some species of the
genus to which it is at present referred.

MeaniDyiiunts.

M.

Width of the lu.stal Ijoiic (». 02:50

Tliickiios.s of tlic costal Iioiic 0. (!(K!;{

Niiiiilicr of ])it» ill l)'".0U1, (1.

Lignite Cretaceous of Colorado; also, several fragments from Long
Lake, " Nebraska,' from Dr. H.iyden.

?Plastomenus in.signis, xp. nor.

Represented hy u portion of the right hyposlernal l)one of a tortoise
about the size of the la.st species, and from the same locality. The specimen
resembles, in its sculjjture, such species as the Plasfomenus trioni/choides, and,
in structural character, the species ol' Anosiira, but it is scarcely probable that
it belongs to either genus. It is flat, and has a narrowed, straight, inguinal
margin, at right angles to the fine suture with the hyosternal. The suture
with tlu! p()stal)domiiuil is partially goniphosial. Surface dense, polished,
marked externally with a reticulate sculpture of narrow ridges separating
larger and smaller areas wider than themselves. Inguinal edge thinner.

Measurcmentx.

M.

Length of the hyposternal fore and .ift . 0. 02.">

Thitkucss of the hyposterual at frout 0. 004

Pits in 0™.010, six.

Lignite Cretaceous of Colorado.

TKIONYX, Geoffr

Although species of this genus occur in the greeiisand of Cretaceous No.
4, in New Jer.sey, none have l)een discovered in the West below the horizon
of No. 6, or the Fort Union fresh-water beds. Dr. Leidy has described a 7'.
foveatus from the bad lands of the Judith River, Montana; and 1 have a<ltled
the foHowing:

96

Tkionyx vagan.s Cope.

Bulletin of tlic United States Geological Survey of the Tenitories, No. 3, 1874. — Trionyx ? forentus, Leidy,
Proceeding.s of the Academy of Natural Sciences, Philadelphia, 185G, p. 312.

Represented by a number of fragments of costal bones, and, perhaps, of
sternals, also. The former are rather light or thin for their width, and are
marked with a honeycomb-pattern of sculpture, in which the ridges are thin
and much narrower than the intervening pits. They incline to longitudinal
confluence at and near the lateral sutures. Several areae are not unfrequently
confluent in a transverse direction near the middle of the bone.

Measurements.

M.

Width of the costal bone 0.0370

Thickness of the costal bono 0. 0045

Number of area; in 0'".019, 4 and r>.

This species differs from the T. foveafus, Leidy, in the much narrower
interareolar ridges and larger areae, and in their longitudinal confluence at
the margins, characters exhibited by munerous specimens.

Lignite Cretaceous of Colorado; near the mouth of the Big Horn River,

Montana; Long Lake, " Nebraska;'' found at the last two localities by Dr.

Hayden.

CYNOCERCUS, Cope.

Established on a metapodial bone and caudal vertebrae of a tortoise of
uncertain, but in any case peculiar, aftiiiities. The caudal vertebrae are not
anterior ones, almost lacking diapophyses, but are long and slender, and the
articular faces singularly incised. The form had a tail more elongate than
the snapping-tortoise, and different from it in details of composition, especially
in l)eing of the procoelian type.

Associated with the remains of CUdastes, and other saurians, and at a
distance of two or three Iiundred yards from the locality of the fossil Proto-
stega gigas, were found some vertelirse of a Testudinate reptile, which ap-
proaches the type of Trionyx.

Cynocercus incisus Cope.

The vertebrae have elongate centra concave below, and have well-devel-
oped diapophyses. One vertebra has transversely oval articular extremi-
ties; in another, they are much less depressed. The former is the more an-

97

lerior, being known as sucli by its larger (iiapophyscs and much smaller artic-
ular surfaces for chevron-bones; it appears probable, indeed, that this one has
been without these appendages. It is, therefore, from no great distance
behind the sacrum. Its position being thus determined, it may be described
in detail as follows :

As observed, the centrum is elongate and depressed. Tiie inferior sur-
face at the cup is flat ; it is then arched upward, descending again to the rim
of the ball. The posterior two-thirds has a median groove, which terminates in
a deep notch of the ball, which involves one-third of its vertical diameter, and
widens backward. The ball is transverse oval, and only moderately convex ;
near its upper margin, a small deep pit interrupts its surface, having the
appearance of an unusually large ligamentous insertion ; its border slightly
excavates the margin of the ball. The cup is a transverse oval, wider below.
Its inferior and superior margins are so deeply (but openly) emarginate as to
reduce the concavity in the vertical direction very much. From tlie superior
emargination, a deep groove descends to below the middle, probably for liga-
mentous insertion. The neural canal is subtrilateral. The neural arch is, as
usual in this group, deeply emarginate in front, and much prolonged behind.
The zygapophyses project beyond the ball, and the arch is contracted in front
of them. Its upper surface has neither process nor keel, but is rugose for
ligamentous and muscular insertion. The diapophyses have a wide base, and
are subcylindric.

The surface is delicately reticulate rugose; coarsely rugose on the exter-
nal faces of the zygapophyses. There are several small nutritious foramina,
the largest being in the bottom of the groove of I he lower face.

Another vertebra differs in being rather more slender, and in having an
obtuse keel of the neural arch. The pit of the ball is wanting, and the iiifc-
rior emargination. The chevron-articulations are larger; and the groove of
the cup occupies its middle, instead of its upper half

Measurements.

ii.

Length of tlio ccntrnm (greatest) 0.027

Diameter of the cup, vertical 0.01

Diameter of the cup, horizontal 0. 017

Elevation of the top of the neural arch above the floor of the noaral canal 0. Oi;5

L"ngth of tbe arch on the median line above 0. 02

Width of the arch in front of the posterior zyg.apophysos 0. 012

A metacarpal or metatarsal bone was found near, though not willi, the
vertebra;, and probably belongs to the same animal. If metatarsal, it is much

98

stouter than in Trionyx, but is more likely to be a metacarpal. It is about
as long as the vertebrjE, centrum and arch together. The proximal end is
transversely truncate, compressed L-shaped ; the shaft compresscd-subquad-
rate; the articular extremity hour-glass-shaped, with an inferior projection for
the insertion of a flexor-tendon. This bone is not that of a marine turtle,
but ot a species of riparian or terrestrial habits. Length, 0™.034; proximal
diameter, 0'".013.

These vertebrae indicate a genus with elongate tail like that of Cheljdra,
or probably longer ; but they differ from those in that genus by their procce-
lian character. An approach to their incised margins is to be found in Trio-
nyx; but in those of that genus, where this character appears, the diapophy-
ses are largely developed. The genus is evidently quite distinct from any-
thing known, and we await further remains with interest. The species is
much smaller than the Protostega gigas, and about equal to the Mississippi
Macrochelys.

TOXOCHELYS, Cope.

This genus is represented by a single species as yet, which reposes on
a number of specimens. These indicate a structure in many respects similar
to the genus Chclone, but sufficiently different to belong to another genus

The mandibular ramus is slender, and has a narrow, flat, alveolar surface.
The coronoid process is moderately elevated, and is excavated behind by the
anterior extremity of the elongate and deep dental foramen. The cotylus is
depressed, and the articular bone ossified; the angle is not produced. The
coracoid bone is long and spatuliform, like that of the marine turtles. A
fragment of the same skeleton supports a condyle with slender, subcyliiidric,
slightly-curved bone attached to one side of it. Were it not a turtle, I should
say that these are the extremity of the quadrate, with a slender jugal or
malar bone adhering.

The slenderness of the mandibular rami resembles the form in Chelydra,
but it differs in the absence of the alveolar cutting-edge of the latter. The
]>lialanges are broad and flat, and not unlike those of Protostega.

ToXOCHELYS LATIREMIS, CopC.

The symphysis mandibuli is very short, and the upper face horizontal.
The angle is truncate beliind, flat, and not produced beyond the cotylus.

99

Measurements.

M.

Length of tlio co.acoid 0.250

WiiilU of tlie coracoiil distally 0.080

Lenstli of the ramus of the inaudible 0. 157

Length of the sym)>hy8i8 of the mandible 0. 031

Elevation of the coronoid of the mandible 0.043

Width of the- angle of the coronoid 0. 023

The axes of the mandibular rami produced unite at an angle of 65°,
indicating a muzzle of intermediate length.

The size is about that of the existing loggerhead (Careiia caouand).
Found by Professor Mudge near the forks of the Smoky Hill River.

Fig. 2. — Sketch of the large radiated bone of I'rotostega gigaa, with other elenxMits, as they were
uncovered by excavation; size much icductd, drawn on the spot. Nos. I, "Vll, VIII, and IX, costal
bones, with the rib-heads looking upward ; X, the coracoid bone ; 5, a marginal bone ; 9 and 10, the lateral
a radiate bones; 16, vertebra, with other bone adherent.

PROTOSTEGA, Cope.

This genus is the type of a new family of tortoises of the suborder
Athccce, characterized by the lack of exj^ansion of the ribs into a ])ony roof,

100

or carapace, and the development of independent superficial dermal bones.
The dermal bones consist of large plates lying above the ribs, which have no
sutural union with each other; of some small vertebral shields on the dorsal
line ; and of tbin, marginal bones, which have no sutural union with each other
or with the other bones. The vertebras pi'eserved possess ball-and-socket
joints, and have flat neural arches, with widely-spreading articular processes.
The humeri are flat, and furnished with an enormous deltoid crest. The fore
limbs were very long, and formed flippers like those of the marine turtles
of the present seas. The bones of the head are very liglit and thin, and
mostly united by squamosal or overlapping sutures. The mandible presents
the elements usual in the marine turtles, and has no angle. It exhibits a
deep pterygoid fossa, and is very light. The constitution of the bones is
rather dense, and there are no medullary cavities whatever. The superficial
layer is very thin and striate. The bones are all very fragile.

The affinities of this genus appear to be largely to the Sphargididce.
This family is represented, in our present knowledge, by but one genus and
one species of the recent seas. It is one of the most generalized, or, in
special characters, the most aberrant, of the order of tortoises, and the dis-
covery of an extinct ally, even as far down In the series as the Cretaceous
period, is not surprising.

The remains preserved belong to a single individual, and include many
portions of the cranium, five vertebrae more or less incomplete, the scapular
arches of both sides, with the coracoid bones; both humeri perfect, with nine
phalanges, ten ribs, one vertebral(?), and ten marginal bones; parts or wholes
of four large lateral (!) dermal bones, with five distinct bones of unknown
reference. There are also some slender curved bones, which probably pertain
to tlie plastron.

As the bones were exposed by excavations in the yellow Cretaceous
chalk, sketches of their positions and relations were made, which aid mate-
rially in the restoration of the animal. The upper layer of bones were those
of most irregular form, as cranial and limb bones. Mingled with these, but
often beneath them, were the ribs ; while underlying all were the large flat
pieces here described as dermal. Adhering to the inferior surface of these
was a layer of thin oyster-shells, with parallel striate surface, perhaps Inoceratni.
The ribs presented their heads upward, so that, taking all points into con-

101

sideration, there is little iloiibt that the reptile was entombed lying on its
back.

The texture of the bones is peculiar. There are nowhere to be seen
medullary cavities, and the bone is spongy, but very finely so; the tubules at
the largest being equal in diameter to an ordinary pin, and generally consid-
erably smaller. They are arranged in concentric series. There is no thick
dense layer of the bone as in other tortoLses, but an extremely thin one,
which is hard, and sculptured on the surface with minute grooves or pits.
The tissue of the bone is very fragile, and has si fracture like the mineral
inclosing it. Many of the bones, especially those of the dermal skeleton, are
extremely attenuated on the margins, being no thicker than writing-paper.

In discussing the affinities of this genus, one doubtful point must be
considered. The large tlat elements described as lateral dermal bones, are
they ossifications of the dorsal or ventral integumentl They were found
below all the other bones, and nearly all the ribs laid on them with their
heads turned upward. This rendered it probable that the shields were
dorsal, and that the animal was entombed on its back; and a coracoid, which
was afterward found lying immediately on the largest bone (No. 10), crossed
in its course parts of two ribs. This could not have been the case had the
shields been ventral. An examination of the shields docs not reveal any
conformity to any known type of Testudinate plastron. The bones radiate
in all directions, leaving no margins for fore and hind limbs, or for a median
fontanelle, still less for suture with each other.

Should these bones then be regarded as dorsal, they constitute a charac-
ter not previously noticed in the order, but one whose homologue is seen
probably in the dermal shield of bony tesselated plates seen in Sphargis.
The other points of affinity to Sphargis are the distinct ribs; the thin lamin-
iforni jaws with cutting-edges; the quadrate bone with such a strong anterior
concavity ; the elevated position of the zygomatic bone ; the form of the
humerus. Points of special reseml>lancc to Chelone are : the short posterior
superior portion of the quadrate ; the entire edge of the maxillary bone ; the
deep dentary. Tlie points in which it diffisrs from both are numcrou.s.
They are : the dorsal shields; the marginals; the notched symphysis, etc.; tlic
shortened articular end of scapula; the elongate form of the carapace, etc.

The constant separation of (he ribs and the shortvertebrae are charac-
ters which arc more like those possessed by other reptiles Ihau (liose cliarac-

102

teristic of Testudinata. The presence of dermal dorsal bones is of the same
kind. Tlie genus Protostega then belongs near the Sphargididce in the sub-
order Athecee, and is in some points to be approximated to the Cheloniidce.

Protostega gigas, Cope.

There are twelve marginal bones. They are all characterized by their
laminar form. The thinnest are those farthest removed from the middle of
the sides. They consist of a single lamina, slightly thickened within the
margin, producing a slight convexity of the lower side. The proximal part
of the bone is an extremely thin plate, with radiating ossification, and conse-
quently more or less serrate margin. It extends some distance over the
extremity of the rib, whose apex is received into a half-pit or acuminate
groove with abrupt termination, about one-sixth the width of the bone from
the margin.

In following the marginals to the middle of the side, the edge, as usual,
increases in thickness. The lower side, becomes more convex, and the upper
slightly concave. The edge is acute, with a very open interior entering angle
at the middle. The lateral extremities of the marginals are irregular, termi-
nating in a double series of closely-packed digitations, which terminate freely,
and enter into no suture. The pit receiving the extremity of the rib ap-
proaches the margin, which now develops an inferior lamina of bone. This
incloses the end of the rib, and thins out laterally in contact with the superior
plate. In some, the inner lamina is short ; in others, it is almost as extensive as
the outer part of the marginal plate, causing the double appearance when frac-
tured. As the marginals thicken, a distinct inferior plane becomes distin-
guished, separated from the interior face by an obtuse angle. The upper face
near the margin is more concave. In the thickest, the inferior face is also
somewhat concave, and the edge quite acute. The lateral extremities consist,
as before, of packages of digitations, which easily break out.

A single nearly bilateral bone of this series appears to be either nuchal
or caudal; but, as it has no sutural connection with any other, it is not easy to
determine which it is. Its marginal length is much less than its transverse
extent, which consists chiefly of a flat lamina. The marginal part is a little
thickened, and bilaterally concave below, and correspondingly convex above.
The margin is thin and acute. A few grooves radiate at a distance from the
middle toward the margin. The lack of concave excavation of the margin
would incline the balanct; in iiivur of the view thai this hone is \\\v caiKlal.

t)

103

A very long, gently-curveil bone is probably the marginal extending on
one side of the nuchal. It is nearly twice as long as the others, and has an
extensive and thin superior lamina. Its ? anterior part is in one plane; but
the margin soon thickens, and displays a rather wide infero-external face. It
appears to have had an inferior lamina on its posterior half, which made an
angle with the face just described. An oval cavity included looks as though
designed for the apex of a rib.

The variation in the lengths of these marginal bones is noteworthy. I
give measurements, premising that a few lines may be added to the extremi-
ties of some for lost digitations.

Measurements.

M.

Lenj;tU of the long anterior (11 inches) 0.28

Widtb of the long anterior (some lost) 0. 135

Length of the lateral with the inferior face — 0. 195

W'iclth of the lateral (mnchlost) 0.110

Length of the lateral with narrower inferior face 0. 1!06

Width of the lateral (much lost) 0.115

Length of one with the interior lamina 0. 14

Width of one (soaie lost) 0. 086

Length of a thinner, no inferior plate 0. 10.3

Lengtli of a still thinner, no inferior plate 0. 1S2

Width of a still thinner (broken) 0. i:!7

Length of the thinnest -._ 0. 1()2

Width of the thinnest (nearly complete) 0. 160

? Caudal length 0.117

'Caudal width 0.150

The shortness of the marginal with large interior lamina is noticeable, as
also the same peculiarity in the caudal. As compared with marine turtles,
difference is to be observed in every particular. Such are the lack of sutural
union; the laminar cliaracter; the great extent of the superior and distinctness
of lower laminfE. There is no trace of epidermal sutures visible anywhere.

A single .symmetrical plate appears to have belonged to the middle line
of the back or nape. It was subtriangular in outline, all tiic margins very
thin, and with an obtuse keel extending on the middle line, on the posterior
(or anterior) two-tbirds to the apex. This ridge disappears in ? front !)y a
gradual expansion The surface is marked by lines of minute pits and
grooves, which radiate from the base at the (I) front of the ridge. Length,
0M35; width, 0'".21.

The lateral dermal bones preserved, are two entire, and large parts of
one. or two others. They have an irregular oval outline, and nrr slightly
dished on the inferior surface or that next the ri])s. The u[)[K;r .surface is

104

more convex longitudinally, from the thickening of the bone. The margins
are irregular, from the projection of many tligitations. Some of these are
broad and flat; others are narrow. They are frequently two deep, and the
fissures separating them occasionally extend far toward the middle of the
bone. The convexity assumes the form of a low ridge toward one end of
the bone. At the point where this reaches the margin, the latter is in all the
four plates, thickened, and composed of several layers of packed osseous radii.
When found, the ribs laid across these shields, one of them occupying the
position of a radius to one of them. These shields are much larger than the
marginal bones.

Measuremenis.

M.

Leugtli of "No. 10" (21 iucbes) 0.535

Width of "No. 10"(brokeD) 0.400

Thickacss at the middle 0. 014

Lenjrth of " No. 9 " 0.530

Width of "No. 9" (much broken) 0.350

Thickness at the middle 0.013

The lengths and breadths given are a little below the truth, owing to
the loss of the exceedingly thin margins.

Turning to the endo-skeleton, the vertebrce deserve mention. There are
more or less complete examples of five of these ; in two, both centrum and
neural arch, in two neural arch, and in one centrum, are preserved. These
have been recognized chiefly by their neural arches, which are separate.
They are in form something like an X, the extremities of the limbs carrying
the zygapophysial surfaces. The only point of contact with the centrum is a
wide process, which stands beneath the anterior zygapophysis, and spreads
out footlike obliquely forward and outward, to beyond the line of its anterior
margin. Its surface extends nowhere posterior to the surface of the zygapo-
physis above it, but a little farther inward. Its outer margin rises ridgelike
to the under side of the neural arch, and each one, forming a semicircle, forms
the boundary of the neural canal, and, turning outward, forms the inner
boundary of the posterior or down-looking zygapophyses. The space between
these apophyses is roofed over, so as to produce a shallow zygantrum, which,
however, only seems to roof over the deep emargination of the neural arch
of the vertebra immediately following. The anterior zygapophyses are often
broken away, so that the neurapophysial supports look like the missing pair,
when the difficulty ensues that both pairs look downward. The top oi' the

105

neural arch is, in two cases, broad aiul flat; in two others, there is an obtuse
keel.

The centra, apart from their arches, are puzzling bodies, especially since
in the present case, they are somewhat flattened by pressure. They differ
materially in size ; one of them being twice the size of the others. The
smaller ones are of the ball-and-socket type, and have a deep longitudinal
groove on each side. The thicker portion of the centrum forms the inferior
boundary of this pit-groove, while a thinner portion, possibly a diapophysis,
limits it above. It is, however, thin, and had no great length. There is no
sign of chevron bones and articulations, so that these vertebrae may have been
cervical. Their bodies are, however, shorter and wider than in those vertebrae
of any known tortoise. A groove on the upper surface represents the neural
canal ; while a flat area on each side, in front, supports the neurapophyses.
The large centrum exhibits the superior groove and antero-lateral platform
for support of the neural arch. One end is cupped obliquely, while the other
is nearly plane, with the same obliquity and a slightly-raised margin. Its
outline is subtriangular. The lower side of this centrum possesses a short
keel posteriorly. The sides exhibit no pit, but have a thin edge, which is
concave behind the middle, and then turned outward. I can see no articu-
lation for a rib.

The forms and characters of these vertebrae resemble Sphargis moi'e
than anything yet described.' Either the large or the small, or both, must
be referred to the dorsnl region ; in this case, the concavity of one extremity
is a new feature among tortoises, so far as known. The great freedom of the
arch from the centrum is very peculiar; while it is probable that the articula-
tions of the ribs were to the middle of the side of the body, and not to the
adjacent parts of two bodies, and may have been (see below) to processes or

diapophyses.

Measurements.

M.

Length of the modium contniin 0. 0:17

Width of the iiiediiim coutrum 0. 0110

Length between tlie margins of the zygapopb.vses of the mcdinm centrum 0. 0(>0

Width between the anterior margins of fhe zygapophyses of the medium centrum 0. 070

Width between the posterior margins of the zygapopliyscs of the medium centrum 0. 047

Width between the anterior bases of tlie arch 0.070

Width of the arch at the middle 0.028

Lengtii of the arch at tlie middle (*• O-JO

Width of the posterior zygapophyses of No. 2 0.048

Width of the arch O.O'i'J

' See Teuiniinck and .Schlegel Tauna Japoiiie.a, Reptiles, Tab. 1-4.

14

106

M.

Length of the arch 0.025

Length of the anterior foot (oblique) — 0. 020

Length of the ceutnim of the large one 0.060

Width of the centrum of the large one 0.094

Width of the neural canal of the large one 0.017

Ten ribs were recovered. These are slender and rather flatter than in
most reptiles, but without the peculiar form characteristic of tortoises and
turtles. They are most expanded proximally; the bone spreading into a lam-
ina from the tubercular region, extending laterally and proximally some dis-
tance beyond the head. The superior plane of this expansion is continuous
with that of the rib, and is flat ; the head of the rib therefore turns down-
ward and inward from it, to join the vertebra. Now, the extent of the inner
part of the lamina is such that, were the head articulated to any of the cen-
tra discovered, the laminae would interfere or overlap. They may, therefore,
have been articulated to diapophyses. The expansions are serrato-digitate
on the margins, and exhibit radiating grooves and ridges in some places on
the superior aspect. The lengths of these ribs are not so great as the pro-
portions of some of the other bones would indicate.

Measurements.

M.

Length of "No. VI "(16 inches) 0.510

Width at the head 0.140

Width of the head 0.040

Width at the middle 0.055

Width at the extremity 0.040

Length of "No. II" 0.300

Widthof "No. II "just below the head 0.100

Width at the middle 0.037

Lengthof "No. IX" 0.380

Length proximal to the head 0. 060

Width at the middle 0.080

In the rib "No. II," the head is turned obliquely to one side, indicating
that the rib diverged at a strong angle from the vertebral column ; in fact, not
more than one of 45°. This is, then, an anterior or posterior rib ; probably
the latter, since the shell is usually expanded chiefly in that direction. All
the ribs are flat above, and convex inferiorly.

Both sides of the scapular arch are complete, except the sutural portions
of the coracoid and scapula of one side. The scapula and procoracoid make a
very open angle with each other, both being stout; the scapula the longer,
with grooved sculpture at its j^roximal end. The procoracoid is a little the
shorter. The glenoid cavity and coracoid suture are almost sessile at the

107

union of the scapula and procoiacoid. The coracoids arc very elongate,
almost equal to the ribs, and not stouter except at the extremity. It is
expanded into an oblique head proximally. The shaft is flat; one edge thick-
ened or truncated; the other thin. The distal portion is scarcely expanded;
being more slender than in any recent Testudinate known to me.

Measurements.

M.

LengtU of the scapula to tlio glenoid cavity 0. 213

Width of the scapula proximally 0.045

Length of the procoracoid to the articular surface - 0. 106

Width of the procoracoid distally 0.060

Length of the coracoid 0. 400

Width of proximally 0.080

Width of medially 0.047

Width of distally 0.080

The elongate coracoid resembles most, among recent Chelonians, the
marine genus Chdone; while the sessile glenoid cavity and short procoracoid
with open angle are entirely different. In these points, this genus is more
like terrestrial forms, as Testudo, or less like Emys.

Both humeri are entirely preserved. They appear to have been some-
what flattened by pressure; but, when unaltered, they were, no doubt, flat, with
stout proportions. They have a globular head, with an immense trochanter,
which projects much beyond it proximally. The shaft is then much con-
tracted, and expands again distally to the broad and very convex articular
extremity Opposite the narrow part of the shaft, the small troclianter
appears on the inner side, forming an elongate ala. Tlie long axis of the
liumerus is not straight; the proximal and distal portions making an angle of

110° with each other.

Measurements.

M.

Total length of the humerus (1 foot) straight u. 3U0

Length of the humerus from the head 0.290

Width at the head , 0. l.-SG

Width of the head 0.073

Least width of the shaft 0.07G

Greatest distal width 0. VM

Length of the basis of the little trochanter 0. Q80

Thickness at the shaft 0. 0:U

The flatness of this element, and situation of the large trochanter in the
general [dane, are characters of the Sphargididce. The great constriction
medially and expansion of both exircmilies romiiul one of tlie niosnsaiu-oid
humerus.

108

Of bones of the fore-arm, there may be one ; but the bones next in
size to the humerus look more like metacarpals or metatarsals. Two of
them were found together in position; and their relations were not like
those seen in the fore-arm of sea-turtles. They measure over seven inches
in length, and are strongly concave on their adjacent sides. One of them is
slightly concave on the outer side ; the other convex, the convexity being at
two-fifths the length from one end. The ends of both are a little expanded ;
and one end of one displays a double or trochlear extremity. The same end
of the other is injured by pressiire. A still larger metacarpallike bone is
relatively more expanded at the ends. The articular surface of one of these
is wide at one end, and much narrowed at the other. The smaller bones,
undoubtedly phalanges, are six in number. They are quite slender, a little
expanded at the ends, and flat.

Measurements. \

M.

Length of the largest • 0. 1C5

Width of the largest at the extremity 0. 0C5

Width of the largest at the middle 0.032

Length of the longest of the pair 0. 180

Width at the extremity 0.C50

Width of the shaft 0.027

Length of the phalange 0. 132

Width at the extremity 0.032

Width of the shaft 0.018

Thickness of the shaft 0.011

These measurements indicate, for the fore-limb, a total length of 4.52 feet
(1".347) if proportioned as in Chelone; this would give an expanse of 11.3
feet. If, however, it was constructed on the plan of Sphargis, the expanse
would be nearer seventeen feet.

Several instructive cranial bones were preserved. These are the maxil-
lary and distal part of the dentary of the left side ; the posterior part of the
left mandibular ramus ; quadrate bones and adjacent pterygoids and squamo-
sal, one side with the columellar plate; right postorbital bone and part of the
left ; also, some probably hyoid elements.

The maxillary bone and the dentary present a considerable extent of the
alveolar margin. This is remarkable in being thin, sharp, and elevated;
without horizontal portion. The former bone is but little incurved to the
premaxillary suture ; its anterior outline is elevated and vertical, the nostrils
entering opposite the probable middle of the orbit. The palatal plate of the
maxillary has no great antero-posterior extent, so that the inner nares arc

109

opposite the anterior part of the orbit. The latter presents only the anterior
and inferior outlines in the specimen. The part of the maxillary below it is
very narrow, and weaker than either Sphargis or Chelone. The cutting-edge
has a very open sigmoid flexure, the suborbital part being turned inward, the
anterior part a little outward. The osseous rim of the orbit projected out-
wards considerably beyond the plane of the maxillary anteriorly.

The dentary hone is very deep anteriorly, and, like the maxillary, is a thin,
vertical lamina. The lower anterior angle is truncated by an acute, concave
margin. This is the anterior extremity of the symphysis. This suture
occupies the inner face of a triangular area, which extends but a short
distance on the lower margin of the ramus, and then passes upward and
backward for a short distance on the inner face of the ramus. That portion
above the symphysis diverges outward; thus producing a deep notch at the
symphysis, as though designed to receive a beaklike projection of the pre-
maxillaries. The cutting-edge has a slight sigmoid flexure, corresponding
with that of the maxillary ; it rises into a projecting angle.

The posterior part of the ramus displays the cotylus, and, in front of it,
a deep, long fossa behind the articular bone. There is no angle nor coronoid
bone, as in all marine turtles. The superior margin of the dentary is thicker
posteriorly than in front; and its outer wall is produced backward as a thin
lamina, covering the surangular almost to the posterior edge of the ramus.
The angular is, as in recent forms, a narrow, wedge-shaped piece below the
dentary and surangular. The posterior edge of the surangular projects behind
the dentary, and exhibits an acute, convex edge rising forward. It supports
a small part of the articular cotylus on its inner face. Most of this portion
occupies the extremity of the articular. The latter sends a stout lamina
obliquely upward and forward to the lower posterior part of the dentary.

The quadrate bones are of peculiar form. They exhibit the usual jJoste-
rior curvature above, with a shallow funnellike fossa for the tympanic cavity.
It presents two strong ridges anteriorly, an inner and an outer, which inclose
a deep, vertical concavity. The inner exhibits the suture with the pterygoid
bone; the outer, with the zygomatic. The superior border of the quadrate
within the squamosal is massive, and not inflated. Its surface is thickest
where the usual articulation with the opisthotic exists. The posterior hori-
zontal is short and deep. Tiie transverse part of the bone which support^
inferiorly the exterior i)art of the condyle is thin, and disappears above to the

110

antero-posterior portion. From its middle upward, it supports the zijgomatic.
The latter has no great extent anteriorly to its malar suture ; and its inferior
margin arches high above the line of the condyles of the quadrate.

^\i& pterygoid bones are subtriangular in outline, with concave sides, an
emarginate base, and a very obliquely truncate apex, which articulates low
down on the quadrate bone. Both margins are thickened and rounded ; the
superior as a boundary of the foramen ovale. The posterior margin of the
platelike columella overlaps it on the inner side, deeply notching it ; on the
outer side, the suture is zigzag and transverse. The superior part of the
bone is produced like a flat rod, and, at its end, exhibits a squamosal suture
for union with what is, in the snapper, a postero-inferior rod-like prolongation
of the columella. No such process of the columella appears to exist in this
species. The columellar plate is half as large as the pterygoid, and exhibits
the obUque suture in front for the descending lamina of the parietal.

The postfrontal hone of the left side is preserved entire, and the inferior
portion of that of the right. The inferior margin for the malar is the longest,
and is straight. The orbit is excavated in part from its anterior margin ;
while the supero-posterior is a continuous curve. The inferior suture is a
groove, whose inner bounding wall is convex, but rises past the straight outer
to an inner ridge, which probably approaches the ectopterygoid region.
A large sutural face for the zygomatic exists at the lower posterior angle, and
an elongate one above for the parietal. The inner face is concave, indicating
a large temporal fossa, as in Sphargis and Chelone.

Two bones, of opposite sides of the cranium, are either those portions

of the pterygoids which bound the temporal fossa below in front, or those

portions of the maxillary bounding the palatine foramen. As the free margin

is much thickened, they are probably the former. Their inner, or thinner,

lamina is marked for squamosal suture with other bones, perhaps columella

and palatine.

Measurements of the cranium.

M.

Depth of the preniaxillary sutnro of the maxillary 0. 060

Length from the premaxillary suture to the iuner nares 0.008

Depth of the maxillary below tho orbit 0.035

Depth of thedentary at the symphysis 0.078

Depth of tho notch of the deutary at the symphysis 0.044

Depth of the dentary behind the symphysis 0.09,5

Depth of the deutary at the coroiioid region 0. 085

Depth of the ramus at the front of the cotylus U.OGl

Length of the pterygoid fossa of the cotylus 0. OGO

Ill

M.

Length of the cotyloid fossa 0.070

Length of the postfrontal on the inferior suture 0. 195

Depth of the postfrontal at the honndary of the orbit 0. 13G

Thickness of the postfrontal of the lower suture 0.019

Length of the postfrontal from the orbit (^oblique) 0. 115

Length of the right quadrate 0. 140

Width (antero-posterior) 0. 110

Width of the condyle 0.0G4

Length of the right pterygoid superiorly 0. 155

Depth of the right pterygoid at the inner columeUar angle 0. 100

Leegth (oblique) of the columella 0.085

Restoration. — Better materials exist for the restoration of this species
than is usual in the case of most extinct Testuclinata. The cranium was
0°.50, or 24f inches, in length. If the neck and carapace were related to it
as in the genus Chelone, the total would be as follows :

Cranium 24f

Neck and carapace 1391

Total, 12.83 feet lC2f

an extent not far from the expanse of the flippers above given, viz, 11.30 feet.
The shortness of the cervical vertebrse indicates that the proportions of the
neck were not dissimilar to those of the existing marine genera. The flippers
were probably similar to the same ; of the hind limbs, nothing can now be
stated. The shortness of most of the ribs, considered in connection with the
length of the marginals, is remarkable. Thus, the longest rib measures O^.Sl,
or 16 inches; width of lateral marginal beyond apex of rib, 2.25 inches;
width of vertebra, 3 inches, which is, however, covered by the expansion of
the rib, included in this case in the length, 16 inches; total width of cara-
pace at middle, 36^ inches; length of carapace, estimated from cranium, 118
inches; or, width, 3 feet i inch; length, 9 feet 10 inches. An outline, twice
as long as wide, is justified in measure by the size, especially the lengths, of the
marginals, which, if placed end to end, would measure on one side of eleven
pieces, if each were as long as the median, S-.O inches X H = 7.8 feet. Some
of the posterior raargiijpls are shorter than 8.5 inches, while some of the anterior
appear to be longer. The length, 8.5, may then be take.i as an average.
But they formed the circumference of an open arc, so the axial length of the
carapace should be placed at a lower figure than the above. Tliis proposition
may be offsetted by the fact that the marginals were not united to each other,
and exhibit no indications of contact. The length of seven feet for tlie cara-
pace is not, llioi), too much, and, estimating from the size of the head, is too

112

little. We can then safely conclude that the carapace of this turtle is more
elongate and narrowed than existing forms. Thus, in Chelone mydas, the
carapace is six-eighths as wide as long.

It remains to discuss the question of the age of the specimen. It migkt
be objected that the absence of carapace, and the radiate character of the
margins of many of the bones, indicate that our type-specimen is young. To
this it may be replied, first, that it is in the (?) sternal bones unlike the j'oung
of any known type, when certain of their bodies do at all times exhibit
smooth margins as boundaries of the points of exit of the limbs ; moreover,
it is possible that these plates were dorsal ; secondly, the superior or inner
extension of the marginals exceeds that of any known tortoise in the adult
condition ; thirdly, the articular bone is ossified ; fourthly, separate ribs should
be discovered among extinct tortoises as an adult character, on theoretical
grounds, the more as it exists in one recent genus {Sphargis, fide Wagler).

Distrihiition. — This fossil was found near
Fort Wallace, Western Kansas. It was entirely
recovered by excavating. The edges of one of
the large bony shields were seen projecting from
a bluff near Butte Creek, and was followed into
the chalk-rock with pickax and shovel with the re-
sult already indicated. The large bones were ex-
posed in an entire condition, but were mucii frac-
tured in the attempt to lift them from their bed.
Though carefully packed, the transport of fifteen
hundred miles still further injured them, and
the portions described were reconstructed of
over eight hundred pieces by myself. One of

Fig. 3. — Costal bouos of a younc ,, , i, ii • , -, r\n •

Testudopoiyphemm from heio^; nat- t^^ bouy plates was broken mto 108 pieces,
""•"i"*^"- the ribs into 183, the marginals into 146, &c.

A second species oi' Protostega^ appears to have existed during the Cretaceous
period, as indicated by a humerus from near Columbus, Miss., sent by Dr.
Spillman to the Academy of Natural Sciences. With it were received bones
of the mosasauroid Platecarpus tympaniticus. Cope; and Dr. Leidy, who
described them,^ regarded all as belonging to one animal. On this basis, he

' Proceedings of the American Philosoiiliical Society for 1871, p. 173.
'Cretaceous Eei)tiles of North America, p. 42 : Smithsonian Contribntions, 18C4.

expressed the opinion llmf ibc fore-limbs of the Pylhonomorpha were natatory.
Thai tliis view wiis correct I provful by study of the skoh'toii of Clulastes pw-
python; and it now appears that the tore-limbs of the latter wert; the (irst
ever described.

The hnmenis of the Mississippi Protostega (see Lcidy, /. c, I'l. viii,
Figs. 1-2) is more elongate than that of tlie P. gigas, is less contracted
medially, and the (great trochanter or) deltoid crest is longer and stouter.
This I called Protostega tuberosa} (Proceedings of the American Philo-
sophical Society for 1872, p. 433.) I also pointed out the existence of a
third species, as distinguished by the form of the humerus, in the green-
sand of New Jersey, which had been referred by Leidy to the "great
Mosasnurys^ From this specimen, Leidy inferred the natatory character
of the limbs of Mosasaurus. The New Jersey species I called Protostega
neptun'ia. A name had l)een already proposed for it; i)ut, as it was unac-
companied with specific or generic description, I did not adopt it. This was
d(me in accordance with the well-known rule that such names without descrip-
tion are useless in nomenclature; and I conceive it to be not only a privilege
but a duty to ignore names put forward in this manner.

The custom of giving generic and specific names without corresponding
diagnosis has only recently been introduced, and has no claims to respect. It
will, if continued, render the science of paleontology .accessible only to a i)riv-
ileged class, who may have control of museums, or who can adopt' a nomadic
lite in traveling from one museni to another. In the case of I lie V rotoMega nvp-
luiiia^ had Professor Leidy, who figured and described the specimen for th(!
lirst time, adopted tiie name already given, 1 should have WM bound lo em-
ploy tlie latter, ascriljing it to Leidy as tlie author; but, as he lefl il among liu;
synonymy of the Mosasauroids, I hav(^ thought il aiUisalile to follow Inni.

PYTHONOMORrHA.

TIh' characters which distinguish this order are the following:
1. The (|iiadrate bone is attarlK^d to the cranium by a giniflymoid articu-
lation, admitting oi' free movement.

^1. The ribs an; attached by simple articulations lo sini,de arlicular i'acets
oi ditipuphyses springing from the bodies of the vertebra;.

' I iiuiilvcrtciitlv wrote I'Uitenirpni tiihiroKiin in tlii' abovi^ essay.
If) V

114

3. There are two pairs of limbs, whicli form packllcs, having the elements
arranged in one plane, and incapable of rotation or Hexure on cacli other.

4. There is no sternum. '

5. The scapular arch consists of scapula and coracoid only.

6. There is no sacrum.

7. The pelvis consists of slender claments, of which the inferior are
nearly transverse, and meet without uniting on the middle line below.

8. The opisthotic bone projects free from the cranium as tlie suspenso-
rium of the quadrate 1)one, and is supported and embraced l)y a pedestal pro-
jecting from the cranial walls, composed of the prootic in fnjnt and the exoc-
cipital behind.

9. The stapes lies in a groove on the posterior side of this suspensorium,
and is produced to the os quadratum.

10. There is no quadrato-jugal' arch.

11. The parietal bone is decurved posteriorly, forming the cranial wall
in front of the prootic.

12. The brain-chaml)er is not ossified in front.

13. The squamosal bone is present, merely forming the posterior part of
the zygomatic arch.

14. The mandible is composed of all the elements characteristic of rep-
tiles: the articular and surangular distinct; the angular represented l)y its
anterior portion only ; and the coronoid present.

15. The atlas consists of a basal and two lateral pieces only ; the odon-
toid is distinct, and is bounded by a free hypapo|)hysis, besides the hypapo-
physis of the axis.

16. The caudal vertebrte support chevron-bones.

17. The teeth possess no true roots.

The free quadrate bone and simple costal articulations at once refer this
order to the Strcptostylicate division of the Repfi/ia, which cml)raccs only the
three orders of Lacertilia, Pythonomorpha, and Ophidia. There are several
characters, however, in which it resembles some orders of one other i)rimary
group, viz, the Synapfosauria, which embraces the Sauropterygin, Testut/i-
nata, and Rhynchoccphalia} In the al)sence of sternum, it resembles tortoises
and Plesiosaurs, and differs from lizards. It resembles the tortoises in the;

' See Procecdiugs of tbo Aiiioiican Association for tUo Advaiicuineut of Sciouce, vol. XIX, p. 2;i3,
1870.

115

posterior decurvature of the upper portion of the quadrate bone, which thus
partially incloses the auricular meatus in a manner not seen in lizards and
serpents. To the orders of the oti^^r primary divisions IcIitJn/opteri/g'ui and
Archosnur'ia, as the Ichthyosauridce and Crocodilia, there; is not the least affinity.

The remaining characters above enumerated ally the Pythonomorpha to
both serpents and lizards. A's there are many Laccrtilla without limbs, and
some serpents with them, their presence in this order is irrelevant in this
connection, especially as the arches supporting them are most like those of
tortoises and Plesiosaurs. In the absence of sacrum, it resembles both the
associated orders, though the same character is universal in serpents, as the
presence of limbs is general in the lizards. The manner in which the opis-
Ihotic bone projects from the embracing bones is a decidedly ophidian feature,
while the production of the exocci[)ital and ])ro5tic is lacertilian. The posi-
tion of the stapes and absence of quad rato-ju gal arch arc characters common
to both orders. The lateral decurvature of the parietal is a character of th(!
Ophidia, and not of the Lacertilla; while the failure of this l^onc and the
frontal to complete the cranial chamber in front is a lacertilian feature. The
composition of the jiosterior part of the lower jaw is like that in the lizards
in the distinctness of the articular and surangular Ijones ; in the presence of
chevron-bones, it differs from snakes; the atlas and axis are those of both
snakes and lizards, and entirely different from those of CrocodUia. In the
absence of true roots of the teeth, these animals differ from all LacerlUki, and
more nearly resemble, without being identical with, the OpJiidln. Thus it is
evident that the Mosasauroids and their allies represent an order of reptiles
distinct from any other, and I have called it Pythonomorpha, from those points
in which it resembles tlx; Opiiidia.

There are innny other eliaracters common to all the known species of
this division, which arc not pn)lial)]y of ordinal cliaractcr, and wliieli T \\\o-
ceed to enumerate. Among tliein will be found some known elsewhere in
the Oplddia, and others which relate them to lacertilian groups.

Craniu7n.

The skull, ill tlie known species of this ordfM-, is wedge-shaped, and
generally elongate. Posteriorly, it presents postfronto-squamosal and paricto-
qualrate arches ; in some spccif^s, also, a malar arch is thouiriit to exist.

Teeth. — These exist in a single row on (he dentary, palatine, and

IIG

itiaxillary hones, aiKi in two rows on llio prcmaxillary. The crowns are simple,
and offer various modifications of" the cone. Their dentinal sul)stance is
confined to the crown, and is attached, at the base, to a pedestal of ostein,
which occupies the alveolus of tlic jaw, and projects al)ovc it. This is
stated, by Cuvier, to be an ossification of the tissue surrounding the circu-
latory vessels and nerves which penetrate to the crown, and which remain
unossified in serpents and most fishes, but are surrounded l)y true dentinal
roots in most vertebrates of the land. Hence, the teeth, in this order, do
not possess tru(; roots. The crowns are covered witli enamel, and their
forms indicate the carnivorous habits of these reptiles.

The premaxillari/ is a narrow, simple element, one-half of a cone
anteriorly, and much attenuated posteriorly, sejiarating the maxillaries above
by the widtli of its spine only. Its extremity projects considerably beyond
the latter. The antcM-ior extremity bears two teeth on each side in tiic
known species.

The maxillary bones are widely separated on the palatal surface, in
front, by the vomers; behind, l)y the nares and palatine bones. They termi-
nate in a narrow j)rocess behind, whose extremity is broken in the specimens
at my disposal, Init which may have supported a malar arch, probably slender,
as in the dolphin.s. The nostrils are linear and superior, and separated by a
septmn composed of coossified na.sal and frontal bones. Tlie nares extend
to a point in advance of the anterior margin of' the orI)it. The prefronlals
are largely developed, and margin the posterior part of the nares. In many
species, their posterior exterior margin projects strongly in tlie plane of the
muzzle, and has caused the orbit to be horizontal, and the range of vision
vertical, as in some aquatic serpents; while, in others, it is decurved, as in
land-vertebrates.

The frontal is a wedge-shaped, flat bone, and presents lateral descend-
ing alae medially. The post-frontals are large, flat, and prominent, and project
beyond the process they send, posteriorly, to join the squamosal. Posteriorly,
they embrace between them a broad, rectangular jn-ocess of the parietal,
which, in Clidastes, contains, near its front suture, the })arietal fontanelle. In
Platccarpus, the foramen is, usually, in or nearer to the suture.

The parietal lias two broad lateral wings, which advance on the frontal,
and form posteriorly the broad anterior margin of the temporal fossa. The
|)ari('tal crests are separfited by a plane whicii is narrowed, or Ihey unite into

117

a median crest po^ti-riorl}'. Two aulero-supcrior ijrojcctioiis ol' the sii|)ra-
occipital embrace the parietal on each side below fhe crest; wliilc il is over- Ij
lapped, just below, by Ihe anterior extremity ol' the pioiitie. l^his does not \^
extend solar lorward as tiie siipraoccipital. In I'ront ol", and below, this
point, the parietal is deeurved, and tbrms a considerable part of the lateral
wall of the cranium, tliongii with but moderate antero-posterior extent.
The lateral wall extends to the body of the sphenoid, where extensive sutural
surface has received if. I can find no suture crossinsi; it; and it is ap|)arently
all alisphenoid or all parietal. A [)art of the parietal is, however, undoubt-
edly deeurved in front of the alis[»henoid. The structure is quite as croco-
dilian as ophidian in this point.

The postero-lateral angles of the parietal s(;nd the parieto-quadrate
arches to the o[)isthotie. which sends an ascending process to meet tho
parietal, as in lizards. It dill'ers from most of tlies*; in the j)resence of an
intermediate i>one, which has been observed l)y Mar.-.ii, and which \\iiuld
a[)pear to Ije a dismemljerment of one of those with which it is in contact.

TIh! anterior ala of the prootk overlaps the alisphenoid largely. Its
posterior lamina may, or may not, meet the expansion of the exoccipital on
I he upper face of the sus|)ensoriuni. Interiorly, it is in contact with the
outer and posterior base of the sphenoid.

The supiaoccipltdl is roof-shaped. The posterior extremity of fhe parietal
rests npou it, sending lateral aniies to tlui opisthotic, as in most Lacerlilla.

The exoccipital is distinct, and bears a very small segment ot" tiie
occipital condyle.

The opislhotic stands obliqnely upward and t<)rward, and furnishes a
glenoid cavity for the articulation of the (piadratum. It lias a process, directed
upward and forward, which occupies a concavity on the inner fiice of the
.squamosal, which has the same direction.

The squamosal is a subrhondjic bone in a vertical plane, and is flat
below, and proximally presents a longitudinal external angle; disfally, it is
slender and [)rolong(!(i, and receives the posterior process of the post-frontal.

The hasioccipital presents a strong transverse condyle. It is a mnssivc
bone, and presents infero-laterally two ])owerful processes, which diverge pos-
teriorly, and present l)road, rugose, ovate tiu'cs of insertion. There is an
obtuse keel on the middle line l)elow, which bifiircatcs posli'riorly to eacli

118

of" the hileml processes. The distal portions of these processes are over-
lapped by corresponding ciiplike processes of the basisphenoid.

Tlie basisphenoid is distinct from the basiocc-ipital, and undrrlaps the
latter almost to its middle. It is longer than l)r()ad, and sends two processes
latero-anteriorly to support the pterygoids. These are not so long as in most
lacertilians. Latero-sujieriorly, it presents a broad surface, on each side of
the brain-case, for support of the upper side-walls. Postero-externally, it
supports the prootic. It thins out anteriorly, and overlaps the alisi)henoid.
The suture for this bone widens anteriorly; inwardly, it is elevated into a
low crest of the sphenoid.

The lircsphcnoid appears to have been distinct ; its base was small ; it
is readily lost, and I have not seen it.

Tiie floor of the cranial cavity indicates that the medulla oblongata
[wssessed the downward flexure characteristic of reptiles ; but it does not
take |ilace till the middle of the length of the basioccipital is reached. The
posterior margin of the sphenoid is marked l)y a deep pit; its median floor is
a transverse elevation ; it then descends again, and terminates in a deep longi-
tudinal groove.

The roof of the brain-case is marked on the parietal bone by two obtuse
divergent ridges, which leave its posterior margin and embrace the fontanelle.
The grooves for the olfactory pedicels are narrow and well separated, but
they unite and are entirely inclosed by inferior processes of the frontal bone,
as in serpents and Varani. These ridges then separate, and leave the bulbi
exposed below. Behind and between the narcs, the median ridge again
appears, separating two strong grooves.

The vomer is divided, and is composed of two slender compressed bones
in contact.

The imlaline hones have a short lateral union at their anterior end with
the maxillaries, and possess a great extension posteriorly, lieing separated from
tlie quadrates by the short pterygoids only. They are free on both sides
behind the maxillaries, and are flattened either transversely or vertically, or
both,- and support a series of strong teeth, generally similar to those of the
jaws. Near their posterior extremity, they send outward and forward a strong
and generally long process, the distal connection of which is uncertain.
These bones are the pterygoids of Cuvier. The true j^terijgoids are rather
short, compressed bones, which are united by suture to the borders of a con-

119

r.avity oi' the palatine. Tlicy arc toothless, and have no sutural connection
with the ossa quadratu. They present no sutural iiicet for a columella; hut
Golclfuss and Marsh believe that the latter element exists. I have not
seen it.

The OS quadratum is a stout bone, and one of the most uniiormly pre-
served among the bones oi fossilized individuals. In general, it forms a halt-
disk ; the convex border thin, one side concave, and the posterior border thick-
ened. The proximal end is produced backward beyond the line of this
border, forming a hooklike process, which is dccurved, nearly inclosing the
auricular meatus. Just anterior to the latter, on the inner face of the thick-
ened portion, is a i)it, which received the end of the stapes or stapedial carti-
lage. The vertical plane of the inferior or mandibular condyle is oblicpie to
that of the superior.

The superior extremity of the us iiuadratum appears to have had consid-
erable motion on th(! opistliotic. Its extent is so much greater than that of
the cotyloid or glenoid cavity, applied to it, as to indicate a gliding motion,
especially as it constitutes an extensive arc, possessing grooves of attachment
for articular cartilage throughout its Icuifth. This arc is ])ent or curved in
the horizontal plane, which would result in a twisting of the os quadratum
round its long axis, should the motion I suggest have taken place. Such a
twist would throw the proximal portion of the ramus of the jaw outward, a
motion quite necessary to the horizontal flexure of the ramus at the splenial
articulation, which no doubt took place in swallowing any large object. The
extent of this outward deflexion of the articular, coronoid, &c., portions of the
jaw, was measured by the outward concavity of the proximal end of the
quadratum. Thus, this is least in M. depressus, and greater in 31. dekayi and
3L maximus (see cuts. Fig. 48); the great projection of the external angle in
lAodon gave that species an excessive power of dislocation, and the same
peculiarity in Clidastes was followed by the same effect.

As the development of processes and ridges on the ossa quadrata differ
in the different species, they may be named as follows : Tlio proximal articu-
lar surface extends over the internal angle and over the upper edge of the ala,
forming the alar 2>rocess (see plates). Below the, meatus and knob, on the
postero-extcrnal margin, there is a ridge, which terminates in a process in some
species, to be called the median posterior ridge. In some, a ridge rises from
the outer angle of the; distal articular face, extending outside the ridge just-

120

mentioned, toward the pit, called the distal internal longltudmal. In Cront of
this, on the inner face of the quadrate, behind or near the origin of <he ala,
may be a ridge called the internal ridge. (See Plate xxxvii.)

The cotylus of the mandible, is also obliqnely transverse ; the iinicr porlioii
deeper, in order to receive the large condyle of tiu- (piadratum

The mandible, of course, partakes of the elongate form of IIk^ cranium.

The fossa t()r tbe temporal muscle is large and deep, but without inner
wall. Tile coronoid process is elevated, convex, and rugose interiorly, and with
a deep longitudinal groove exteriorly. The superior margin of the coronoid
bone is longitudinally concave and obtuse. Tlie dentary terminates in a
peculiar striate plug, posterior to the last tooth. The distal tliird, or les.s is
strongly grooved itir Meckel's cartilage; proximally, this is concealed by (lie
very long laminiform splenial.

The si)leni(d \s largely developed on the inner lace of the ramus, wliere
it articulates by ball-and-socket juint witli the angular.

The angular has a narrow and inferior exposure on the external face of
the ramus, and overlaps the articular by extensive squamosal suture. Interi-
orly, it is a little more elevated, but only opposite to the coronoid bone ; behind
and aljove this it is restricted by the long anterior process of the surangular.
Anteriorly, it is terminated by the squamosal suture of the splenial, just below
the beginning of the coronoid bone.

The articular furnishes the floor of the cotylus lor the quadrat um and tlie
large angular termination of the jaw. Inwardly, it is largely exposed; exte-
riorly, it is extensively. concealed.

The surangular is the largest bone l)ehind the dentary. It is convex
externally, and sends a longitudinal ridge from the cotyhjid cavity to that of
the coronoid, thus inclosing a large shallow fossa. It suppoiis the outer or
vertical half of the articular cotylus of the mandible.

The coronoid is a longitudinal bone attached by squamosal suture only to
the surangular. It is easily separated, and its form diflers in the genera. It
is always more obtuse anteriorly, and more projecting and aliform posteriorly,
where its superior margin is rolled over to the outer side.

In the splenial articulation, the angular bone ])resents the condyloid ; the
splenial, the cotyloid face. The former narrows and retreats upward and
backward. Tlie articulation allows of a rotary motion inward and upward;
the alveolar margin of the dentary bone being thrown upward and (nitwiird.

121

Tliis motion is permitted by the laminar cliaraeter of" tiu' overlapping margins
of" the splenial, etc., as toUows :

The principal Ijody of the dcntary comes to an ol)tiise but grooved pos-
terior termination. Its external wall is prolonged more posteriorly, the inferior
margin fitting a rabbet of the outside of tiie splenial. The superior margin
of this thin plate is nuicli lower than th(; truncate extremity ; and its margin
gradually rises to meet the outer margin of the latter. The section of the
splenial is U-shaped, much thickened at the turn. The inner lamina is more
elevated than tlie outer, and is concave, turning outward above to conform to
the dentary. A narrow laminar prolongation of the articular is observed In;-
tween the folds of the U. An outwardly convex, wedge-shaped termiiuis ot
tlie surangular is included between the inner lamina of the coronoid and the;
onlcr lamina of tlie dcntary, moving liccly on the latter. There is, then,
nothing that prevents tliis Irom being a complete articulation, except the
lamina of the articular, which is about half a line in thickness, and pnil)al)ly
llc.\il)le in lit"e.

The superior margin of the coronoid is convex outwardly, and is not con-
tinuous with that of the dcntary, when the elements forming the splenial
articulation are in line. When, however, the process of the articular is
l)roperly applied to the dentary, and the coronoid and splenial are in line, as
they no douljt were under ordinary circumstances in life, the curvature of th(;
upper margin of the ramus is continuous and normal. At the same time, tiic
splenial articulation is strongly flexed, and the inferior outline of the ramus
angulate at that point.

We have in this feature one of the most extraordinary peculiarities of
(his remarkable order. The mandibular arch, in its usual relations, inclosed
a diamond-shaped area, open behind, the portion anterior to the lateral angles
the longer, and only closed by ligament in front. The structure is an element
of weakness, though, indeed, without such an articulation, such a light and
slender jaw would be particularly liable to iracture. There was, no doubt, a
strong ligamentous union of the parts, as the grooved adjacent margins testify;
but for any supernumerary muscles to flex the dentary bones I can fuul no
provision.

This structure was no doubt designed to effect the deglutiti(m of large
bodies, which would readily pass between tiie expanded mandibular rand.
This lateral extension is necessary to reptiles, which, like the snakes, swal-
IG c

122

low large animals wliole, but arc not furnished at the same time with tlie
arrangement of the suspensorium, the quadrate, and mandible, by which they
unfold downward, thus increasing the vertical diameter of the pharyngeal
cavity. The palatine and maxillary arches not having the mobility seen in
snakes, the mandibles possess increased adaptation to the necessities of doubt-
less similar habits.

The accompanying cut shows the appearance of the normal flexure of
the ramus: c is the splenial articulation ; d, the coronoid process; and e, the
quadrate cotylus.

Fig. 4. — Right inaiulibiilar ramus of CUdastcs pvojxjthon, Coiic, ouc-thiril natural size : a, from tbo
inner side; fc, from above. Fig. .5. — Left ramus of ioxocmiMs htcoloi; a pjtbouid from Central America,
inner side, uatural size. Fig. 6. — Riglit ramus of liryx johuii, Euss., from India, inner side, natural sizo;
1, articular; 2, surangular; 3, angular; 4, corouoid; 5, splenial; G, deutary.

Vertcbrcc and ribs.

Vertebra. — As has been already pointed out l)y Cuvier, the vertebra} in
Mosasaurus flill into cervical, dorsal, sacro-lumbar, and caudal series. The
cervicals are either round or depressed ; they are arbitrarily characterized by
the presence of an obtuse hypapophysis, which has an articular surface for a
separate continuation of the same. The latter may be compared to short,
compressed, ungueal phalanges. The articular extremity of some is nearly
plane; of others, conic, with antero-posterior enlargement. They are directed
posteriorly, and have a broad, obliquely ovate outline on the lateral view.
Their extremities are rugose.

The atlas consists of the three pieces, the l)asal and two lateral. The
axis supports a large odontoid proces*;, which is bounded below by a three-
sided piece, which is provisionally called its hypai)ophysis. Besides these,
he axis has its own (jroper fixed and corresponding free liypapophyses.

123

The dorsals liave no liy[)ai)opliy.sis, and llic diapopliyses decnsasc in ver-
tical extent toward the posterior part of the scries. The greatest variation
is presented by the different species in the long series preceding the caudals,
which do not present zygapophyscs. The posterior of this series are much
shorter than the anterior; the former having the form of the dorsals, the lat-
ter of the caudals. In the Mosasaurus dekayi, M. gracilis, and M. giganteus,
none of these are depressed ; the shorter are subpentagonal in section. In
M. deprcssus and 3f. missuncnsis, the longer are depressed, while the dcprc-.s-
sion of the shorter diminishes I'egularly to the distal caudal series. In M.
hrumhyi, the long vertebrce are flattened to a still greater degree. (Sec
Gibbes' Monograph on Mosasaurus and its allies.)

The caudals are divided into three series by Cuvier, viz : those with sep-
arate chevron-bones; those where the latter arc united to the centrum; and
those without them. Passing posteriorly, these vertebrae become gradually
shorter and more vertically ovate in form. The more posterior are less nar-
rowed in the M. maximus, M. oarthrus, and M. missuriensis {vide Leidy's
work) ; while, in M. dekayi and the Liodons, they are rather more narrowi'd
vertically.

The characters of the diapophyses are marked in different parts of the
column. In all the species of the family, they descend from an elevated posi-
tion on the cervicals and anterior dorsals to an inferior one on the lunibars.
They never spring from the neural arch, as in the Archosauria, but always
from the base of it. On the median dorsals, they originate from the middle
of the side of the centrum, and, on the lumbo-sacrals, from the i)lane of the
inferior surface. They diminish in size, and, as soon as the articulations of
the chevron-bones appear, begin to ascend again. Ou the anterior caudals,
they rise to near the middle of the centrum, and gradually disappear at difi'er-
ent points in the different species.

The chevron-bones are free throughout the anterior part of the caudal
series in the M. giganteus, and confluent with the centrum in the posterior
portions. This is probably the case with many species of the genus. During
immaturity, they may be all distinct in Mosasaurus, while in the genus Liodon
(vcl Macrosaurus) this condition is permanent throughout life, and thus
characteristic. Thoy have two short ha'mapophysial limbs, and a very long,
tapering spine, which is grooved in front. On the middle caudal of certain

124

species it is much longer tliau the centrum and neural spine, and as long as
4.5 centra adjacent. All the lisemal arches are directed ohliquely posteriorly.

Although I do not possess any specimen with complete vertebral col-
umn, an appro.ximate idea of its length nuvy be gained by comparison of parts
which are more or less complete in different species. The cervicals are all
preserved in a s[)ecimen of Cliclastcs propytlion from Alabama, and numbei'
eight,Mncluding atlas and axis. In the type-individual of C. *'^<='WjfAs, seven
may l)e counted; in a Platecarpus ic/ericus, seven; in a Liodon dyspelor, seven;
of which (he last two are without free hypapophyses, while in the two species
preceding, but one without the free hypapophysis is preserved. In Cuvier's
Mosasaurus giganfeus, the dorsals number forty-three; in an undetermined
Liodon, from Kansas (alluded to in a former article as L. lulispinus), there
are preserved the seven terminal dorsals and thirty-two caudals, with <liapoph-
yses, which exhibit little diminution in size; the last with stout but ri'duced
diapophysis. Caudals without diapophyses, in a supposed species of /Vft/e-
carpus, number twenty-seven; and there were at least as many, judging from
the rate of diminution, beyond these. A specimen iti' L7i(/a.st<'.s vipnan'ii has,
according to Marsh, eighty-one with chevron-bones.

The ribs commence at the axis, which bears a small one. There is no
distinct parapophysis ; hence each ril)-liead is undivided, but is Hattened
vertically. The anterior may be known l)y the greater compression of both
head and shaft. The rib of the third vt;rtebra has a narrow, convex, articular
surface, and is concave on the anterior face. Those of the dorsals are much
wider, and with more truncate head.

Tlic limbs and linib-girdles.

The limbs in all the members of this order are very small in jiroportion
to the size of the body and tail, and .the bones of the scapular and pelvic
arches of projwrtionate small development.

The scapula is a broad segment of a disk, differing in form from that
seen in any other order of the Rcptiliu. It only presents facets for the eora-
coid and humerus. The coracoid is a similar bone, but ernijraces a larger

' Professor Marsh (American Journal of Science and Arts, 1872, p. 454) quotes me as assigning ten
cervical vertcbne with articulated hyiiapopby.ses to tliis species. This I have not tlone, but state (Synop-
sis of the Extinct Batradiia and Keptiles of North America, \>. 221) that it possesses six such verteline.
I'rofessor Marsh's statement and eouseriuent sniiposition that he lirst determined the uumher of cervical
vcrtebrie in the genus (.7ii7uste are the result of a misappreheusiou.

125

portion of a disk. In sninc jrciirni. it is dcpply emarginato, as in LacertUia;
l)iit it is usually entire, as in Sauropteri/giu.

'Vhv liumcriis is a small bone, exceedingly wide in ('liddstes and Platc-
carpus, and narrower in Llodoit. It is wider distally than proxinially, lias a
Hat shaft, and presents no condyles, l)ut elongate articular surl'aces only.
The radius is also a wide l)one, especially dilated at its distal and exterior
border. The ulna is nuieh less expanded ; the extremities being subecpial,
and tlie shall contracted, l»ut flat. The carpals are small, ilat, and lew in
number; they are sui)r()und or hexagonal in outline. The phalanges, meta-
carpals, and metatarsals are llattened near the ear|)us, l)ut soon become less
expanded and more cylindric at the extremities and at the shaft. The term-
inal ones are flat.

The ]>elvic elements arc slender, and the inferior but loosely-united on
the middle line below. The ilium is the longest, and is cpiitc attenuated
above, and witliout immediate contact with a vertebra. The pubis is clavate
and flat; the v.'ider portion next the ilium, and pierced with tin; tbramen
observed in LacertUia. The ischium is broader, and has an angulate poste-
rior outline. The femur is equally or more slender than tlie humerus (see
the plate oi' Flatecarjms crassartus), and, in Liodon, resembles it in form. It
is Hat, without condyles, wider distally, and with a trochanteric tuberosity at
the proximal end. The fil)ula is a very wide bone, sometimes constituting
three-quarters of a disk. The tibia is, like the ulna, a more slender bone
than its companion, with contracted shaft, and subequally-cxpar.ded extrem-
ities.

The phalanges much res('nd)le those of marine turtles, and the pes and
manus are of a less rolnist type than in any other order ol' marine p'ptiles.

jtj/inities.

The significance of the ordinal characters lias been already pointed out.
There remain a iiund)erof peculiarities, not certainly of ordinal value, which
are, nevertheless, necessary to consider in estimating the relations of these
reptiles to others :

1. The form and position of tlie coronoid bone are those seen in eryci-
!i)rm serpents.

2. The articulation of the sjjlenial with the angular is only paralleled in
the pvllionoid and allied seiiiciits.

126

3. The extensive freedom of the palatine is less complete than that
of the same element in the Ophidia, and much more so than in most Lncerlilia,
resembling, in this respect, the Varanida.

4. The close articulation of the parietal with the supraoccipital is seen
in the serpents and the amphisbfcnian section of the lizards, but not in true
lizards; wliile the supraoccipital crest is a character of serpents, not of lizards
or saurians.

5. The presence of fronto-parietal fontanelle is very general in Lacertilia
and various extinct saurians, while it is unknown in serpents, crocodiles, and
other groups.

6. The parieto-quadrate arch belongs to several groups of saurians and the
Lacertilia, but not to serpents.

7. The squamosal bone has a similar distribution among reptilian orders.

8. The firm attachment of the maxillary to tlie other bones has the same
significance.

9. The want of sutural symphysis is seen in snakes, the varanian lizards.
Ichthyosaurus, &c. ; but not in Lacertilia generally, Testudinata, Sauro- ■
pterygia, or Crocodilia.

10. The simple premaxillary is general in the streptostylicate orders, and
uncommon in the others.

11. The underarching of the olfactory lobes by frontal laminae is char-
acteristic of some lizards (the Varani) and of all snakes.

The above characters are of very unequal value, and may, in many cases,
be departed from by forms hereafter ihscovered without invalidating their
ordinal relations with the Mosasauridce.

As a conclusion, it may be derived that these reptiles are not nearly
related to the Varanidm as has been supposed, but constitute a distinct order
of the streptostylicate group ; that they are primarily related to the Lacer-
tilia, secondarily to the Ophidia, and thirdly to the Saurop>terygia ; that they
present more points of affinity to the serpents than does any other order ; and
their nearest point of relationship in the Lacertilia is the Varanidce or
Thecaglossa.

Experience in paleontology has shown that generalized orders have been
the predecessors of the special groups of the existing fauna. The structure
of the Pythonoinorpha, which has so much in common with orders well dis-
tinguished from each other, offers a hint of the character of the primary

127

group from wliicli the latter have spninij. That tliis order is not tliat unknown
type is clear; !)ut the indication of alKnity to it is equally unmistakable.

Restoration.

The proportions and appearance of the Tythonomorpha can be determined
from the remains which have been jjrocured. The body-cavity is more
elongate than in any group of lizards and sauriahs, excepting the Amphis-
h(enia, l)ut not so long as in serpents. 'J'he tail is excessively long and
flattened. The head is also long — in some genera, as Clidastes, very slender,
and always flat, with the eyes nearly vertical. There is no distinction between
neck and body, but a contraction behind the head. The limbs arc very small
for the size of the animal, forming broad paddles, with but little peduncle;
the hinder limbs well behind tlie ribs, and often smaller than the anterior.
The general effect of the more slender of llicsc animals was that of gigantic
eels, or some of the snakclike lizards of the present lime, so that they are
veritable sea-serpents of the Cretaceous ocean, and would doubtless be
described as such were any perchance to be found to be still in existence in
the depths of modern oceans.

Several peculiarities affecting the appearance of these animals may be
derived from the peculiar articulation of the lower jaws. The position
of these articulations in advance of the pharynx indicates a baggy extension
of the gular walls, to permit the passage of large bodies between the jaws to
the (Esophagus. This had, perhaps, the appearance of the posterior part
of the pouch of the pelican. This arrangement necessarily requires that the
larynx should be, as in the serpents, in the middle or anterior part of the
mouth ; for, as in those animals, the delay involved in deglutition would cau.sc
suffocation were the glottis immediately below the descending mass. T'his
structure requires another, namely, the anterior position of the tongue ; and
this organ, unless very small, would have to be received into a sheath beneath
tlie larynx and opening anterior to it. Ensheathcd tongues, among reptiles,
tend to become cylindric in proportion to the completeness of the sheathing,
for obvious mechanical reasons. It is almost certain that the I'ytliononwrplia
had tongues of this kind ; for their nearest living allies on l)()tli sides iiave
them, viz, the serpents and the varaiiian or thecagloss lizards. These have
the tongue cylindric, ensheatlied, and forked at tlu^ cud, and project it as a
delicate tactile organ.

128

It, is slated by Professor Marsli that these reptiles possessed scales ; that,
in Clidastes, a complex pattern was ])roduccd l)y alternate rows of scutes
of different shape and size. They are osseous, but thin, and generally united
by beveled edges. Such structures would produce a distinct effect in the
living animal. I had already inferred the existence of dermal scales, and
figured them in a restoration of Mosasaurus, published in the American Nat-
uralist,^ but had not suspected osseous scuta. It is a little singular that I
have never detected them among the numerous skeletons of these reptiles
which I have exhumed in Kansas and elsewhere ; nor have any been sent me
by Professor Mudge.

The proportions varied somewhat in the known genera. Thus, in
Clidastes, we see the greatest attenuation of form ; while, according to Marsh,
the caudal series is less elongate in Liodon. In Clidastes and Platecarpus,
the humerus is very short and wide, broader than the femur. In iodon,
the humerus is more slender, and not very diiferent in proportions from tlic
femur; hence, the flippers were more distinctly pedunculate.

•

Classification and distribution.

The well-distinguished genera of the order known from North American
strata are the following:

I. Cervical hypapophyses separate, articulating:

a. A zygosphenal articulation :

Chevron-bones coossified with centra Clidastes.

Chevron-bones free Sironecfes.

aa. No zygosphenal articulation :

Teeth subcylindric conic; humerus short, platelike;

chevron-bones free Platecarpus.

Teeth mostly compressed, cutting; humerus with nar-
rowed extremities ; chevron-bones free Liodon.

Teeth subcylindric faceted ; chevron-bones in part

coossified . . Mosasaurus.

II. Cervical hypapophyses continuous and (nitire :

No zygosphene Baptosaurus.

The material obtained in Kansas, during the autunm of 1871, by the

writer, proved conclusively that this order of reptiles attained a predominant

'18C0, \^. ai.

129

importance during the Niobrara epoch of the Cretaceous period. This is
indicated by the great profusion of individual remains and specific forms.
Although occurring in America, wherever the Cretaceous formation appears,
they are, so far, more numerously represented in Kansas than elsewhere.
Though not rare in New Jersey, crocodiles and tortoises outnumber them ;
but, in Kansas, all other orders are subordinate to the Pythommorpha. As is
now well known, since 1868,' the seas of the American continent were the
home of this order; while they were comparatively rare in those of Europe.
In the latter country, we have four species only determined by paleontologists,
viz :

Mosasaurus 2

Liodon 1

(?) Saurospondylus 1

In North America, the species have been exactly determined from three

regions, as follows :

Crreensand of New Jersey.

Mosasaurus 6

Baptosaurus 2

Clidastes 2

Liodon 4

(?) Diplotomodon 1

15
Rotten limestone, Alabama.

Mosasaurus 1

Holcodus 1

Liodon 3

Clidastes -• 2

7
Chalk of Kansas.

Clidastes 10

Sironectes 1

Platecarpus - 11

Liodon 4

26

' See Traiisactious of tUe Americau Philosophical Society, vol. XIV.

17 c

130

Wg have additional species from —

North Carolina (Mosasaurus) ■ 1

Mississippi (Platecarpus) 1

Nebraska (Mosasaurus) 1

Making, with the others from —

New Jersey 15

Alabama .• 7

Kansas 26

A total of '. 51

Of these, I am not acquainted with any species which extends its range

into two of the areas above named ; while some of these districts possess

peculiar genera. It is, nevertheless, premature to draw any conclusions as to

geographical range, as most of the species are known as yet from but few

specimens.

HistorT/.

The knowledge of the structure of the animals of this order has been

due, almost exclusively, to the labors of three paleontologists, viz, Georges

Cuvier, Prof 0. C. Marsh, and the writer. Cuvier determined the characters

of the dentition, and of the anterior regions of the skull, and of the vertebral

column. Professor Groldfuss added little to this beyond the description

of the parieto-quadrate and postorbito-malar arches. The writer ascertained

the structure of the posterior part of the skull, including the walls of the

brain-case, the suspensorial apparatus, the pterygoid and palatine bones, and

the median hinge of the lower jaw ; also, of the scapular arch and fore and

liind limbs, except the phalanges. On this basis, the determination of the

affinities of the order was made. Professor Marsh determined, for the first

time, the presence of hind limbs, and described the pelvic arch ; he also

determined the relations of the phalanges of the fore limb, the presence

of two small supernumerary cranial bones, and the probable existence

of dermal scuta.

CLIDASTES, Cope.

Vertebral column exceedingly elongate, the vertebrse united by the
zygosphenal as well as the usual articulation ; the zygosphene elevated but

i;n

little above tlic plane of the zygapophyses. Chevron-bones coossified!
cervical hyjjapophyscs free. Humerus short and very wide. Parietal fonta-
nclle pierced in the parietal bone.

Tiie sijecics of this genus are, so far as known, the most elongate iu the
order. The quadrate bone, in the most typical species, exhibits a very
prominent internal longitudinal ridge, indicating a great degree of rotation
of that bone, and hence external flexure of the mandibular rami.

There are specific differences in the form of the palatine bones in this
genus, from the more transverse or expanded type of the C.propytlwn, to that
of the C. j>la?iif/otis, where they are narrowed posteriorly in some degree, to
the C. tortor, where tjiey are vertically placed in the posterior half. Clidastes
is nearly allied to Platecarpus, with which Sironectes associates it as an inter-
mediate genus. The numl)er of species already known is considerable, and
the genus is divided into sections, for convenience of reference:

A. Centra of dorsal vertebrae depressed:

a. Frontal bones with median keel:

Clidastes tortou, Cope.

A slender species of some thirty feet iu length, w ith a narrow, pointed
head of two and a half feet. Its teeth are compressed, and with a cutting-
edge fore and alt, and were eighteen in number on the under jaw: the palate
was armed with eleven teeth.

The frontal hone is light, and witli thin margins; it is keeled above for
the anterior three-fourths of its length. On its inferior face, the olfactory
groove is closed by the apposition of its lateral bounding ridges. Posterior
to this point, the latter diverge and disappear; and a median ridge, with an
acute edge on each side, carries the deep median olfactory groove, but disap-
pears with the lateral ridges. There is no grooved triangular area in front
of the foramen parietale. The imrietal bone below- presents a V-shaped,
rounded ridge; the limbs embracing the narrow and small parietal fontanelle
in front. The bone, in general, is' broad and expanded laterally. Its antero-
lateral ala is largely underlaid by the postfrontal, and presents a transverse
ridge for its boundary. The superior surface of the bone presents a flat,
longitudinal surtiice; the angles bordering the temporal fbssse being obtuse,
well se|)arated, and low. Bases of slender parieto-squamosal arches project
from it l)eliin(l.

132

The postfrontal is large and jirominent, and terminates in the usual
posterior process, connecting with the squamosal.

The prefrontal is a slender, flat, triangular bone. Its orbital portion
projects at right angles to the orbital margin of the frontals, and is, in its
direction, at right angles to the exterior margin. The latter is in the hori-
zontal plane, and is transversely plicate. A strong process projects inwardly
from the lower side, and is continued across the latter as a curved, flat-topped
ridge, to which the maxillary bone is articulated. It soon reaches the outer
margin of the prefrontal.

The suspensorium is flat (perhaps on account of pressure). Tlie opis-
thotic sends a flat process to the parietal. The squamosal is, as usual, a
sickle-shaped bone, with a flat extremity for articulation with the opisthotic.
Unlike what is observed in Platecarpus and Mosasmirns, it presents no concave
articular face below for the articulation (by ginglymus) of the quadrate.
The exoccipital extends to near the end of the suspensorium, and terminates
in a flat extremity with truncate border. The proiitic, on the other hand,
terminates near the middle of the length of the suspensorium by a transverse
suture. The basis cranii I have not yet found among the debris of this

skeleton.

Quadrate bone with long internal angle, and rather thick anterior ala,
with broad, rugose margin. A prominent, obtuse ridge is continued from the
internal angle to the inferior articular extremity; the distal portion being
more acute. A rugose process projects at the point where the posterior hook
approaches the body, and is continued, as an elevated, narrow ridge, parallel
to the one previously mentioned, to the distal articular surface. A button-
like knob appears on the posterior margin of the hook opposite the meatal
pit. A strong ridge extends, on the outer face of the bone, from opposite the
end of the hook to the base of the great ala. The distal articular surface
presents two planes : the narrower at the end of the posterior pair of ridges
above described ; the larger considerably less distal, like a broad step.

The maxillary bone descends regularly in front, uniting with the premax-
illary by minute suture. Its posterior extremity is slender and acute. The
premaxillary is short conic ; not particularly prominent. The palatine bone
lias a slight expansion on the inner side ; on the outer, the margin is very

narrow.

Tlic teeth number seventeen on the maxillary bones. They are com-

133

pressed, least so anteriorly, and with a cutting-edge from base to crown as far

as the fifth from the front ; in those anterior to that point, the posterior edge

is discontinued. There are sixteen palatine teeth, which are smooth, and

without anterior cutting-edge. The frontal hone has a low carina along the

median line of its anterior portion.

Vertebra of the cervical and anterior dorsal regions with round articular

faces, not emarginate for the spinal cord. The bodies are elongate and

somewhat contracted, and marked everywhere with finer and coarser striae.

Hypapophyses prolonged on the cervicals ; the free one of the atlas with a

j)rolonged keellike process.

Meaiuremente.

M.

Length of tho axis witb tbo odontoid process 0.078

Diameter of the ball of a cervical, vertical 0. 02U

Diameter of the ball of tho same, trausverso 0. 026

lixpause of the diapophj ses of tho same 0,084

Len};tb of the centrum of the same 0. O.Vi

Length of the maxillary bono 0. 3G1J

Length of the ramus mandibuli behind the dentary 0.310

Length of the promaxillary 0.040

Total length of the cranium (2.33 feot) 0.713

Length of tho pterygoid and palatine 0. 315

Length of the centrum of the posterior dorsal vertebra 0.066

Diameter of the ball, vertical 0. O.iy

Diameter of the ball, transverse 0. 038

The bones of this species are all light and slender. The elongation
of the vertebrae indicates that, if their number was of the usual amount, the
animal was of more than usually slender proportions. The position in which
it was found was a partial coil; the head occupying the inside of a turn of
the dorsal vertebra?. As compared with E. dispar and E. ve.lox of Marsh, the
present difi'ers in the lack of depression of the centra of the vertebrae,
especially the anterior, and in some details of structure of the quadrate bones,
as well as tiie larger number of teeth.

Discovered in Fossil Spring Canon, in the gray limestone, by Martin

Ilartwell and Sergeant William Gardner. But one specimen was found, which

includes the greater part of the cranium, with the vertebra? as far as the

lumbar region.

Clidastes stenops. Cope.

Indicated l)y a large part of the skeleton of one individual, and fragments
of two others. The first includes a large part of the cranium, willi l)()11i
quadrates, and fifty vertebra-, including the a.vis. The characters arc similar

134

to those of tlie preceding species; but all the bones are more massive, though
of the same dimensions.

The teeth are strongly compressed with cutting-edge fore and aft, and
with the surfaces distinctly faceted; there are seventeen on the mandible.
The palatine bones are stouter than in C. tortor, but the teeth are not larger,
and are probably as numerous, as they are similarly spaced.

The i)arietal is thicker than in C. tortor, and flat above. The decurved
lateral portions are short antero-posteriorly ; behind these, there is a promi-
nence on the inferior face, which is broken, but probably ends in the acute
median termination of the bone. The apex of an inferior V-shaped ridge is
])reserved. The parts of the frontal preserved show the olfactory groove
nearly closed, and its division in front into two contiguous lateral grooves.
The middle of the upper surface is plane; its anterior part, with a low keel.

The prefrontal is of peculiar form, and displays the greatest difference
from that of C. tortor. Instead of being a horizontal bone, it is so oblique
as to be nearly vertical. From this follows an alteration of the relation of all
the parts. The squamosal suture with the frontal, which is marked by pecul-
iar concentric rugosities in both species of this genus, instead of being on the
upper, is nearly on the under surface, though oblique to both. The lateral
margin is subinferior and plicate ; the crest of the inner side bounding the
maxillary projects far below it in front. In consequence of the form of the
bone, there is less expansion of the face in front of the orbits than in other
species of the order; whence the face is much narrower, and the name stenop!<
is appropriate. The portions of the prefrontal of the other side whicli are
preserved are similar to those described. Tiie characters of the suspensorium
are in the main as in C. tortor.

The quadrates, hke those of the last species, have a very prominent
internal angle. They present various differences, which may be regarded as
only individual: for example, the edge of the great ala is not expanded
inward, but only outward; the distal articular extremity is wider; the poste-
riorly decurved hook is more contracted, forming a deeper internal concavity
behind the internal angle. The button on the posterior aspect of the hook
is wanting; its place being taken by a recurvature of the smooth articular
face along the margin. Characters of more importance are, the lack of
the two ridges which bound the posterior face of the distal end of the
bone, that face being thus convex instead of concave ; and the process below

135

(lie meatus is isolated, and not continued into a ridge, except externally, where

it gives rise to the heavy ridge which extends to the base of the great ala.

Tlie vertebrte exhibit round articular surfaces ; those of the dorsal region

being rather stouter than the cervical, though the difference does not appear

to be so marked as in the preceding species. The anterior caudals possess

wide diapophyses. The articular faces are a vertical oval, a little contracted

above, sometimes by a straight outline. They preserve a peculiarly elongate

form.

Measurements.

M.

Length of the axis (aloiio) 0. OCO

Diameter of the ball, vertical , 0. 0ii7

Diameter of the ball, horizontal 0.027

Length of a posterior dorsal 0. 000

Diameter of the ball, vertical 0. 03;{

Diameter of the ball, transverse 0.038

Length of the caudal with the flat diapophysis 0. 033

Depth of the cnp of tho caudal 0.031

Width of the cup of the caudal 0.030

Length of the mandible (28 inches) 0.720

Depth at the coronoid process 0. 150

Depth at the proximal end of the dentary 0.074

Depth at the distal end of tho dentary 0.030

A fine specimen of this species was found by Martin V. Hartwell near
Fossil Spring; and portions of a second were found by Lieut. James H.
Whitten, on a bluff on Butte Creek, during my expedition of 1871.

aa. The frontal bone without median keel:

Clidastes planifrons, Cope.

A large species, represented by large portions of the cranium, including
quadrate bone; by cervical and dorsal vertebras, and fragments of other ele-
ments, all belonging to one individual. They are well preserved, and have
suffered but little from distortion.

The frontal hone is especially massive, and is plane on the supericM-
surface. The superciliary borders are strongly concave, a feature either littU;
or not at all marked in other species known to me. It is thickened ; but the
(bssa for the postfrontal bone extends far toward the front and middle on the
inferior surface. In front of the prefrontal angle, the frontal contracts, nar-
rowing regularly to the line of the nares. The prefrontal has the remarkable
form characteristic of Clidastes stenops ; that is, with the exposed face subver-
tical or steeply roof-shaped, instead of horizontal. A groove descends on

136

each side to each nareal orifice, and the intervening longitudinal ridge is deeply
fissured hy a parallel groove. The parietal fijntanellc is entirely in the parietal
bone. The postfrontal is massive.

The quadrate bone presents a very prominent internal angle, as in other
Clidastes, and has the posterior hook much prolonged downward and inward,
with a button and surrounding groove on the inner side of one, but none on
the other. The stapedial pit is a narrow oval, as in Lwdon proriger. The
median posterior ridge is prominent, and united with the distal internal longi-
tudinal, extending to the narrow posterior angle of the distal articular face.
The internal ridge is prominent, dividing the internal face of the bone into
two planes, the posterior of which is but little concave along its upper poste-
rior border {i. e., on the inner side of the hook) ; this region is very concave
in some species of the genus. There is a strong transverse obtuse ridge, which
extends along the outer side, turning backward into a rough process opposite
the origin of the base of the ala. Between this and the distal articular face
is a subtriangular rugose area. The palatine bone has its anterior and poste-
rior extremities broken away, the fragtnent supporting six teeth. The bone
is flat, much as in the species of Plalecarpus, the tooth-line passing from the
inner margin behind to the outer before, the roots being more exposed on the
external side ; the external process is stout. The crowns of the palatine
teeth are curved, with lenticular section, one face being much more convex
than the other ; the enamel is shallowly striate-grooved.

The articular faces of the cervical vertebrce are all transversely oval, not
much depressed ; those of the dorsals are also transverse, but less so than the
cervicals. Five cervicals and nine dorsals are preserved. The hypapophyses,
both fixed and free, are very large and stout. The odontoid is large and prom-
inent, and deeper than long. The diapophyses are short, and send a narrowed
extension forward to the rim of the cup on all the cervicals and three dorsals.
The vertical portion of their surfaces diminishes anteriorly as the horizontal
extends, till, on the axis, it is horizontally subtriangular in outhne. . The
zygosphen is smaller on the anterior than the posterior vertebra; ; on the
latter, the zygantrum possesses special facets for it. The cups, especially
of the dorsals, are emarginate for the neural canal. A smooth band borders
tli(; circumference of the ball in front. The surfitce, in general, is smooth,
will) rugose lines and grooves extending to the articular face of the fixed hypa-
[M)pliysis and ;i|)c.\ of flic iVcf, and on I lie upper rool-likr smlacc ol ilir [mjs-

137

tcrior zygapopliyscs. The inferior surfaces of the centra display a more or
less prominent longitudinal median ridge.

Measurements.

M.

Length of the frontal bono to tho narea 0. 150

Width of tho frontal bono i)08tcriorIy 0. 110

Width of the frontal bono at thu orbits 0. 104

Width of tho frontal bono at tho uares 0.040

Width of tho palatine at the third tooth in front of tho transverse process 0.037

Width of the palatine just behind tho transverse process 0.035

Depth of the mandible at tho cotylus 0.040

Depth of tho splenial condyle 0.032

Width of the proximal articular surface of the quadrate (transVerse) 0.023

Width of the distal articular surface »*^ fjiri"^

Length of the third cervical centrum 0. 0G2

Depth of the ball of tho third cervical centrum 0.030

Width of the ball of the third cervical centrum 0.037

Length of tho articular end of the fixed hypapophysis 0.025

Length of tho free hypapophysis 0.029

Length of a median dorsal centrum 0.072

Width of tho ball of a median dorsal centrum 0.048

Dei)th of the ball of a median dorsal centrum 0. 040

This species need only be compared with the Clidastes stenops, Cope,
which exhibits the same peculiarity of roof-shaped prefrontal bones. That
species has the cervical articular faces entirely round ; the frontal bone is
keeled in the middle, and the palatine much more vertically comjiressed.
The quadrate bone differs in various respects ; among others, in the round
form of the stapedial pit. As compared with the species described by Pro-
fessor Marsh as Edestosaurus dispar and E. velox, it differs in the form of the
(piadrate, which, in these species, is much as in C. tortor and C stenops; i. e.,
with sliort proximal hook, oblique inferior articular surface, round pit, &c.
In this species, the quadrate is truncate distally, &c. This fine s^jecies was
discovered by the veteran geologist, Prof. B. F. Mudge, during his annual
expedition of 1873.

AA. Centra of anterior dorsals compressed.

Clidastes cineeiarum. Cope.

The largest species of tliis genus, as indicated by the zygosphen articu-
lation of the vertebrae.

The region whore it was foimd is (he same as the last, but the speci-
mens were taken from I ho gray bed, perhaps thi^ santc that ]ir<tdiiced llie
Elasmosauius platijurus, Copo. They consist of vcrtcbriu and iilcryiroid
18 c

138

teeth. There are two anterior dorsals, three lumbars, and one caudal. The
articular faces of the caudals are broad vertical ovals. They increase in
width on the lumbars till, on the last of these, they assume the subpentagonal
form characteristic of many species, and which is still more marked on the
caudal. The centrum of the anterior dorsal is much compressed; inferiorly,
slightly concave longitudinally, regularly and prominently convex transversely.
Conversely, the rims of the cup and ball are strongly expanded; the latter
with surrounding groove. The diapophyses of the lumbars are of consider-
able length, exceeding, in this respect, those of Mosasnurus we possess where
•^h^'Ji''" parts are preserved. On the median of the lumbars, the inferior sur-
face of the centrum first becomes truncate or plane, and separated from that
below the diapophyses, which become slightly concave. The expansion of
the ball becomes more abrupt and striking on these vertebrae. The caudal
is a little more compressed than the lumbars, and presents the character of
coossified chevron-bones. These are slender and longitudinally grooved.

A single pterygoid tooth was found in the matrix on one of the dorsals
The basis is short and much swollen ; the crown curved, acute, a little com-
pressed, and with an obtuse cutting-edge posteriorly.

Measurements.

SI.

ADterior dorsal, leugth of tbe centrum 0. 0608

Anterior dorsal, depth of the articular ball 0. 038

Anterior dorsal, width of the articular ball 0. 038

Anterior dorsal, diameter behind the diapophyses 0. 021)

Anterior dorsal, depth of the articular face for the rib 0. 032

Lumbar, length of the centrum 0. UGO

Lumbar, depth of the ball 0.037

Lumbar, width O.O.TJ

Lnmbar, length of the rerau.ant of the diapophysis 0. 046

Lumbar No. 2, length of the centrum 0. 055

Lumbar No. 2, width of the zygosphen 0. 0182

Caudal, length of the centrum 0.041

Caudal, depth of the cup 0.040

Caudal, width of the cup 0.040

Caud.al, width of the basis of the diapophysis 0.0245

Caudal, width between the chevron rami 0. OUu

Pterygoid tooth, height of the crown 0.0125

I'tcrygoid tooth, diameter of the pedestal 0.013

This species was found by Professor Mudge near the locality of the
Platecarpus mudgei, six miles south of Slieridan, Kansas.

It is only necessary to compare this species with C. intertnedius, Leidy,
as the G. iguanavus and C. propython have depressed vertebral centra. Those
of 1li(^ first are rounded ; of the present one, compressed. The C. intermedius

139

also agrees with the two others in the obliquity of the articular faces to the
vertical transverse plane of the centniin ; in the present species, these planes are
parallel. This species is also larger than the C. iguanavus, Qo^^e; the C.
intcrmedius is smaller.

A smaller specimen, apparently of this species, was obtained by Profes-
sor Mudge during his expedition of 1872. The small distal caudals exhibit the
coossified chevron-bones, and the articular faces broader than long; the ante-
rior caudals do not exhibit the two angular ridges of the lower side, seen in
so many species. The lumbars are short, and the dorsals rather elongate,
with the slightly vertically-oval articular faces and contracted sides of the

centrum.

Measurements (No. 2).

LeuKth of the centrum (with ball) of the posterior dorsal 0.060

Depth oftbe cup 0.042

Width of the cup "-O^O

Leusthofalnmbar "-040

Depth of the cup "• f ^^

Width of the cup 0-0^7

Leugth of the caudal without tho diapoijhyses 0.021

Depth of the cup 0.024

Width of the cup 0.028

Length of a more distal caudal - 0. OUt

Depth of the cup O-O'^Jl

Width of the cup 0.022

This is the specimen mentioned under the head of Plesiosaurus gulo as

having been probably swallowed by the latter. Found near Sheridan, Kans.,

ill the graj' shale.

SIRONECTES, Cope.

The characters of this genus are such as to unite closely that wliicli
precedes it with that which follows it in the present enumeration. It is
more nearly aUied to Platccnrpus in the only species known, where the
zygosphen is weak, but articulates with special facets on the lateral walls
of the zygantrum, which are not known in that genus. The form of tlie
bones of tlie liiuljs is unknown.

SiRONECTES ANGULIFERUS, Cope.

Established on a portion of the left mandible, with a series of thirty-one
vertebioe, of a single individual, discovered by Prof B. F. Mudge in the gray
calcareous shale of Trego County, Kansas.

Some of the vertebrae have suffered from pressure; but the centrum

140

of an anterior cervical is little, or not at all, distorted, as are also many of the
caudals. All the dorsals and cervicals have transversely-oval articular faces,
openly notched above for the neural canal. The fixed hypa]3ophyses arc
large ; the last one small and subconic, abruptly following a large truncate
one. The three succeeding dorsals are keeled below, the keel of the last
low and obtuse. The zygosphen is weak, and deeply notched in the middle.
On the anterior cervicals, it is rudimental; but, on the dorsals, supports a well-
developed articular facet, which meets a corresponding one of the zygantrum.
The fixed hypaj^ophyses and roofs of the posterior zygapophyses are rugose,
with grooves and ridges. The articular faces of the caudals are broad,
vertical ovals as far as the specimens extend, the series including only a part
of those with diapophyses. On the anterior caudals, the chevron-facets are
compressed. The neural spines are thinned out in front; obtuse at tiic base
behind, but expanding to a thin edge there also. The sides are longitudinally
grooved. Diapophyses on the middles of the sides. Some ribs have tiie
heads not expandad but truncate. The angle of the mandible is produced
backward, and below the plane of the lower margin of the ramus, in a marked
manner. The lower margin and the surface next the smooth edge are

rugose.

Measurements,

M.

Depth of the mandible at the posterior margin of the cotyliis 0. 058

Length of the centrum of the anterior cervical 0. 070

Diameter of the ball, vertical 0. 028

Diameter of the ball, transverse 0. 040

E.\i>an8e of the diapophyses of the anterior cervical 0. 09.")

Expanse of the anterior L'ygapophyses - 0. 064

Expanse of the diapoi)hy8es of the anterior dorsal 0. 115

Length of the centrum of the anterior dorsal 0. 071

Length of the centrum of the anterior caudal 0. 053

Depth of the cup of the anterior caudal 0. 044

Width of the cup of the anterior caudal 0. 041

Width of the neural spiue 0. 03;i

Elevation of the neural si)ino (apex lost) 0.070

Diameter of the centrum of the twenty-fourth caudal, vertical 0. 039

Diameter of the centrum of the twenty-fourth caudal, transverse 0.040

Length of the centrum of the twenty-fourth caudal 0. 042

This species appears to have had proportions not unlike those of Plate-
carpus coryphceus ; the specimen described being larger than that on wliich the
latter was based. It is also ratlicr larger tlian the Clidastes planifrons, Cope,
llic largest of its genus, Init which, since its caudal vertebfie are unknown, may
yd Ik; I'ouiid to he a Sironectes. It differs specifically fruiu the S. angullferus
ill llic less (Icscliipinciit of I he zygi)s[>iicii, especially im I he aiilciior verlebtje,

141

and its deep emargination in front, where well developed. From its general
characters, I anticipate that the quadrate bone of this species will prove to
be more like that of the Platecarpi; that of G. planifrons is that of the genus
to which I have referred it.

PLATECARPUS, Cope.

Vertebrae very numerous ; caudal series very elongate ; zygapophyses
strong; zygosphen wanting, or very rudimental. Chevron-bones free. Teeth
siihround in section, acute, and curved. Humerus short and wide.

While the form of the humerus in this genus is that of Clidastes, the
vertebral articulations are those of Liodon, the zygapophyses being, however,
stronger. The teeth are different from either.

Besides the characters assigned to this genus in the analytic table
already given, Platccarpus is characterized by the position of the fronto-
parietal fontanelle on, or very close to, the coronal suture, instead of in the
broad plate of the parietal bone posterior to the suture, as in Clidastcs. The
genus is also characterized by the form of its teeth, which are neither com-
pressed, as in Liodon, nor broadly, angularly faceted, as in Mosasaurus (and
Holcodus, fide Marsh), but are curved and subcircular in section. The
exposure of the roots of the palatine teeth is largely less, or scarcely greater,
on the outer than on the inner side. The form of the cranium is, in this
genus, less elongate than in Clidastes, and the muzzle is often quite short
and obtuse. The caudal vertebrae of the type-species P. tympaniticus are
unknown; but its quadrate bone and the forms of its cervical vertebrse and
palatine teeth are quite similar to those of the other species here referred to
Platecaipus. Professor Marsh has referred the same species to a genus,
Leslosaurus, Marsh, of which the species L. simus, Marsh, is regarded as type,
and which he regards as nearly coextensive with the genus I previously
de6ned under Dr. Gibbes's name Holcodus. He finds the coracoid of L. si?/ius
to be deeply incised, as is often the case in Lacertilia.

The name which I formerly used for this genus was originally applied
by Dr. Gibbes,^ of Charleston, to a species represented by teeth from the
Cretaceous of Alabama, but of which no other portions were known. Tiie
teeth of the Kansas species referred to it are somewhat similar in character
to tiiose described by Gibl)os; but it is evident that the latter belonged to an

' Oil Mosasam-us and aUies, in Smithsnniiin CtintriUutiona to Knowlculgp, IH.'il, ; plntus.

142

animal more nearly allied tothc true Mosasauriis. The place of P/aiecarpus
is evidently between Clidastes and Mosasaurus; the palatine bones being those
of the former, and the vertebral articulations being identical with that charac-
teristic of Mosasaurus. In all of the sjiecies, traces of the zygosphen appear;
but in the H. coryphcRus, Cope, the rudiment amounts to a short process
directed forward at the base of each anterior zygapophysis.

The species known, as yet, are of medium size in the order.

The species P. mudgei and P. tectulus resemble each other in the form
of their quadrate bone, and are referred to this genus provisionally only.
The P. simus, Marsh, resembles the P. crassartus, Cope, but differs in the
depressed, instead of circular, articular faces of the dorsal vertebrae.

It is probable that this genus had a considerable geographical distribu-
tion. P. tympaniticiis has been found in A.labaraa ; and I suspect that other
species from other localities belong to it.

Platecarpus coryph/Eus, Cope.

Characters. — Cervical and dorsal vertebrae with the articular surfaces
depressed transverse, slightly excavated above for the neural canal. The dia-
pophyses not continued inferiorly to the rim of the cup on the cervical ver-
tebrse, and not receiving from it a cap of articular cartilage. Occipital crest
much elevated. Quadrate bone small, the meatal pit depressed between
bounding ridges above and below. Rudimental zygosphen not uniting into a
keel above. Teeth slender, less curved than in P. ictericus.

Description. — This species is chiefly based on one specimen, which
includes the greater part of the cranium, and seventeen vertebrae, with ribs.
Isolated portions of other individuals were also found in the same region of
country.

The disproportion between the diameters of the cervical and dorsal ver-
tebrae is more marked here than in the species of Clidastes. The centra are
also less elongate, though with larger diameter. The cranium is relatively
much smaller ; the teeth absolutely smaller, though the quadrate bones are
of equal size. The genei'al character of the species is stouter, but less
strongly armed, and less elegantly built.

The hypapophysis of the atlas has a short small keel below. The neu-
ral spine of the axis is elongate, but less so than in the two Clidastes, trun-
cate behind, with a median groove, into which the anterior kocl of the neurnl

143

spine of tlic lliiid fcrvical vertcljra is applied. Tlic diapopliysis of this ver-
tebra has a short vertical articulating surface, and is continued into a longi-
tudinal keel, which disai)pears before reaching the edge of the cup. The
same process of the axis has a longitudinal parallelogranimic articular sur-
face.

The supraoccipital is very thick, and is roof-shaped, the keel rising
nearly perpendicularly from the foramen magnum. The suspensoria are
directed both upward and backward, at about an angle of 45° in each direc-
tion, and support, on their extremities, the squamosal bones. These are pro-
longed, each forming a part of its appropriate arch. Tlie occipital condyle is
transversely oval. The sphenoid bone embraces, as usual, the basi-occipital
protuberances ; it is not carinate on the median line below. It sends out, on
each side near the anterior extremity, a subhorizontal, laminar process.

The quadrate hone is much like that o{ P. ictericus, but is relatively
smaller. While the teeth in that species are smaller, the quadrate is larger;
hence, the difference in the species is, in this point, quite striking. The
internal angle is prominent, but very obtuse, and is the summit of a very
thick, obtuse ridge, which extends to near the distal articular face. Tbe
posterior hook is much prolonged downward, and has no buttonlike process
or extension of the articular surface on its posterior face. This face presents
a strong rib along the meatus, ahd, disappearing above the pit, throws the
latter into a depression. This is increased by the swelling of the internal
angular rib. A prominent knob, very rugose at the extremity, rises beneatli
the end of the hook, and bounds a concavity between it and the internal rib.
The latter closes the concavity by curving round toward the knob above
mentioned. A keel rises interior to the rib, and below it, and continues into
the internal angle of the articular extremity. Another very prominent keel
extends from the knob beneath the hook to the base of the great ala. The
articular extremity is transverse, and in one plane.

The maxillarij bone is marked with shallow longitudinal grooves. It
supports eleven teeth, and has a rather steeply-descending premaxillary
suture in front. The nareal expansion in front occurs opposite the fourtli
tooth.

The teeth are rather long, slender, and incurved and recurved. There is
a distinct cutting-edge anteriorly, and on a greater or less part of the length
of the posterior face. Tlio crowns are four or five faceted on the outer face:

144

the inner face is more numerously faceted, and striate-grooved. The section
at the base is subcircular; higher, the outer face is iiatter, the inner more con-
,vex. The apex is acute, and the cutting-edges strong.

The frontal is narrow, and differs from the other Platecarpi here de-
scribed in having the olfactory groove closed by contraction behind. Both
palatines are preserved. They support tw^elve cylindric conic teeth, which
have recurved apices and striate enamel. The section of the bone is a flat,
transverse oval, where the external transverse process is given off". The shaft
of the bone is much expanded inwardly, with a thickened margin ; exteriorly,
the margin is thin, and is nearly followed by the series of teeth whose bases
are exposed externally, and are therefore pleurodont. The emargination for

the pterygoid is very deep.

Measurements.

M.

Length of the axis with the odontoid 0.074

Length of the third cervical 0. 048

Diameter of the ball of the third cervical, vertical 0. 031

Diameter of the ball of the third cervical, transverse 0.033

Elevation of the spine of the third cervical from the centrum 0.046

Length of the posterior dorsal 0. 068

Diameter of the centrum, vertical 0.033

Diameter of the centrum, transverse 0.048

Length of the basioccipital and basisphenoid 0. 084

Elevation of the occipital crest above the floor of the foramen magnum 0. 030

Length of the suapensorium from the foramen ovale 0. 090

Length of the os quadratum 0. 073

Width of the distal extremity 0.036

Length of the os maxillare 0.210

Depth of the os maxillare at the third tooth 0. 036

Length of the fourth tooth 0.032

Length of the crown of the fourth tooth 0. 021

Length of the pterygoid bone 0. 155

This fossil was foftnd by the writer projecting from the side of a bluff
in a branch of the Fossil Spring Canon, near the mouth of Fox Cafion. The
l^luff" was from eighty to one hundred feet in height; and the Platecarpus v/as
taken from a position forty feet below the summit, from the yellow chalk.

Platecarpus ictericus, Cope.

Characters. — Internal angle of the os quadratum close by the meatus, and
continued as a rounded ridge separating the anterior and posterior internal faces
of the bone. Median posterior ridge not prominent. Centra of dorsal ver-
tebras depressed. Humerus broad, short.

Description. — This species is represented by portions of a cranium, as post-

145

frontal, suspe.usorial, pterygoid, articular, and qiiadialc Ikhics; hy parls or
wholes of" several vertebni;, which are all dorsals; and Ijy soapida and cora-
coid, with many elements of the fore linil). The latter include luimenis,
radius, a carpal, and numerous metacarpals and plialansrcs.

The species is tirst well characterized by the form oC tin; quadrate bone.
The specimen lacks a portion of the ala, and the postero-superior decnrved pro-
cess, but is otherwise perfect. Its form is intermediate between that in Liodon
vaiidus, Cope, and Mosamurus depressu,s, Cope. Its internal any;le of the
proximal extremity is posterior to its usual position, as in tlio former species,
but is less ])r()niinent than in it. It extends to near the distal end, disappear-
ing between tiie extremities of the median posterior and the distal longitu-
dinal angles. Tiie former of these is short, and it disappears l)y a gradual
descent distally in a very rugose margin. The distal longitudinal is short
and acute, not prominent at the distal extremity. From the posterior posi-
tion of the proximal internal angle, the alar articular surface is somewhat
elongate. The postero-internal face above the meatus is pro[)ortionateIy
short. The meatal pit is scarcely one-fitth the usual size, so far as determin-
able from the present surface; but it is |)ossible that tin; greater part is filled
by an impacted mass of bone derived from the adjacent ridife. The margins
of the articular extremities and of the ala are striate and papillose ruyose.
No meatal knob.

The sus{)ensorium is slender. It is peculiar in the great extent of the
exoccipital element, which covers the whole superior surfiice, and extends
externally over the opisthotic to the squamosal, concealing the former, excejit
its anterior margin. The prootic sends a small proximal portion only to the
superior face.

The pahitine has been free from its fellow medially, A distal and
median portions have hecn lost ; the remaining fragments presents i)a.ses and
alveolse for eleven teeth. Tiie fangs are rugulose and I)ut little swollen ;
probably five to .seven .«tood on the lost portions. The i)ases of the crowns
are circular. The external process of the bone is slender and flat.

The portion of the mandible preserved includes much of the articular,

and adiicreiit parts of the angular. The latter forms a narrow band on the

lower edge of tlie external face, and one twice as wide on the inner face.

The only characteristic feature is the lowness of the ridge which descends

19 c

14G

aiul extends iiuteiiuily tVoiii tlie uiilcrioi- uiargin of the cot} Ins I'ur I he qua-
drat um.

Of the vertebrge, several are so distorted by pressure as to be uncharac-
teristic. Two well-preserved anteiior dorsals have transversely oval articuhir
surfaces excavated openly above I'or the neural canal. One is from a position
anterior to the other; and these surfaces are less oval, (hough still transverse.
The centra of both are very concave in |)rolile l)el()w, and expand botii infe-
riorly and laterally to the edge of the cup A deep groove surrounds the
base of the posterior face. In the anterior dorsal, the neural arcli is presiMved.
It exhibits an approach to a zygosphen articulation, more marked than in any ■
otiier Pldtccarpus, and is hence nearer Sironecte.s in tiiis respect, as well as in
the slender pterygoid. A zygosphen is not separated lioni llie zygapophyses,
owing to their connection by a lamina of bone. The notches at the posterior
end of the arch for tliis prominence are marked. The neural spine had a
long anterior ala, the i)ase of which extends 1o tlie summit of the neural arch.
It presents a fine slriation vertical to the centrum and oblique to the edge
of the bone, as is seen iu C. propython, Cope. The diapopliysis on this ver-
tebra looks obliquely upward, and carries a vertical articular surface, which is
concave behind. The line of its lower extremity tiiUs the depth of the neural
arch below the latter, and of its upper reaches the apex of the canal in front.
The more posterior vertebra has, as usual, a broader articular rib-surface, the
diapophysis being flattened above and below. The marginal and angular sur-
faces are striate-rugose on these and the other vertebrsTe. One of the free
hypapophyses of a cervical is preserved. It has a subtrigoual section, and is
longer than wide, and obtuse. Its posterior faces are exceedingly rugose.

A cervical rib is compressed and short. Head narrow, large, simple ; the
adjacent sides striatc-rugosc. Sides with a shallow groove.

The scapular arch is represented by an entire right scapula and proximal
part of right coracoid. The former is broader than in any of the species in
which I have seen it, and is Hat and thin above. Its anterior extension is
greatest below; its posterior above, at the superior angle. The lower poste-
rior margin is strongly concave and thickened. The antero-superior margin
is a regularly convex arc of more than 180°. The lower portion in front
is on a different plane, and is the rudimental acromion. Tlie articular surtiuc
is rugose, and the glenoid cavity not less so.

The proximal portion of the coracoid is flat. It presents the usurd fo; a-

147

men iieir tlic antt'rior iiiiugiii; and tin- shorter concavity of tlu; aiiti'rior margin
leads to the belief that tin; aiit(!rior extremity of the bone is the more pro-
longed, as in Clida^tes 2»'opytho7i.

The glenoid cavity is not concave, but merely two adjacent llattened
rHgose surfaces.

Consequently, llic liuments has no head, i)ut merely an elongate articular
surface, which e.\hil)its a median keel and a shoit angular expansion near the
middle. This \nnw. is of remarkable form, more resembling that I have
descril)ed in C pro2)ytfion^ than any other, and very diiTerent from that
described in Lindoii diispelor. It is a broad Hat bone, expanded at the
extremities, and in one iilaiir distally, so as to be as wide as long. In the
present individual, it is crushed by pressure, so that its thickness is not
readily deterininable. Its external surface rises into a crest medially at the
narrowest portion, which continues to the lateral angle of the proximal end,
following parallel to one of the l)orders. A moderate thickening exists on the
opposite side, a little beyond the extremity of the crest. Strongly rugose
striae extend to the edges of tlie articular faces. An oval rugose muscular
insertion exists on the least prominent of the distal angles, and not on a
process, as in C propythoii.

A bone, whicli, from its analogy to the radius of the last-named species
I supjiose to be that bone, accompanies the others. It is tlat, truncate prox-
imally, and with nearly parallel borders on the proximal half. Distally, it is
obli(|uely expanded ; the outline forming a segment of an ellipse, whose axis
is oblique to that of the bone. Its extremities are rugose-striate.

One carpal remains; it is a quinquelateral bone, one side being marginal
and concave. Perhaps it is the intermedial. There are several elements,
which arc pro!)ably metacarpals. Tiie general struclnrc of the wiiolc liml)
may be determined from these and from the numerous ])i'ialangr.!<. Tiic i()rmcr
are flattened and with oblique extremities; the latter more cyliiidric, with a
transverse truncation. Both have a median contraction, which becomes less
marked in the distal ones; these are also more cylindric, entirely so at the
distal extremities, whicli are concave. All of these elements are rod-like,
much more slender than any of those figured by Cuvier or Leidy. Tliose
immediately following the metacarpals are flattened, i>ut thickened distally.

The number of disds cannot he readilv determined, I)ut i()ur nuiy i)e

' Sec TrniiRactioiiB of thu American Philosopliical Society, 18fi9, p 219, Table XII, tig. 17.

148

certainly clistinguislieil. The genenil similarity in construction of the ninims
to that of a cetacean mammal is noteworthy.

Measurements.

M.

Length of the suspensorium anteriorly 0. 1 1 1

Width of the suspensorium, medially 0.031

Quadrate, greatest length 0. 09'J

Quadrate, width of the ala 0. VCM

Quadrate, thiclcness behind 0. o;!0

Quadrate, length of the distal extremity 0.043

Palatine, length of six alveoli 0. 0."-.&

Anterior dorsal, length of the centrum 0. 050

Anterior dorsal, width of the eup 0. 0,')1,''>

Anterior dorsal, dejith of the cup 0. 03rt

Anterior dorsal, expanse of the posterior zygapophyses 0. 03yr>

Anterior dorsal, expan.se of the diapophyses - 0. OUl

Anterior dorsal, width of the neural canal 0.0135

Anterior dorsal, deptli of the neural oaual 0. Oil

Posterior dorsal, depth of the ball 0.049

Posterior dorsal, width of the ball : 0. OJ-2.^)

Posterior dorsal, length of the centrum 0. OS.':^

Posterior dorsal, expanse of tlic diapophyses 0. 088

Scapula, length 0. 14.''>

Scapula, width, proximal 0. 070

Scapula, width, median 0. 112

Coracoid, width, proximal 0. (KiCi

Humerus, length 0. l.'>4

Hnmerus, width, proximal 0. Hi)

Humerus, width, median 0. (I7.>

Humerus, width, distal (restored from C. propijlhon) 0. 15M

Radius, length 0. 111".

Radius, width, proximal 0. 001

Radius, width, distal (oblique) 0. 105

Carpal, length 0. 040

Carpal, width 0.037

Metacarpal, length 0.005

Metacarpal, width proximally 0.045

Metacarpal, width medially 0.018

Metacarpal, widih distally 0. 034

Phalange (medial), length 0.085

Phalange (medial), width proximally 0. 027

Phalange (distal), length 0.059

Ph.alange (distal), width dist.illy 0.0082

Ramus niandiljuli, dei>th in front of the cotylus 0.056

Cervical rib, lengtli 0.074

The total lenylli of the anterior limb could not have been less than 0'".fl(),

which allows of live phalanges in tiie longest digit. There may have been
more. That tlie digits were of unequal lengtli is indicated liy portions ot
two in matri.\ accompanying (he specimens, where th(^ articulation of two

phalanges falls opposite the shaft of one of the adjoining digits. The pha-
langes were separaled liy a short interval oi cartilage.

149

The size of this roittili' was near that of Liodon validus. pfiliaps tliirtv- -
five to forty feet in length.

The specimens on vvliich this species rests were discovered hy l^rof. B.
F. Mndgc, formerly State geologist of Kansas, now professor of geology in
tlie State Agricultural College of Kansas, on the north bank of the Smoky
Hill River, thirty miles east of Foi-t Wallace, Kansas.

Numerous fragments of auollu^r larger individual were found by Professor

Miulge near the same locality, which belong probably to tlie same species.

Among them is a portion of the maxillary Ijone, will) ])ases of two teeth; the

bases of the crowns wiiere broken olV are not com|)ressed, but slightly oval.

A radius is a ilal lione. moic dilated al one extremity than that of Clidaskn

pro])yt}ioii.

Measvrtmentx.

M.

Lci)j;lh iil'tbo nuliiis ^- '0"^

Wiatli of tbr nuliiis at t-lu; niiriow extremity ''•"'"'*

Width of till? riiiliiis ;it the wiilei' iixtieiuity O.OdO

Wiiltbof thr iiulius nicilially ''•''■'-

In addition to the two individuals proeure.l by Prof. B. F. Mudge, the
writer obtained a considerable part of a third from a low blufTon Fox Caflon,
soutli of Fort A¥allacc. This iiududes seventeen lumbar, dorsal, and cervical
vertebrae, including a.\is, with rii)s, and a large part of the cranium, with both
quadrates, occipital and periotic regions, etc. Its characters may be briefly
pointed out as follows :

Articular surfaces of dorsal and cervical vertebrte transverse oval, exca-
vated above lor neural canal ; diapophyses not extending below to the edge
of the cup, hence not receiving an area of articular cartilage continuous with
the rim. Occipital crest low, oi)lique; quadrate bone larger; the meatus
depressed between ridges. A l)iitton of articular surface on posterior face
of hook. Scarcely any rudiment of zygosphen. Teeth small, ranch incurved,
faceted, and striale-ridged.

Some characters, additional to those already derived from the first-known
examples, may be added. The mandible supports only twelve teeth. The
palatine bone is shorter anterior to the external process, and longer behind it
ihan'niH.coryphceus. In our specimen, the posterior extremity is broken
off, yet shows no indication of the (nnargination for the pterygoid bone an inch
behind the position of its anierior extremity in H. con/p/ueiis. There are ten
teeth on the part preserved, four in front of transverse process (six in H.
cori/phfEii.-i), and six (proij-^bly seven) Ix'hind (fix in H. cori/ph(eus).

150

The plate is more expanded than in tlie last-named species, especially
tlie thickened inner margin, which only approaches the basis of the last tooth
(reaches the tooth-line at the (ifth in H. coryphceus).

The ocdintal crest is low, and directed obliquely forward from the foramen
magnum. The suspensoria are stout, and directed at an angle.of 45° in lioth
the superior and posterior directions. Tlie basisphenoid is strongly keeled
below. The qiiadratum is like that of H. coryphoius in its massive external
angle and ridge, but differs in the shorter hook and the non-interruption
of the groove between the internal angular ridge and the knob below the
meatus. The cervical and dorsal vertehrcE displa}' the same disproportion in
size observed in H. ionjpliceus.

Measurements.

M.

Length of the os quadrat um 0. C81

Width of tho articular extremity of the os qnadriitum 0. 038

Length of the dentary boue 0. s;80

Length of the tooth of the dentary bone, third from behind 0.022

Length of the tooth, crown only 0.016

Length of the siispensorium from the foramen ovale 0. 108

Total length of the cranium (23 inches) 0. ;3go

This species cannot be confounded with the Liodou pruriger, Cope, and
L. congrops, Cope, owing to its depressed vertebral centra; from L. milchillii,
DeKay, the equal and numerous pterygoid teeth separate it at once.

PLA.TECARPUS CUKTIROSTRIS, Cope.

The specimen below described was found by the writer on the denuded
foot of a blufi" on the lower part of Fossil Spring Canon. The posterior part
of the cranium, with several vertebrge, was found exposed, and many other
bones, including the cranial, were found only covered by the superficial
washed material. Other portions were exposed on excavating the blue-gray
bed of the side of the spur adjoining. The name has reference to the abl)re-
viation of the head and jaws. These are relatively shorter than in any other
species here described where these parts are known. The end of the muzzle
does not overhang, but descends gradually to the tooth-hne. There are but
ten maxillary teeth and two premaxillaries on each side. Size about that
of P. coryplicnus.

Characters. — Cervical and dorsal vertebrae with transversely oval articular
faces, which are little depressed, and, though not continued to the base of the
neural arch, are scarcely excavated above for the neural canal. The diapophy-

151

sis with stout iiitVriur horizoiitul IjraMcli, which is capped h_v an cxlcnsion of
the articular cartilage from the rim of the cup. Occipital crest elevated, sub-
vertical. Quadrate broad below ; pit sunk between Ijounding ridges.

Description. — There is a great disproportion in the sizes of the cervical
and posterior dorsal vertebrae ; the centra of the latter are rather more de-
l)ressed than those of the former. They are similar in proportions to those
of the other Platecarpi, and shorter than those of the Clidastes. The short
axes of the articular faces are sub vertical. The rudiment of zygosphen is
seen in the sliglit anterior prolongation of the roof of the neural canal. The
keel of the hypapophysis of the atlas is short and obtuse.

The greater part of the cranium is preserved. The supraoccipital keel is
vertical, and furnished at the summit with a plicate knol) for the insertion of a
Hgamentum nuchce. The thickness of the walls of the bone is not equal to tliat
in H. co7-yphcBus, and the suture for the parietal is a double squamosal; i. e.,
with a groove along the middle of the edge. The basisphenoid is but slightly
keeled below, and is distally expanded into a horizontal plate on each side.
The parietals are, as usual, conHuent, and send off two light arches postero-
laterally tor union with the squamosal bone. Between their origins are two
subparallel ridges, which disappear, the transverse section of the narrow part
of the parietals being rounded. The lateral ridges within the temporal fossae
are obsolete, while the convergent angles which bound the parietal table pos-
teriorly are strongly marked. This table is nearly plane, and the foramen
parietale is large. The frontal is narrowed in front, and has an elevated keel
along its anterior half The ollactory groove is not much contracted behind,
but is closed by the apex of the rugose area in front of the foramen parietale.

The palatine bone is narrow, and the external margin is very slight, the
bases of the teeth being exposed in that direction. The inner margin is
much thickened downward, but not so as to be a vertical plate. The hinder
part of the bone is flat and horizontal, with a long maxillary process. The
pterygoid notch falls opposite the second tooth from behind. The whole
number of teeth is eleven.

The J aics are represented hy the greater part of all of the tootli-beariug
portions. The maxillary bone is shallowly sulcate on the exterior face. Its
proportions arc quite similar to those of the H. coryphcBus, but the teeth it
supports are larger and fewer. There are none missing from the extremities
of the specimen, the wliole number being ten ; in the H. corijphmtif, there are

152

eleven. The crowns iirt- incurved, t'acctcd externally, aiul striate ymoverl
internally; there are cutting-edges in front and rear, both strongest near the
apex ; the anterior continued to the base.the posterior wanting on the basal third
on the median maxillaries. The anterior nareal excavation marks the fourth
tooth from the premaxillary suture. The premnxillary hone is remarkable for
its shortness and flatness at the extremity ; tins part being depressed and
scarcely projecting at the lower margin in front of the anterior teeth. These,
as nsual, number four.

Both quadrate hones are preserved nearly entire. They have the same
general character as those of PL ictericus and Fl. coryphcBus ; reseml)ling rather
the ktter in the great length of the posterior hook, which is without posterior
marginal button. The proximal internal angle is large and obtuse, and is
continued into a prominent thick ridge. The latter divides below, the thick
extremity turning outward and ceasing ; an acute ridge continuing inward, and
joining the interior acute extremity of the distal articular surface. The sub-
nieatal knob is broad and thick, and not prominent; and its extremity turns at
an acute angle forward on the outer face, and forms the commencement
of the great ala. The articular surface is straight crescentic, with an expan-
sion on a tuberosity on the inner face (concave of crescent). The meatal pit
is sunk between the ridges surrounding, one of which is on the outer margin
of the posterior hook.

The mandihle is nearly perfect. The dentary bone bears thirteen teeth,

and, at the extremity, is contracted in both directions, and not prolonged beyond

the base of the last tooth. The ridge, which descends from the cotylus along

the inner face of the articular bone, is not nearly so strong as in the P.

mudgei.

Measurements.

M.

Length of tbe axis with tbc odontoid 0.06'i

Elevation of the neural spine of the odontoid at the middle 0. 04(i

Lenj^th of the third cervical (body) 0. OoO

Diameter of the ball, vertical 0. 0i5

Diameter of tbe ball, horizontal 0. UlJ'i

Length of a posterior dorsal 0. Oii.'i

Diameter of the ball, vertical 0. Uod

Diameter of the ball, horizontal 0. (•.')()

Length of tbe basis cranii 0. U'JO

Length of the suspensorium 0. 105

Elevation cf the occipital crest above the floor of the foramen magnum 0. (i 15

Length of the tooth-lino of the palatine 0. 115

Length of the maxillary bone U. LIU

Length of the preniaxilUiry laterally O.0;l5

153

M.

Widtb of tho piemaxillary at tho seuond tooth I). 041

Lcugtli of tho deutary 0. 2'ir>

Lengtli of a luasillary tootb 0. OCO

Length of tho maxilldiy tooth, crown only 0.023

Length of the os quadratum 0. 077

Width of the OS quadratum distally 0. 045

Length of the parietal 0. 085

Length of the frontal to the nares (median) 0. 110

Width of thefroutal hetweeu the orhits 0.077

Total length of the eranium (18.7r> iuches) U. 473

Platecarpus crassaktus. Cope.

This saiiriiui, which is of larger size than the last, is represented by a
series of dorsal, lumbar, and caudal vertebrEe, with some bones of the liinl)s

The vertebrae are as much distinguished for liieir shortness as those
of P. lolisplnis are tin- their elongation. Tiie articular faces are but little
broader than dcej), and their planes are slightly oblique. They are very slightly
truncate aljove by tiie neural canal. The interior face is somewhat concave
in the longitudinal direction. The zygapophyses are stout, and there are no
distinct rudiments of zygosphen.

The dorsal vertebrae best preserved are those in which the diapophyses
reach the middle of the sides of the centra, and have no horizontal limb.
The latter are luirrow, and have not extensive articular extrcmi'al siirliices.

The lumbars and anterior caudals have round articular surfaces. One
of the latter, with strong diapophyses, but posterior, is subpentagonal in
outline of cup.

This species furnished the materials for the first description of the
posterior extremities in this order of reptiles. The humerus is a remarkable
bone, having the outline of that of Clidastes irropython^ Cope, but is very
much stouter, the antero-posterior dimensions of the proximal extremity
being greatly enlarged. The long diameters of the two ('xtiemities are, in
fact, nearly at right angles, instead of in (he same plane; and the outline
of tlie proximal is subtriangular, one of the angles being prolonged into a
strong deltoid crest on the outer face of the bone, which extends half its
lengtii. The inner or posterior distal angle is nuicli produced ; while the
distal extremity is a flat, slightly-curved, diamond-shaped surface. The fibula
is as broad as long and three-quarters of a disk. The phalanges are stout,
thick, and depressed, thus differing much from those of Platecarpus ictericus.
A bone which I cannot assign tit any other position than that of femur has
20 c

154

a peculiar form. It is a stout bone, but more slender tban the humerus.
The shaft is contracted and subtrilateral in section. The extremities ore
flattened, expanded in directions transverse to each other; the proximal
having, however, a lesser expansion in the plane of tlie distal end. The
former has, therefore, the form of an equilateral spherical triangle ; the apex
inclosing a lateral fossa, and representing probably the great trochanter The
distal extremity is a transverse and convex oval. This bone is either ulna,
femur, or tibia, judging by form alone. Its greater length, as compared with
the fibula, forbids its reference to the last; the trochanter-like process of the
head is exceedingly unlike any examples of the second bone I have seen.
Its reference to femur is confirmed by its presence with the caudal vertebrae
of a similar species from near the Missouri River, Nebraska, and its resem-
blance to the femur o^ Liodon dyspelor.

Measurements.

ii.

Length of the humcius P- 10"

Proximal diameter of tlio humerus 0. 005

Distal diameter of the humerus 0. 102

Length of the femur <l. 080

Proximal diameter of the femur 0.0(35

Median diameter of the femur 0.035

Leugth of the centrum of a. dorsal vertebra without the ball 0. 061

Transverse diameter of the cup 0- 060

Vertical diameter of the cup 0. Ooil

Length of a lumbar (total) 0. 055

Diameter of the ball of a lumbar (transverse) ti. 060

Length of a caudal 0. 041

Depthof the bull of a caudal 0. 0,)2

Width of the ball of a caudal •'• 0-'>'2

The form of the humerus is something like that ui' Jchthi/osauius. Both
this element and the femur are remarkable for their small size. They are
scarcely half the dimensions of the elements of the anterior liml) of Plale-
carpus ictericus; and are even less than those of L. dijspelor in proportion to
the animal^ size

Some of the ribs preserved exhibit cylindric shafts.

The form of humerus nearly resembles that of P. sinms as figured by
Professor Marsh, and it is probable that these species are very nearly allied.
The vertebrfE indicate the largest of species of this genus.

The remains above described were obtained by Prof B. F. Mudge, near
Eagle Tail, in Colorado, a few miles west of the line separating that Terri-
tory from the State of Kansas.

155

A series of twenty-nine caudal vertebrae, with and without diapopliyses,

from a bliiti' on Butte Creek, belongs perhaps to this species. Tiic proximal

specimens, at least, cannot he distinguished from those of Professor Mudge's

collection. Tiie distal ones cannot readily be distinguished from tiiose of

L. proriger.

Platecakpus latispinis, Cope.

The remains representing this species consist of seven cervical and
dorsal vertebra^; live of them being continuous and inclosed in a clay concre-
tion. It is a large species, nearly equaling the L. mitchiUii in its dimensions,
that is, forty or Hfty feet in length, and is intermediate between such gigantic
forms as hiodon dijqjelor and the lesser Platecarpi. The type-specimens were
found by Prof B. F. Mudge, one mile southwest of Sheridan, near the "Gypsum
Buttes." These display the elongate character seen in Liodon hevi.t, &c. ; l)ut
the articular surfaces are transversely oval, thus resembling tlie P. iderkus.
They are less depressed than in L. peiiatus and L. dyspelor. The cup and
ball of the penultimate cervical are a little more transverse tiian those of the
fourth dorsal, and none of them are excavated above by the neural canal. The
last cervical is strongly keeled on tlie middle line below, and with a short,
obtuse hypapophysis marking the beginning of the posterior third of tlie
length ; the median line of the first dorsal has an obtuse ridge. There is uo
keel on the fourth dorsal, but the lower surtace is concave in the antero-
posterior direction. The diapophyses on the last two cervical and first three
dorsal vertebrae have great vertical extent; the articular surface lor the rib
is not bent at right angles on the first dorsal. Neural arches and spines are
well preserved in most of tlie specimens. There is no trace of zygantrum.
The neural spines are flat, and have considerable antero-posterior extent on
cervical as well as dorsal vertebrae, and are truncate above. The first dorsal

bears a long, strong ril).

Mcasurement>i.

ji.

TrauBvcise diameter ol the cuji of the penultimate cervical vertebra 0. 0.")!

Vertical diameter of tLe cup of the pemiltinuite cervical vertebra 0. (Ml

Leugth of the centrum of the fourth dorsal, without ball (i. UT'J

Vertical diameter of the ball 0. 0-l>=)5

Transverse diameter of tlie ball 0. 0D.")5

Elevation of the front margin of tin' neural spino of the penultimate cervical 0. OSJJ

Anteroposterior diameter of tin; neural spine of the penultimate cervical 0.050

There are smooth bands around the balls, and the surfaces of the centra
are striate to these.

166

The clej)ressed cups of the cervicals and auioriur dorsals distinguish this
species from the Liodon validu.s, L. pror/g(u-, and P. viudgei. The same ele-
ments are much larger and more elongate than in P. iclerkiis. It differs
especially from these species of Platecarpus in the elongate form of the ante-
rior dorsals. In four of the latter, at least, tlie inferior limb of the dia-
pophysis is turned forward to meet the rim of tlie cup ; while this feature
ceases with the last cervical in L. latispinis. The articular surfaces have
planes at right angles to the axis of the centrum, and are not prolonged above,
as in P. glandiferus. The last hypapophysis is very short, with the anterior
margin transverse and elevated, as in the last-named species.

Platecarpus glandiferus, Cone.

A smaller species than the last, with apparently a greater llcxiljilily ot body,
as indicated by the forms of the vertel)ral centra. It is represented l)y por-
tions of two individuals frpni localities twenty-tive miles apart. There is,
unibrtunately, in each case, only a cervical vertebra; but they agree in pos-
sessing such peculiarities as distinguish them widely from anything yet
known to the writer.

One is an anterior, the other a posterior cervical. The articular sur-
faces are transversely elliptic, and completely rounded above; that is, n('itlier
truncated nor excavated for the neural canal. Their shoiter axes are oblique,
i. e., make less than a right angle with the long axis of the centrum; and th<;
articular surface of the ball is thus carried forward on tlie upper face to much
nearer the base of the neurapophyses tiian usual, iu the anterior vertebra
nearly touching them. The ball is, likewise, more convex than in any other
species, having a slight central prominence in the posterior vertebra. There
is no annular groove round the ball. In both, the articular surface of the
hypapophysis is truncate and Ijounded by an elevation in front, a peculiarity
not observed in any of the species already described There is uo trace of
zygosphen in either. In the anterior vertebra, the diapophyses are nearly
liorizontal; the posterior portion slightly thickened and oblique." The ante-
rior ])oi-tion is thinned out, and very rugose above and below, and does not
continue its margin into the rim of the cup. In the second vertebra, the dia-
pophyses are very large, vertical, and with a horizontal portion rising in a
curve to join the middle of the lateral margin of the cup. Neural spine nar-
rowed upward, and keeled behind.

157

McanKrcments.

u.

Length of tlio (■fiitnim <if (lie antfiior viMtcbni 0. Ofi-l

Diameter of tlie ball, vertical 0. 0:!0

Diameter of the linll, transverse 0.039

Length of t lie posterior 0. 0(j-1

Diameter of the ball, vertical 0. 0:!0

Diameter of the ball, horizontal 0. Oi:i

Expanse of the anterior zyfjapojihyses 0. 05r>

The first vertel)ra was louud by (he wriler at tlic loot ol' a bliill' on tlio
lower part of Butte Creek; the second was procnicd bv Prot". B. F. Mndire,
from a poinl one mile sonihi'ast of Slicridan, iicai" the North Fork of 1hi-
Smoky Kivcr.

It is this species that T compared with the Mosasaurus deprcssiis, Cope,
in a report on (Jie collection made by Profpssf)r Miidge (Proceedings of (he
American Philosophictd Society, 1871. p. 1G8). The size is similar, but tin;
form of the articular surfaces of the vertebrae is very diflferent.

Platecarpus mudgei, Cope.

The characters dislinguishiug this saurian are the following: Vertebrae
without rudiinenlal zygospheii; ([uadrate bones with plane surfaces from the
proximal articular surface and the external obtuse-angled ridge to the meatal
pit; tiie latter, therefore, not sunk in a depression as in the other species.

'Vhe determination of I his species rests on a series of specimens from the
yellow chalk at a point six miles south of Sheridan. Kaiis. They consist
of three vertebrie and traguieuts of atlas, with numerous por(i(jns of cranium
and proximal extremity of scapula.

The iiarts of cra/^wm preserved are the frontal bone without the anterior
extremity, and with the adjacent parietal almost complete; j)arts of the basi-
sphenoid ; the suspensorium ; the ossa tpiadrata; and the greater part of tlie
surangular. The frontal is flat, witii thin edge, longitudinally hollowed on eatdi
side of the median line, wiiicli is marked by a low but acute keel. There is
an abundance of foramina and delicate grooves on the surface, and. posteriorly
elevated striae, which converge to the median keel. Thi; median square ex-
tension of the i)order of the parietal :s in tidvance of the lateral portion ol'tiie
same, and not behind it as in Clidastes i»opython. The t()ntunclle is large.
A marked feature is that the parietal crests unite into a low median ridge a
short distance behind tlu^ f()iitan(^lle, and are not. as in ('. jiropi/thon, separated
by a horizontal plane. The sntnn^s of the l)ones fonninir the side oi" the brain-

158

case are very obscure. Nevertlieless, it .ippt-ars that, the descending margin
of the parietal does not descend to the front of the alisphenoid, but is mar-
gined interiorly by the latter to the postorbital expansion. No part of the
inferior margin of the alisphenoid can have reached the sphenoid, as it ter-
minates in a thin edge, except for a short distance medially, where it is
broken off.

The inferior aspect of tiie parietal and frontal l)ones presents a furcate
keel corresponding to the divergent parietal crests, and a very large funnel
for the epiphysis of the l)rain. The olfactory groove- is deep and regular, and
open like that of P. ir/eria/s.

The si/rangv/ar hone is characterized by the pioniinent longitudinal crest,
which descends on the inner side, from the front of the glenoid cavity to
below the posterior attachment of tiie coronoid bone, where it terminates in
a thin edge; also, by the short distance between the margin of tlie glenoid
cavity (cotylus) to commencement (or end) of coronoid suture, indicating a
shortening of the posterior part, at least, of the cranium. The bone is con-
tinued forward only immediately under the coronoid (ciV. F. ictericux).

The proximal extremity of the quadrate is characteristic, and exhibits
features intermediate between those o( Fkitecarpi/s icterkiis, Cope, and the typ-
ical species of Mosa-saurus, as M. fulciaius, M. dekayi, etc. The proximal artic-
ular face is much like that of i)/. depressus (Transactions of the American Phi-
losophical Society, 1S6J, p. 181, Fig. 48, No. 3). The internal angle is much
smaller than in the Liodons, and more anterior; nevertheless, it is continued
distally as a ridgelike angle, separating the antero-lateral from the postero-
lateral faces as in them, and not presenting the gradual blending of the two sur-
faces characteristic of the genus Mosasaurus. The postero-lateral face is thus
flat proximally ; and the meatal pit, which is well developed, cannot be seen
irom the antero-lateral tace. The distal part of the quadrate is lost, so thai I
cannot determine the character of the ridges there.

The basal element of the axis bears a strong hypapophysis without artic-
ular faces, but very rugose surfaces The same portion of the atlas is a con-
vex parallelopipedon, with median rugose tuberosity and very rugose exlrem-
ities. Its surflice is not separated from its liody anterioily by a deep groove,
as in P. ictcricus.

The iirtirul.nr f\iccts of the srapu/a arc much lircKKlcr than in the other

159

species here described, indicating a iK-ad oi- widci- ;iiliciilatii)ii of liiuncnis.
No limb-hones are preserved.

The vertebra) are too miicli injnrud to be characteristic. One posterior
dorsal now has a compressed centrum, or, at least, not depressed. The infe-
rior face is convex transversely, and tliere is a slight concavity below each
diapophysis; but it is clear that il has been so modiOed by pressure as to
render its normal «hape a matter of uncertainty.

Measiircmenfu.

li.

Piuictal, h'Uiitb 0.074

Pariftiil, widtli bet wren tlic anlci ior iiids nf tlio crests 0.048

Parietal, least widtli 0.022

Frontal, intcrorbital width 0. Od:i

Qn.-idrate, width above 0.020

Quadrate, leiij;th from the pit to the proximal end 0. 023

Sui angular, length of the lower e<lj;o 0. 01.'')

Siiranguhir, depth in front of the cotjhis 0. liuO

Snrangnlar, dejith at the end of the roronoid 0. 0")!j

Posterior dorsal, length 0. O-lf.'i

Kcripnla, proximal width 0. 051

Tins species difi'cr.s from all ihor^c oi' 3Iomsaurus and Liodo)! \\) wiiich
tii(! form of the quadrate is known in th<; character of that bone. From 7^.
/■(Bvh: and L. congrops, in which that element is imknown, it difiens in the
stouter or less slender vertebrae; from L. proriger, in its much smaller size.

Its size is a little less tiian the P. ktericus or L. mlidiis. It is dedicated
to Professor Mudge in reco^rnition of tiie valual)le results of his investigations
a:-r'8tate geologist of Kansas.

PlATECARPUS TliCTULUS, Copo.

Established on a number of cervical and dorsal vertebrae of smaller size
than those characteristic of the other species of the genus. The centra have
no', suffered from distortion under ])ressure. The articular surfaces arc de-
pressed transverse-elliptic in outline, with a slight superior excavation for
the neural canal. A well-marked constriction surrounds the ball.

There is a rudimental zygosphen, in the form of an acute ridge, rising
from the inner basis of the zygapopliysis, and uniting with its fellow of the
other side, forming a j.roductionof the roof of the neural canal. 'I'he combined
keels become continuous with th(> anterior acute edge of t.iie neural spine.
Thus, the form is quite different from that seen in the previously-described
species, and constitutes a lower jrrade of rudiment. The fact that this zvgo-

IGO

sphenal roof is separated on cacli side from the zygapophyses l)y an acnfe
groove gives the former a distinctness more apparent than real.

The fixed hypapophyses are short and broad. The centra are not elon-
gate. Those of the anterior dorsals present an obtuse keel below.

Alcnsitremcnts.

ST.

Length of a median cervical 0. 04-

Diameter of the ball of a median cervical, vertical 0. 0;.i9

Piameter of the ball of a median cervical, transverse 0. 0?3

Length of the anterior dorsal 0.\l4vJ

Width of thecnp 0.0:W

P'onnd by the author on a low bluff", or "break," on Bulte Creek, four-
teen miles south of Fort Wallace.'

A second specimen of this saurian has since been discovered by Profes-
sor Mudge. The frontal hone is thick, and |)reseuts a median keel. The
quadrate is flat on the j)osterior inner face, so that the stapedial pit is exca-
vated in a plane surface; the internal proximal angle is nearly right. The
vertebrse are small, and the hypapophyses short, and with htn-izontally trun-
cate articular faces, as in the lype-speciineu.

This species is the smallest known P///fecar/»/s.

LIODOX, Owen.

Vertebnx^ without zygosphcnal articulation, and with very weak zyga-
pophyses; the chevron-bones not coossified to the centra. Teeth with
opposite acute edges, compressed, lenticular in section. Humerus small,
narrow.

The quadrate bone in the known species of this genus is of small pro-
portions, and presents a very prominent internal proximal angle and longitud-
inal ridge. The structure of the cranial Ijoucs is Hght. The zygapophyses
are weaker, and disappear more anteriorly than in any other genus, approach-
ing nearest to Mosamurus in this respect.

There are four species of the genus known from the Kansas chalk, all
of which have the end of the premaxillary bone protuberant and truncate. A
species of similar character has been described by Mr. Hector, from the Cre-
taceous of New Zealand. Three species from the greensand of New Jersey
arc similar in many respects; but the forms of their muzzles are unknown.

The typical species of this genus {Liodon aiiceps^ Owen) is scx-^^ little
known, but few remains liavinii so far been ohftiincd liom the English (dialk,

101

its locality and horizon. Nuinc rons Norlli American spocies rescmhle it, in
the I'ornis of tlio crowns of the teeth, and it is probable, though not certain,
that they agree in other respects also. Several names have been proposed
for our species, the earliest of which is Macrosaurus, Owen. This name
applies to species with compressed dorsal vertebra?, as L. Icevis and L. mi.tch-
illii, both from the New Jersey greensand. For the species with depressed
dorsal vertebrae, as L. validus from New Jersey, L. perlatus from Alabama,
and L. proriger from Kansas, the name Nectoportheus was proposed, and
briefly characterized (Extinct Batrachiaii Reptilia of North America, 1870, p.
208). Professor ]\larsh subsequently gave the Kansas species the name
of Rhlnosaurus, which name being preoccupied more than once, I changed it
to Rhampliosaurus} This name will remain for species of the type of L.
proriger, if they be found to represent a genus distinct from Nectoportheiis or
Liodon, of which there is as yet no evidence.

LlODON PROKIGER, Cope.

Tiie original description of this large Mosasauroid was based on material
in the Museum of Comparative Zoology, Cambridge, Mass., brought by Prof
Louis Agassiz from the Cretaceous beds in the neighborhood of Monument,
Kans., and near the line of the Kansas Pacific Railroad. It consists of the
greater part of the muzzle from the orbits, with the right dentary and left
pterygoid bones nearly complete; one cervical vertebra (with hypapophysis);
one dorsal; one caudal with diapophysis; and ten caudals without diapophysis.

The characters presented by the vertebral column indicate an exceedingly
elongate reptile ; the transverse diameter of one of the distal caudal vertebrae
is less than one-fifth that of a jvoximal with short diapophysis ; while four
consecutive ones of the former show but little variation in dimensions. This
diminution amounts to two-sevenths of a transverse diameter of the larger
form. With this ratio as a basis, fifty-three two-thirds vertebra? would form
a complete series from caudals one-half the diameter of the last of the four
to the proximal caudal above mentioned. There have been, no doubt, several
caudals in advance of the latter, as the diapophyses are small. From the
slow rate of diminution of the columns of other species examined, it may be

' This name was applied by Fitzinger to two spucies of lizards which had already received generic
nonies, and hence becatiio at once a synonym. Fnrtlicr he did not characterize it. For these reasons, tlio
name was not prcoccnpied at the time I employed it as aliove ; hence there is no necessity for Professor
Marsh's subscqiieut name Tylusanrus, given on the snpposition of preoccupation.

21 c

1G2

supposed that sixty caudal vertcbraj is below rather than above the true
iiumbei'.

The cervical and dorsal vertebrae have been slightly crushed, as they laid
on the side, and present a narrower diameter than is normal ; the cup of
the cervical has not been distorted, and is deeper than wide, presenting the
character of Macrosaurns. The rudimental zygosphen consists of a continu-
ation of the roof of the neural canal in front, to adapt itself to the inner face
of the down-looking zygapophysis of the preceding vertebra. The latter is
thus received into a groove on the inner side of the up-looking posterior zyga-
pophysis. The dorsals and cai'idals exhibit, with the cervicals, that minute,
sharply-defined rugosity which characterizes all the projecting margins,
especially those of the hypapophyses and diapophyses, in this genus and
Clidastes. The whole surface of the cervical is marked with either inoscu-
lating striae or impressed punctae. The same character marks the cranial
bones, though they do not present such rugosity as the vertebrae.

The proximal caudal presents a subhexagonal section, of which the infe-
rior and supero-lateral sides are longest; articular faces about as broad as
high. A broad, smooth space between the chevron-bones. Diapophyses with
broad, ovate transverse section.

A caudal without diapophyses, anterior to the middle of the series, esti-
mated by the size, is but slightly deeper than long, and with parallel lateral
outlines of the articular faces. The neural aixh is very much narrowed antero-
posteriorly, but has a greater transverse extent at its lowest part ; above, the
spine is much compressed, but not widened. The zygapophyses remain as
rudiments just above the small neural canal, but do not probably touch each
other. There are two anterior and two posterior narrow ribs on the upper
portion of the neural spine. The more distal caudals have wider neural
spines, and the arch has also a greater antero-posterior extent. The zyga-
jjophyses are scarcely traceable, and the neural spine is strongly striate. The
reverse arrangement is observed in Clidastes propython, where the neural
spine of the proximal caudal has considerable extent, while those of the pos-
terior and distal vertebrae are almost cylindric, especially the neurapophyses.

Dimensions.

Inches,

Dorpal, length 3.25

Dorsal, width of the cup '. ...'. ii..''iO

Dorsal, depth of the cup 2. 77

Proximal caudal, leugth 2. 14

1G3

Inches.

Proximal, wiiltli of the cup ]' '

Proximal, <lci>tli of the cup ' • '■

Caiulal wit bout tlio diapopliyses, No. 1, leiisHi '• '

CaiKlal witliout tlie diapophyses, No. 1, deptli of tliocup '-J- *j-J

Caiulal without tho diapophyses, No. 1, width of the cup 2.6

Caudal without tlie diapophyses, No. 1, height of the neural canal 0.4

Caudal without tho diapophyses, No. 1, antero-posterior width of tho neural spine 0.8

Caudal without the diapophyses. No. 2, length 1-2

Caudal without the diapophyses, No. 2, depth of the cup 2.15

Caudal without the diapophyses, No. 2, width of tho cup 1-S<5

Caudal without tho diapophyses. No. 2, width of the neural spine (antero-posterior) 1.07

Caudal without the diapophyses, distal, leugth 0.5^

Caui'.al without the diapophyses, distal, depth of the cup. 0.8.5

Caudal without tho diapophyses, distal, width of tho cup 0.64

Caudal without tho diapophyses, distal, antero-posterior diameter of the neural spine 0. 40

The points of attachnieut of the chevron-bones on the distal vertebrae
are strongly-marked pits ; on the anterior, the anterior margins of the pits
are raised and continuous with the chevrons.

The muzzle presents the usual characters uf the large Mosasauroids, but
adds a peculiarity in the prolongation of the premaxillary bone into a cylindric
mass, forming an obtuse beak beyoud the premaxillary teeth. The bone is
narrowed anteriorly, and does not descend regularly, as in Mosasaurm, sp.,
but continues to its abrupt and narrowed termination described. The
extremity is deeper than wide. Immediately in front of and between the
anterior premaxillary teeth, a short acuminate projectioQ interrupts the
surface, and, in front of this, a transverse depression. Above, the surface
becomes flattened, and presents two shallow longitudinal depressions con-
tinuous with the nostrils. Where the premaxillary rather suddenly contracts
into its spine, it is materially wider than the maxillary on each side of it ; in
M. missuriensis it is narrower, according to Goldfuss. The maxillary border
of the nares is rather suddenly concave at the anterior extremity of the
nares, narrowing the maxillaries. The latter graxlually wulen by the expan-
sion of their inner margins.

No part of the fronlals is preserved, but a considerable part of the left
prefrontal remains. It unites by a very coarse, overlapping suture with the
maxillary, whose outline forms an irregular chevron, with the apex pointing
forward in the middle of the maxillary bone. This, it will be seen, is very
different from the form given by Goldfuss in the M. missuriensis, where the
most anterior point of the suture is on the narcal margin. The external
margin of the boue behind, is contracted considerably within the maxillary

164

border previous to its outward extension towai-<l the orbit. This is mucli
less marked in tlie Clidaatcs propython, but is distinct in M. missuriensis.

The maxillo-preniaxillary suture gradually descends toward the alveolar
border, at the extremity of the maxillary bone, where it descends abruptly,
forming an interlocking suture cpiite different from that squamosal type
already observed in other species of the order. The length of the premax-
illary anterior to this point is three-fourths the length of the same to the
beginning of the nares.

The number of teeth on the maxillary bone was probably thirteen ;
twelve alveolae and bases remain, and I add one in the position of the
posterior tooth of M. missuriensis, if such existed. This may be questioned,
in consideration of the small number of mandibular teeth. Premaxillary
teeth, two on each side, the anterior with bases separated only by a groove.
Throughout the whole series, the bases of the teeth are considerably more
exposed on the inner than the outer side.

The crowns are everywhere subcylindric at the base; the inner face more
convex than the outer. Posteriorly, there is a posterior cutting-ridge, as well
as a marked anterior one, both minutely crenulate ; but the former gradually
disappears till, in the anterior teeth, there is only an anterior edge, the pos-
terior face being convex and continuous with the inner. There is a trace
of cutting-edge on the outer portion of the extremity of the crown in the
most anterior teeth. The anterior ridge remains very strongly marked. The
surface is quite rough with longitudinal ribs, of which eight may be counted
on the outer aspect of the second maxillary. These are not strongly marked,
and are separated by concave facets. The basal part of the crown is marked
by numerous fine, sharp striae, which are most distinct on the inner face.

The external face of the maxillary bone presents three series of foramina.
These rise superiorly on the premaxillary, and increase in number, and become
irregular on its extremity.

The ramus of the mandible is massive, and differs from that of Mosa-
saurus giganteus in continuing its proportions to its extremity. Its depth at
the latter point is as great as the sixth tooth from the front. It is prolonged
beyond the first tooth in correspondence with tlie prolongation of the pre-
maxillary. This extremity is conqiressed and obtuse; its inner face is very
rugose, as though there had been a closer union at the symphysis than usual,
though it would not appear to have been other than ligamentous. The groove

1G5

for Meckel's cartilage is very large, aiul is exposed Uclow the last two teeth,
as the spleiiial terminates at the tliinl. Two series of foramina on the exter-
nal face of the ramus. There are alveolae and bases for thirteen teeth on the
dentary bone. This, it will be observed, is one more than in M. gracilis Owen,
and one less than in other species of Mosasaurus. The posterior extremity
of the dentary shows its marks of reception into the notch of the coronoid ;
it is more compressed and less club-shaped tiian the corresponding part
of L. mitchillii., and would indicate less lateral flexibility than in some other

types.

The right palatine is of less elongate form than in some other species.
It presents the sutural face for union with tho maxillary on the outer anterior
extremity, and narrows to an apex a little in advance. Tlie dentigerous face is
widest at the anterior third of the length, where the outer margin is expanded.
This then contracts, and is compressed vertical at the tenth tooth, wliere it is
broken oflf. The transverse process is given off a little anterior to the ninth
tooth. The interior face of the bone is a vertical plane, without projection,
except a slight obliquity at the anterior extremity ; and it is clear that there
has been some interval between this palatine and its fellow. The superior
margin is obtusely rounded.

The bases of the palatine teeth are exposed for two-thirds their length on
the outer side of the bone. The antero-median are large, and the anterior
most closely placed. Their crowns are strongly recurved, round in section,
and with a fine sculpture of straight striae, most marked near the base and on
the inner side. They are more spaced posteriorly than any otlier species
except L. mitchillii, and are relatively larger than in any except the same
species. They have not the compressed form with basal shoulder character-
istic of the M. dekmji.

Measurements of the muzzle. ^ ^

Inches.

, , , 31.

Length of the fragmeut ^^^ .

Length from the eud of the muzzle to the prefrontal • ^

Length from tho end of tho muzzle to the nares ^' '_'_

Length from the end of the muzzle to the maxillary ^^•^■*

Length from the end of the muzzlo to the first tooth -j^

Width of tho muzzle at tho end "

Widthof the muzzle at the anterior extremity of the nares

Width of the premaxillary at the anterior extremity of tho nares -'^

Width of the maxillary between the tenth and tho eleven! Ii toot h |^- -

Depth of tin' inandilili' at the extremity •• ~'^

Depth of the mandible at the sixth tooth ^^

Depth of tlx' palatine at the transverse process.... ""■'

Width of the palatine at tlie transverse process '

166

Inches.

Widtb of tbo palatine in front 2.2

Lcugth of tUe palatine anterior to the transverse process 7.2

Length of the crown of the fifth pal.atine tooth 1.

Length of the crown of the second maxillary tooth 1.9

Diameter of the crown of the second maxillary tooth at the base 1. 1

The vomers are, as usual, separate and narrow. They are in close contact
from the second maxillary to the second premaxillary tooth. Throughout
this part of their length, tliey are embraced by posteriorly-produced vertical
laminae of the premaxillary bone. These laminae unite anteriorly just behind
the second premaxillary teeth, and foi-m a single prominent keel, which disap-
pears between the first premaxillaries.

The fine specimen which is the subject of the preceding description was
discovered by Colonel Connyngham and Mr. Minor near Monument Station,
Kansas, and sent by them to Professor Agassiz.

This is the most abundant of the large species of the Kansas chalk.
The WTiter found a muzzle consisting of premaxillary and portions of maxil-
lary and dentary bones in the spur of the lower bluffs of Butte Creek, and
numerous fragments of cranium and vertebrse on a denuded tract in the same
neighborhood. Both of these belonged to individuals of smaller size than
the type, the opportunity of examimng which I owe to Professor Agassiz.
The more complete Butte Creek specimen belongs to a huge animal ; the size
is grandly displayed by a complete premaxillary bone, with its projecting
snout, and large fragments of the maxillary. These furnish characters con-
firmatory of those already given as above. The vertebrse are remarkable
examples of flattening under pressure, without fracture ; some of them having
a vertical diameter no greater than one's hand. The cervicals are less flat-
tened, and give the impression that they were not transversely elliptic. This
is consistent with our knowledge of the perfect specimen, where it is, as
described, furnished with vertically ovate articular surfaces. In this, the cup
is symmetrical and apparently not distorted, but the ball is a little com-
pressed by pressure.

The most important addition to the knowledge of this species furnished
by the Butte Creek specimen is the character of the quadrate bone.

The internal longitudinal angular ridge is very prominent, and extends
to the distal end. It supports a hook-like prolongation of the proximal artic-
ular surface; almost as large a one as in Clklas/es irrapython, and more nar-

1()7

rowed. Tlie ridge is so proiniiicnt, as to create a wider face or surface behind
the basis of the great ala tlian exists between the latter and the edge of the
auricular meatus. This basis is quite convex inward, and embraces a relatively
smaller space than in any other Pythonomorph. A section of the bone at the
meatus is subtrilatcral, with a notch behind. The distal articular surface is
prolonged below the origin of the great ala, and receives the keeled termina-
tion of the internal ridge.

Measurements.

u.

Total IciiKtli of tlio, qiiiulrato 0. l.')3

Length from tbo superior to tlio inferior orijiin of the great ala 0. OdO

Length of the external angle from the basis of the ala 0. 05'i

The two usual ridges pass inward and downward from the meatal knob.

The above quadrates are flattened from within outward by pressure.

A portion of the palatine l)one, supporting the teeth, displays the char-
acters of the type, viz, the inner face vertical and deeper than the outer,
and forming a strong parapet of bone on the superior or toothless aspect ; the
outer face a little expanded laterally ; the bases of the teeth exposed. It is
proper to add that the locality ascribed to the type-specimen, "near Fort
Hays, Kansas," which was originally given rae on inquiry, is erroneous.

LlODON DYSPELOR, Cope

The type-specimen, which first indicated the characters of this species,
was obtained from the yellow beds of the Niobrara epoch of the Jornada del
Muerto, near Fort McRae, New Mexico. The greater part of the remains
have been described by Professor Leidy/ and a few only of the vertebrae
came under my inspection. A second specimen, more complete in all respects,
was discovered by my party during my expedition from Fort Wallace, Kansas,
in 1871, which is fully described and figurcil in the present work.

In the first specimen, the centra of the dorsals are much depressed, qiiilc
as in L. perlatus, Cope, and Mosasaurus hrumbyi, Gibbes. Their articular
faces are of transverse lenticular form, the superior arch being a little more
convex than the inferior, and obtusely emarginate for the floor of llie neural
canal. The superior outline is thus bilobed ; the lobes rounded. The trans-
verse curvature of the articular ball is quite regular, and not, as in Arosasaurun
7}ia.vimu.s, more steeply inclined at the external or lateral angles. A rather

' Report of the United States Geological Survey by Haydeu, vol. I, p. 'i7l.

168

broad, sinoq,th band separates tbc edge of tbe ball from the surfaces of the
ceutruin adjacent. The latter are rather finely striate-ridged from the edge
of this band. The inferior outline of the centrum is strongly concave, and,
with two venous foramina, separated by a wide interval. The basis of the
diapophysis on a lumbar is very broad, measuring more than half the length
of the centrum. In general characters, this lumbar resembles the dorsal,
including the emargination for the neural canal, but is shortened in relation to
its length. The depressed form of the lumbar centra gives place gradually
on the caudals to a more elevated pentagonal outline, which is still more
reduced in width in more posterior regions. The hajmal arches are articu-
lated on the anterior caudals to slightly elevated bases ; on the more pos-
terior, the bases are reduced in height and more widely and deeply exca-
vated. I have not seen the most distal caudals, and hence cannot determine
whether their chevron-bones articulate in pits, as is the case witli those of L.
perlatus, L. prorige?-, etc. On a caudal, where the depth of the centrum a
little exceeds the transverse diameter, the diapophysis has become narrow
and thick. The excavation for the neural canal is strongly marked on the
more anterior caudal. The smooth border of the articular ball is here narrow,
and the superficial rugae are fine, and confined to the anterior part of the

centrum.

Measurements.

M.

Transverse diameter of tbe ball of tbe posterior dorsal 0. 144

Vertical diameter of tbe ball of tbe posterior dorsal , 0.097

Vertical diameter of the anterior caudal 0. 094

Transverse diameter of the anterior caudal 0. 107

Length of tbe centrum of tbe caudal ,. 0. 071

Transverse diameter of the neural canal 0. 0145

Transverse diameter of tbe basis of tbe diapophysis 0. 032

Transverse diameter of the basis of tbe diapophysis of a more distal caudal 0. 0'J7S

Longitudinal diameter of tbe chevron articulation of the caudal 0.023

Length of tbe centrum Q. QGg

Depth of the ball of the centrum 0. 093

Width of the ball of tbe centrum 0. 091

Length of tbe centrum of a lumbar 0. 106

Width of tbe articular ball 0. 125

The characters of the Fort Wallace saurian are as follows :
The fronto-nasal septu?n is convex in transverse section. The maxillary
bone is much attenuated anteriorly, and supports thirteen teeth. The ramus
mandihdi is light and slender ; the angle is quite produced, and the median
articulation indicates considerable mobility. The palatine bones are narrower
than in any of the species previously described. They are deeply notched

169

for imioii with the pterygoids, and the superior posterior process terminates
in ail acute cone. In front of tlie articulation, the bone is a vertical plate,
slightly concave on the inner side ; the anterior half is subquadrate in section ;
tiie outer ftice subvertical ; the inner regularly rounded. Tiie inferior surface
is marked with a groove which passes from the inner side to the outer. The
portion on the outer side of this groove is on the distal third of the bone pro-
duced downward into a prominent keel or ridge. The anterior extremity is
an acute point. Each bone bears eleven teeth, all of which have the external
faces of their roots exposed. The bones are curved outwardly from the fourtii
tooth from behind; opposite the sixth, there is a longitudinal concavity on
the inner face.

The occipital region and suspensoria are not i)rcsent, Init both qundrutes
were found perfectly preserved, excepting the thin ala. They present marked
characters, being most nearly allied to those of L. jfroriger and L. validus.
The proximal articular surface exhibits an obliquity iu the transverse direc-
tion. It presents a large internal angle, which, instead of being nearly at
right angles to the axis of the main portion of the surface, is nearly in the
same line. The decurved posterior hook is very short. The distal articular
surface has, like that of other Liodons, a small transverse extent, and is divided
by a concavity into two tuberosities. The outer of these receives at its angle
the prominent narrow portion of the internal ridge, which extends from the
internal proximal angle. The prominence of this ridge is greater than in any
other species except L. j^roriger; it is acute throughout its length, and has a
gentle sigmoid flexure. The basis of the great ala includes a smaller area than
usual, and is continuous with a prominent narrow ridge, which proceeds from
outside the meatal crest. This meatal crest takes the place of the " knob '' in
such Mosasauri as M. dekayi; it projects strongly backward and outward as an
angle of two ridges, the inferior being acute and curved, and terminating above
the middle of the distal condyles. The meatal pit is not concealed between
ridges, but is internal ; its form is peculiar, being a narrow oval, three times
as long as wide, directed downward and forward. Thus, the characters of
this element are well marked among those pertaining to the other species.

The teeth are not much compressed, and have a cutting angle on tiic
anterior and posterior margins, which separate nearly ecpial faces.

The vertebral centra ciiangc iu fi)rm liom the anterior to llie posterior
regions. The ball of the axis is round ; those of the vertebrae early succecd-
22 c

170

ing are moderately depressed. The balls of the dorsals are transverse elliptic,
with a slight coiicavity for tiie neural canal; the plane a little obhque to that
of the long axis. The centra are more depressed posteriorly, where the balls
of the dorsals present rounded lateral angles. On the lumbars preceding the
caudals, the base of the neural canal becomes more elevated, and the articular
faces assume a slightly pentagonal outline. This form continues as far as our
specimens of caudals extend. On three lumbars, the centra present two longi-
tudinal angular ridges below, at whose jiosterior ends the chevron articular
surfaces appear on the first caudals. All present an incised annular marginal
groove to tlie ball. The surface, especially the inferior, is strongly rugose up
to this groove, especially on the dorsals.

The axis is much shorter than in any other species here noted where
known. The neural spine has a very oblique superior margin, and is ex-
panded behind. The diapophyses are narrow, and continued as vertical plates
to the inferior face of the centrum at its anterior margin. The diapophyses
of tte other cervicals have the usual horizontal limb, which is, however,
shorter than the vertical. In the anterior dorsals, they are directed more
obliquely upward, and are longer. These aud all other dorsals maintain a
connection between the rim of the cup and the anterior basis of the diapo-
physis by a smooth area apparently capped by cartilage in life, as exists in F.
curtirostris. As we pass posteriorly, these processes descend, and become
narrower, until finally they thin out and lengthen into the ribless diapophysis
of the lumbars. Those of the caudals are long and subcylindric. Their
extremities are deeply striate-grooved. The neural spines of all the vertebrae
are longitudinally striate-keeled. The zygapophyses are remarkable for their
narrow form and surfaces.

The atlas is shorter on the outer and longer on the inner face than in L.
validus. This is caused by the fact that the posterior articular face is not
transverse, but very oblique, and, instead of being vertical and narrow, is
obliquely longitudinal in its long axis. It is separated from the inner face by
a wide rugose groove behind; its lower edge sends a keel downward. There
is no process at the thinned infero-anterior angle.

The scapular arch is small, especially the scapula, which is absolutely
smaller than that of the Platecarpus ictericus, a very much smaller reptile.
The iwsterior nmrgin is thickened, the anterior thinner, and less elevated.
The superior is arched upward and backward. The general form is less

171

oblique than in P. ictericus. Tlic coracoid is twice as large, and is ilat and
thin. Its inner margin is regularly convex; the posterior concave and thin ;
the anterior thickened. The foramen is present.

The humerus is diflferent in form from that observed in P. crassartus, P.
ictericus, Clidnsles, &c. It is relatively less expanded proximally and espe-
cially distally; there is but one deltoid crest, which is proximal, and near one
extremity of the articular surface, and disappears into the general plane above
the middle of the shaft. The general form is flat, partly due to pressure.
The distal extremity is but little convex, and displays the terminal muscular
insertions but little produced. Near the inferior end, there is an external
expansion for articulation with the ulna.

The radius is lost. The ulna, or a bone which is like that, regarded as
such in several species described by me, has the extremities in different planes,
which cross each other obliquely. The proximal is triangular and very wide;
too wide for the humeri in their present state. It is also too long, leaving but
little space for a radius. The distal extremity is as expanded, but much nar-
rower, and presents two articular surfaces, a large and wide and a narrow,
connected by a wide isthmus. The bone was taken out near a humerus, but
not in position.

The pelvic arch, as above remarked, was found perfect, and with all the
elements present, with a femur with the head in relation with the acetabu-
lum. The articular extremities are somewhat depressed, and do not precisely
fit. The pubis is a straight, flattened bone, dilated moderately at the proximal
articular extremity, where it is pierced by a foramen. It is coarsely rugose-
striate at both extremities. The ilium is a longer bone than the pubis ; is more
slender, and more expanded at the articular extremity, where it is also thick-
ened. The shaft is curved so as- to be oblique in position ; it shows no trace
of union with a vertebra.

The ischium is a broader bone, with the axis transverse to that of the
body, and sigmoidally curved, iirst slightly forward, then gently backward.
The common suture is about as wide as the proximal extremity. The ante-
rior margin is somewhat tiiickened ; the posterior is produced into a short
process directed backward, which is the homologue of that seen in the Testu-
dinata, and which is connected with the distal end by a thin concave margin.
The /ew?/7- is rather more slender than the luinierus; the distal extremity
is about as much dilated; (he liead less so. The great trochanter is a thick

172

convex ridge, witlr a truncate discoidal articular extremity, which is nearly
separated from the head by a groove. Both extremities are moderately con-
vex. The Jlbula is similar to that of other species in its broad three-quarters
discoidal form. Both articular extremities are strongly convex, and are con-
tinued on the inner side on the thinned inner border. The external margin
is thickened and deeply concave, and without tuberosity. The iibia is a more
slender element, with subcylindric shaft and much expanded extremities.
The proximal is oval, and is continued as a narrow ridge on the inner side,
for contact with the corresponding ridge of fhe fibula. The distal extremity
is an equilateral spherical triangle, of which the inner angle is on a different
plane from the remainder.

Fig. 7. — Sketch of the pelvic bones and luljaccnt vcitcbni; in tlio relative positions iu which (hoy-
wero founil on reinovin;^ the superinciiuibont rock; 1, ilia ; 2, puhes ; 3, iachia; 4, femur ; 5, posterior
lumbar vortobruj. Size much reduced. '

The phalanges are slender, witli cyliiidric shafts and ex])anded extremi-
ties, which support oval articular surfaces. Tlioso of the two extremities
appear to be similar. The distal ones are extremely small and Hat, willi
o\ itanded extremities.

O'i do iihlftd hones may l)c mentioned two with Hat expanded distal extrem-
ity and thick proximal, bearing an oval articular surface, with an angulate

173

extremity which temiiiiiifcs in ;i Uiiii edge. The form is like Ihal of a nar-
rowed radius of/*, ic/ertrus, hul if is much too shmi lor Ihr ulna. As it was
I'niiiid will) tiic scapula, il is prohaldy a portion of (lie l()rc-liml), and li<;iicc
may be a metacarpal. A somewhat similar but narrower bone may be a meta-
tarsal, A piece which is probably the free hypujmphysis of the atlas is a
transversely elliptic piece, with an oblique smooth articular face at one end,
the posterior face rugose, the inferior with a flat truncate process directed
downward and backward. If correctly identified, its great peculiarity con-
sists in its thinness antero-posteriorly, and the large process. (See PI. xxxi,
fig. 1 d.)

In comparing this species with the L. proriger, its nearest ally, I have
already observed the difference in the form of the articular surfaces of the cer-
vical vertebrae, which are in that species vertically oval, in the present trans-
versely so. The comparison is made between posterior cervicals of both,
which, in L. di/spelor, are less depressed than the others. As it is possible
that the form in the type-example of L. i)roriger may be slightly affected by
pressure, I compare other |)oints. Thus, the palatine bones are more slender
anteriorly, and the outer edge descends lowest in a ridge ; in L. proriger, the
inner is produced downward as a longitudinal rib. In this species, there are
eleven teeth ; in that oi>c, nine. The quadrate bone of Z*. proriger presents a
longer intejnal angle, and more prominent internal ridge, with smaller space
inclosed by the basis of the great ala. My statement, in a pul)lished letter to
Professor Lesley, that the ends of the mandibles were acute, thus dilTering
from h. proriger, is an error, due to my having inadvertently mistaken the i)al-
atines for the dentaries while writing. The posterior extremity of these
bones iu L. proriger is unknown.

The only species whose dorsal vertebrae are known to resemble, in the

stoutness of their form, those a^ L. dxjspdor is the h. crassartus; the manifold

differences of the latter will be at once discovered on reading the description

already given.

Measurements.

M.

Atlas, length of tlio inner articular face _. 0. 0G5

Atlas, length of the posterior articular face 0. 0.')4

Atlas, depth of the posterior articular face 0. 037

Axis, length at the middle of the side 0.075

Axis, depth anteriorly J 0.081

Axis, elovation of the neural spine 0. 075

Axis, width of the neural spine (plane) 0. 045

^xis, diameter of the ball, vortical 0.070

174

M.

Axis, dianicterof tbo ball, borizontal -'. 0.070

Corvical, diamoter of the ball, vertical 0. 006

Cervical, diameter of the ball, borizontal 0. 07()

Cervical, leugtb 0.090

Anterior dorsal diameter of tbe ball, vertical 0. 0G5

Anterior dorsal diameter of tbe ball, horizontal 0.0H7

Anterior dorsal lengtb below (witb the ball) 0.100

Anterior dorsal lengtb of tbe diapopbysis 0.047

Anterior dorsal depth of tbe diapopbysLs 0.040

Posterior dorsal leugtb of tbe centrnm 0.097

Posterior dorsal diameter of the ball, vertical 0. 07G

Posterior dorsal diameter of the ball, borizontal 0. 105

Posterior dorsal height of tbe neural spine (of another) 0.120

Lumbar, length of the centrum 0.090

Lumbar, diameter of the ball, vertical 0.073

Luml)ar, diameter of tho ball, borizontal 0.090

Lumbar, length of the diapopbysis 0. 09G

Caudal (anterior), length of the centrnm 0.073

Caudal (auterior), depth of the ball, borizontal 0.085

Caudal (anterior), depth of tbe ball, vertical 0.075

Caudal (anterior), Icugthof the diapopbysis 0. li.'0

Caudal (posterior), lengtb of the centrum 0. 0G7

Caudal (posterior),lengtb of tbe diapopbysis 0.100

Candal (posterior), height of tbe neural spino 0. 087

Caudal (posterior), diameter of the ball, vertical 0.080

Caudal (posterior), diameter of tbe ball, borizontal 0.084

Maxillary bone, leugtb O.fi.'iO

Miixillary bone, lengtb of the bases of two of the teeth (largest) 0.090

Mandible, depth behind the cotylus 0.110

Mandible, length behind the cotylus 0. 110

Width of tbe nasal septum 0. 0'31

Length of the palatine on the tooth-line 0.380

Depth of the palatine at tho third tooth from the front 0.039

Quadrate, length 0- 1^>0

Quadrate, length of the proximal internal angle .- 0.029

Quadrate, width of tbe face from the meatus to the external ridge 0. 029

Quadrate, width of tbe area of the basis of tboala 0.040

Quadrate, width at the condyles 0.070

Scapula, height (axial) • 0.120

Scapnla, width 0.183

Coraeoid, width 0. 187

Coracoid, length 0.200

Coraeoid, thickness at the cotylus 0.027

Humerus, lengtb .- 0. 189

Humerus, proximal width 0. 120

Hnmorus, distal width 0. 127

Ulna, length 0.179

Ulna, width, proximal - 0. 115

Ulna, width, distal 0. IIG

Ulna, thickness proximally , 0. OGO

Pubis, lengtb 0.245

Pubis, width, proximal 0.039

Pubis, width, distal 0.178

Ilium, length on the cnrvo - - 0. 350

llinm, width, proximal - 0. 018

Ilium, width, distal 0.087

Ischium, lengtb (straight) 0.195

Ischium, lengtb to the posterior process (axial) 0.123

175

M.

Iscliitun, width, ]>i()xiiii;>l 0. •K')

Iscbimn, wiillli, ilistal 0. 07r)

Fciiinr, Icii'^tU _ 0. 185

Femur, width, proximal 0. 0!)3

Foumr, widtli. median 0. 0G4

Femur, width, distal 0. i;iO

Fibnla, Um-xth 0. IIG

Fibula, widtli, proximal 0. 100

Fibula, widtli, distal 0. 118

Fibuhi, [UDximal thickness 0. ().')2

Fibula, median width 0. OSO

Tibia, leu'jth t». 10;{

Tibia, width, pnixiuial •. 0. (I4r>

Tibia, width, median O.dvTj

Tibia, width, distal 0. U,V2

Tibia, thickness, proximal Ul'i

Tibia, thickness, distal 0.0'M

Phalange (posterior), length 0. U-'O

Phalange, terminal, length 0. U15

This specimen is one of the most instructive which has yet been tlis-
covered, including, as it does, fifty vertebrsE from all parts of the column, a
large part of the cranium, with teeth, and both quadrate bones; the scapular
arcli complete, except back of coracoid on one side; botli humeri, radius, and
numerous phalanges of fore limb; the pelvic arch complete, with one hind
limb complete to tarsus, with phalanges. The premaxillary is wanting, but
the adjacent suture of the maxillary remains.

Portions of a second individual of this species, or of L. proriger, were
found On the Fox Cafion. They belonged to a larger animal, one equal to
the New Mexican first described. Professor Mudge has fragments of still
larger specimens.

The principal specimen above described was excavated from a chalk
bluff. Fragments of the jaws were seen lying on the slope, and other por-
tions entered the shale. On being followed, a part of the cranium was taken
from beneath the roots of a busli, and the vertebrje and limb-bones were
found farther in. The vertebral series extended parallel with the. outcrop
of the beds, and finally turned into the hill, and was followed so far as time
would permit. It was aliandoned at the anterior caudal vertebrae.

The outcrop of the stratum was light-yellow. The concealed part of the
bed was bluish. Yellow chalk left on the specimens in thin layers became a
white, or nearly so. The yellow and blue strata are definitely related in most
localities, the former being the superior; but in others they pass into each
other on the same horizon.

In instituting a comparison between this and other known MosnsnurldcB,

176

it will be necessary to consider species referred to Mosasaurus as well as to
Lio(lo7i, from the fact that some of the former may really be Lit)dons. The
Liodoiis with compressed or round dorsal or lumbar vertebras may be dis-
missed from comparison. Of the depressed species, L. pcrlatu.s, Cope, is
known from specimens of one-third or less the size of the present one, which
are further peculiar in having the diapophyses of the lumbars to stand on the
anterior half only of the centrum.

Among Mosasauri with depressed vertebral centra, it is to be noted that
none present so great a degree of depression and I'ateral extension except the
M. bnimhyl of Gibbes. They are all also much smaller. The M. hrumhyi
was founded by Dr. Gibbes on two lumbar vertebrae from the Cretaceous
of Alabama, which resemble those of the AI. dyispelor in form, and also in size.
It is probably its nearest ally, and may be a Liodon. Dr. Gibbes established
the genus Amphorosteus for it, but without sufficient evidence to support it.
The principal point of distinction between it and the L. dyspelor which I
observe is the lack, in the former, of the strong emargination of the superior
margin of the articular surface for the floor of the neural canal, which is so
marked in the latter. I have only the figures of Gibbes to rely on fi)r this
l)articular, and it is scarcely probable that the artist would have overlooked
it had it existed. Should the bounding prominences have been worn ofl^,
then tiie restored centrum would have had a notabl}^ greater vertical diameter
than in the L. dyspelor in the same portions of the vertebral column. As a
second character, I note that, relying as before on Gibbes's figures, the
external angles of the depressed ball are not so extended laterally in M.
hrumhyi.

In size, the vertebrae of the present animal exceed those of the M.
hrumhyi. The latter has been hitherto the largest known species of the
oriler Pythonomorpha, exceeding twofold in its measurements the M. gigun-
teus of Belgium. So the present saurian is much larger in dimensions than
the New Jersey species I have called M. maximus. If, as appears certain,
the Mosasauroid discovered by Webb measures seventy-five feet in length,
and the M. maximus measured eighty, the M. dyspelor must have been the
longest reptile known, and approaches very nearly the extreme of the mam-
malian growth seen in the whales, though, of course, without their bulk.
Such monsters may well excite our surprise, as well as our curiosity, in the
inquiry as to their source of food-supply, and what the character of those
contemporary animals preserved in the same geologic horizon.

177

The Idcnlily wliciicr lliis reptile was first procured is near Fort McRac,
in New Mexico. It was discovered by Dr. W. B. Lyon, surgeon at that post,
and by him sent to the Army Medical Museum, at Washington, whose
director placed it in the collection of the Smithsonian Institution. The
attention to the paleontology of his neighborhood by Dr. Lyon will always
be cause of satisfaction to students, and his name will be remembered with
that of Turner (discoverer of the Elasmosaurus ])latyurus, Cope), Sternl)erg,

and others.

LioDON NEPJt:oLicus, Cope.

Rhamphosaunis nepceoHcus, Cope, MS.

Represented by the niandil)ular and parts of the maxillary and premax-
illary bones, the quadrate, a dorsal vertebra, etc, of a single individual.
These all indicate an animal related lo the large L. proriger, l)ut not more
ihan one-third the size or less. It is about the same size as the L. micromus,
Marsh, but is much more like the L. proriger in characters, so as to rcnd(!r
it important to ascertain whether it be not a young individual of that species.
An examination having convinced me that such is not the case, the points
of distinction will be given farther on.

The premaxillary is very prominent, forming a rostrum, whose inferior
face is narrowed, and suddenly descends to a prominent transverse ridge,
which bounds the anterior alveoli in front. The four premaxillary teeth
stand on an area a little broader than long. Extremity broken. The anterior
suture of the maxillary is vertical and zigzag. It displays a lateral contrac-
tion just, behind the first tooth; while the anterior margin of the nostril is
above the third tooth. The teeth of both jaws have broadly oval bases, and
apices with two cutting-edges and lenticular section. The inner face is more
convex than the outer, most so in the anterior part of the jaws, and neither
is faceted. The enamel is finely striate-grooved, especially toward the base.
The mandible is light and thin, and diminishes in depth posteriorly. The
coronoid is small, and tlic angle is produced backward and but little down-
ward. Tlu! rami are not complete; the large portions preserved exhibit
teeth at intervals of precisely an inch. Professor Mudge, who discovered
the specimen, states that the jaw, when together, measured twenty-six inches
in length, wdiich would leave thirteen inches lor the dcntary bone. This is
not flir from the true number of teeth.

The quadrate resend)les that of /.. di/spdor in various respects. The

23 c

178

internal longitudinal ridge is very prominent, and extends from the proximal
angle to the distal articular face in line with the plane of the short acuminate
hook. The great ala is narrow, and rather stout ; tlie proximal articular face
slopes steeply outward. The stapedial pit is a narrow, straight groove (per-
haps partly closed by pressure). The knob is represented by a longitudinal
crest, bordering the meatus below on the outer side, and not continuing to
the distal articulation. The surface of the latter is crescentic, with an angle
on the outer anterior bolder. This angle is the summit of a short, low, rugose
ridge, which extends j)art way to the knob. Outer edge only of the great
ala radiate-grooved ; posterior angle of distal condyle produced.

The dorsal vertebra is somewhat flattened by pressure; but the ball was
evidently transversely-cordate in outline. The bases of tlie diapophyses are
very rugose : an angle from the articular cartilage is directed toward it from
tlie rim of the cup. Inferior face with an obtuse median keel. The odontoid
hone is deeper than long (fore and aft).

As compared with L. micromus, this species differs in the much less
attenuated premaxillary and maxillary bones, the anterior nostril, and absence
of facets on the crowns of the teeth; from L. proriger in the absence
of narrow concave facets on the anterior teeth, and anterior position of the
nostril ; from L. dyspelor in the less compressed, or less knife-shaped, dental
crowns, and totally different form of the condyle of the quadrate.

Measurements.

M.

Leugth of tUc bases of the two i)reuiaxillary teeth 0.026

Width of the bases of the two premaxillary teeth 0.034

Length of the bases of the two maxillaiy teeth 0.042

Depth of the anterior suture of the maxillary 0. 028

Depth of the maxillary at the end of the uarcs 0.038

Length of the quadrate 0.082

Length of the distal condyle 0.040

Width of the distal condyle 0.017

Width of the great ala on the inner side 0. 032

Width of the inner face above the meatus 0. 0.!7

Length of the hook from the stapedial pit 0.028

Length of a dorsal vertebra 0. 009

Diameter of the ball, vertical ., ^.^~j

Diameter of the ball, transverse 0.048

This species was discovered by Prof. B. F. Mudge, who dug the type-
speciineii from the gray shale of tlie Niohrara Cretaceous, half a mile soutli
of the Solomon River, Kansas.

179
PISCES.

LITERATURE OF THE SUBJECT.

1856. Lcidy, Proceedings Academy Philadelphia, p. 73. Lcpidotus- from

Judith River.

Leidy, /. c, p. 256. Two species of Enchodus and Cladocydus de-
scribed.

Leidy, /. c, p. 312. Mrjlognathus priscus, from the Fort Union epoch,
described.

1857. Leidy, I. c, p. 167. Phasganodus dims described.

1858. MarcoLi, North American Geology, p. 33. Ftychodus lohippleyi de-

scribed.
1868. Leidy, Proceedings Academy, Philadelphia, p. 207. Ftychodus occi-
dentalis described.

1870. Leidy, /. c, p. 12. Xiphactinus audax described.

Cope, Proceedings American Philosophical Society, p. 529. Family
of Saurodontidcs established and defined, and one new genus and
four new species described.

Cope, Hayden's Report on Geological Survey of Wyoming, p. 424.
Apsopelix sauriformis described.

1871. Cope, Proceedings American Philosophical Society, p. 170. Genera

Portheus and Anogmius described.

1872. Cope, /. c, p. 327. Descriptions of twenty species and nine genera;

characters of latter and of families Saurodontidcc and Strato-
dontklce extended.
Cope, Proceedings Academy, Philadelphia, 280. Erisichthe nitida
described.

1873. Leidy, Vertebrate Fauna of Western Territories, in Hayden's Report

of United States Geological Survey, vol. I, p. 288. Twelve known
species redescribed and figured.
Cope, Proceedings Academy Philadelphia, p. 337. Portheus ieslrio and
P. gladius described, and genus Daptinus defined.

1874. (April.) Cope, in Hayden's Bulletin of United States Geological Survey

of the Territories, p. 39. Synopsis of all tlie species ; ten added ;
genus Empo more fully defined.

180

1875. Cope, Proceedings Academy Philadelphia (January). Gar-fishes in
Fort Union beds of Milk River.

In the present work, the families, genera, antl species are more fuliy
defined than has been heretofore practicable.

? PEECESOCES.

SYLLiEMUS, Cope.

Allied to the MugUiflce. A short spinous dorsal fin; ventral fins abdom-
inal, posterior to the spinous dorsal. Pectoral fins subinfcrior in position.
Coracoid bones forming a compressed, keeled body. Scales large, cycloid ;
lateral line present, extending along the middle of the sides. Parietal bones
less than epiotics, entirely separated by the supraoccipital. Frontal bones
large, wide, their common suture distinct.

The opercular apparatus extends obliquely backward, while the man-
dible is produced forward. Hence, the inferior part of the hyoniandibular
and the symplectic are directed obliquely forward. The eiul of the muzzle
is broken off, but the posterior part of the dentary bone does not exhibit any
teeth. The opercular bones are thin, and their inferior borders reach the
median line of the inferior side of the head.

The only species of this genus which has fallen under my observation
is represented by a specimen in which the body posterior to the femoral
bones is wanting. The surface is covered with scales, so that only the out-
lines of the femoral bones can be distinctly seen. These are thickened, and
curved outward ; those of opposite sides are well separated from each other.
The scales exhibit a very delicate concentric line-sculpture.

The very posterior position of the ventral fins distinguishes this genus from
Mugil, while the inferior position of the pectoral fins is not seen in Atherina.
The lateral line does not occupy the inferior position seen in the Scombreso-
cidcB. As compared with Apsopelix, Cope, from the Benton group of Kansas,
%//«??«MA- differs in the absence of continuous dorsal radii or interneural spines
anterior to the ventral tins. There is doubtless some affinity between the two
genera, as the other characters arc quite similar. I was unable to detect a
lateral line in Apsopelix. It is possible that a catalogue-name of Agassiz, viz,
Calamophurus (Poiss. Foss., V, p. 122), refers to this or some allied genus;
but I am unable to discover that it has ever been described.

181

~ SyiJ-.ii:MUS LATIFRONS, Cope.

Keprcsciiied \)y tlic entire 1ic;kI and liody ol' a fish as far as tlic l)asis
of tlic ventral fins, excei)ting the end of the muzzle. The scales are com-
pletely j)reserved, while only the bases of the fins remain.

Tin; l)ody is subcylindrie, while tin; head is broad and flat above. The
inferior side of the head is contracted; the coracoids forming a keel, and the
lower borders of the dentary bones being in contact. The angular portion
of the dentary is strongly grooved on its inferior surface, and the proximal or
anterior parts of the operculum display a radiate sculpture. The top of the
head is smooth, excepting a slight radiate sculpture of the parietals. The
outline of the parietals is subround, and a little more extended than that of
the siipraoccipital, wliich is a short longitudinal oval.

There arc twenty-six or twenty-seven longitudinal rows of scales, or
thirteen on each half; tlio.-^e of the abdomen not differing from fiiose of the
sides. The lateral line runs along the eighth below the dorsal fin, originating
just'above the base of the pectoral fin. There are nine rows of scales between
the occi[)ut and the first dorsal ray. I count the bases of fifteen dorsal radii,
which are all fissured anteriorly, excepting the first, which is rudimental.
The anterior rays are stouter than the posterior, and they embrace the pos-
terior part of the ray in front of them by the basal fissure. The posterior
rays are mucli narrowed, and embrace l)ut little. The pectoral rays are
namerous. The physiognomy of this fish is rendered peculiar by the depressed
form of the snout, with the narrow under jaw. It is impossible to be sure
whether the muzzle was elongate or not.

Measnreinents.

M.

Lciij;tli of the s)peciiiicn (o the l)a.so of tlio veutral liii ('. '*05

Lcimlli (>f tlic specimen to flic base of the dorsal 0.090

I.ciigtli of tlio .'.pccinii'ii to tho base of the pectoral ". 'i7.">

Length of the ppecinicii to the edge of the operculum 0. 071

Length of the specimen to tho eilgo of the preopcreulum CO.").')

Length of the specimen to tho condyle of the mesoptorygoid 0.021)

Lcngtli of tho specimen to the orbit 0.017

Diameter of the front between the orbits 0.020

Diameter of the body at the middle of the dorsal tin 0. 04.''>

Depth of the body at the middle of the dorsal fin 0.050

The specimen was found by Lieut eiianl Marsiiall, of the AV heeler United
Stales Geographical Survey, "near the siimmif of Pike's Peak," Colorado.
The specimen has the appearance ol' having been derived iioui tiie Cretaceous

182

or possibly Juraspic beds at the base of this granitic mountain, and its occur-
rence where found was doubtless accidental.

PELECORAPIS, Cope, gen. vov.

This genus embraces tishes with strongly ctenoid scales and abdominal
ventral fins. There is a spinous dorsal fin, apparently short, and not continued
over the ventrals. The ribs and apophyses are slender, and {\\c dorsal ver-
tebrae short and pitted. The ])uljic bones consist of two antero-posterior
plates, in contact on the middle line. The anterior portion i)r()jects to a
median angle, and there is an angular projection of the lateral border. From
the angle formed by these borders, a long, cylindric rod projects forward
beyond the plate ; those of opposite sides slightly converging.

The general relations of this form are to the tiimilies which combine the
features of the orders of physoclystous and physostomous fishes, viz, Scom-
hresocidcE, Afheriniclce, etc. The pelvis has considerable resemblance to that
of those families, I)ut especially to that of Exocosius. From this it presents
subordinate ditferences.

PeLECOKAPIS VAKIUS, S]>. 7100.

Represented by portions of perhaps two individuals, the larger of wiiich
includes a considerable peart of the body, the head and tail l)eing absent.
On this specimen, it is evident that the scales diminish in size toward the
posterior part of the body, where they are small ; on the anterior region, there
are two scales exposed, in an oblique series, in si.x millimeters; on the pos-
terior region, three and one-half and four in the same. The concealed
portions of the scale are sculptured with minute contiguous concentric grooves,
without any radii. The exposed portion is thickened with a cementum-like
layer, which is marked witii a few radiating lines of p(jres, which sometimes
unite into irregular grooves. Teeth of the comb numerous and strong.
Depth of l)ody at pelvis, 0"'.074 ; length of pelvis, 0"'.040 ; of lamina, (r.022 ;
of rod, 0"'.022 ; greatest width of pelvis, ()"\()23 ; width at basis of rods,
O^.OH; length of seventeen consecutive vertebrae, 0"'.105; diameter of a
dorsal vertebra, 0'".007.

Discovered by Professor Mudge in a lead-colored <'lny, [)rol)ably of the
Benton epoc-ii, twenty feet l)elow the Iitnceramvs bed, two miles west of
Sililev, Kansas.

• 183

ISOSPONDYLl.

SAURODONTIDyE.

This family cmUraces caniivumiis fislics, many of. them of large size
and interesting structure, which have as yet only been discoverea in the beds
„f the Cretaceous formation. Tliey are of interest (o the student of com-
parative anatomy, and also to the paleontologist, as they appear to have been
the predominant type of marine fishes during the Cretaceous period m the
North American seas, and to have been abundant in those of Europe.

The characters already assigned to the family are confirmed by the new
species discovered, and many additional ones added, as follows :

The cranial structure can be nearly made out, and the following points
may be regarded as ascertained. The brain-case is not continued between
the orbits, and the basis cranii is double and witli the muscular tul)e open.
There are no exoccipital condyles, and that of the Ixxsioccipital is a conic
cup. The homologies of some of the bones that constitute the cranial trnlls
are difficult to determine. The basioccipital is longitudinally excavated
below. The exoccipital is probably a small bone, which embraces the basi-
occipital closely, so that it is .litficult to say whether the lK.ne that joins the
opisthotic below is the former or latter. The opisthotic has considerable
transverse extent, and an articular surtacc behind, probably for the posttem-
poral. The supraoccipital is keeled or longitudinally crested al.ove, and is
preceded by, or continued into, a longitudinal median extension, whi<h a.ntin-
ues as far as the frontal bon.-s on the mid.Ue line, separating cnlirely the
lateral elements. It is uncertain whether this 1)0 supraoccipital, or, by
homology with the Siluroids, conjoined parietal bones. If this homology be
true, we can easily refer the elenu-nts which hound the cxoccipitals above,
iind the supraoccipitals on either side, to the epiotics. They occupy the i)osi-
1i,„i „f the epiotics in Salmo, and are produced upward and backward into
crests which have free margins, l)oth on the upper and postero-interior mar-
gins. The appearance of these formerly led me to suspect the presence of a
fontanelle, which 1 am now able to assert has no existence. The extent of
these supposed epiotics anteriorly is limite.l l)y th<' approach of the more lateral
elements to the middle lin<-,. Thes(; elements are wide, and, oiferin- allaeh-
meut to the opisthotic, hyomandibular, and postfrontal, must l)e regarded as
pterotic. The uost frontal is a well-developed l)one. T\v' frontals terminate

184

pusteriorly near the middle of the skidl, aiul are well developed. Thcv are
hounded in front by the prefrontals and ethmoid. The ])rcfrontals are stout
hones, directed ol)liquely downward, and terminate in a large truncate articu-
lar face for a facet of the palatine. The ethmoid is generally wide, and ter-
minates in an apex. It presents a large fiicet downward and laterally I'or
the anterior articular surface of the maxillary, opposing a corrcspondintr facet
of tlie vomer. Anterior and exterior to this point it exhil)its a lateral excava-
tion for the superior condyle; of the premaxillary.

^

Fig. 8. — Craiiimii of I'orthvns molossus, Cope, oiic-foiulU iialiii:il sizu (linear) ; », snpraoceipital
bone; b, cxoceipital ; e, basioceipital ; d, parietal; c, pterotic; /, epiotie ; y, postfrontal ; It, iVoiital ;
i, lirefrontal ; A, etbmoiil; ?, Iiyo-niaiiJibuUir ; ni, metapteryf;oid ; «, quadrate ; o, cctoiiterygoid ; j), pala-
tine ;'/, sclerotic; ?', suborbital ; s, parasphenoid ; ?, jireniaxillary ; », maxillary ; r, accessory maxillary ;
u; deutary ; x, articiilo-angnlar. The opercular bones are ■wanting.

Viewing the cranium on the inferior aspect, the parasphenoid and vojner
are seen to form a stout axis, the former running well posteriorly, and fissured
Iji'hind for the muscular tnlie. Neither supports teeth in any known s[)ecies.
Just Ijehind the line of the orljits, the parasphenoid gives off a latcial process,

185

at the l)ase of which are one or more foramina. The postfrontal shares with
the pterotic the support of the hyomandibular. Tlie prootic is elongate, and
sends a crest downward and forward to the basis of the above-mentioned
j>rocess of the parasphenoid. Superiorly, it bounds, with tiie pterotic and
sometimes {Portheus) opisthotic, a large foramen.

The pretnaxUlari/ hones are short, and form but a small j)ortion of the
upper jaw. The ma.villary is elongate and simple. The hyomandibular is
rather narrow, and does not ])rescnt an elongate support for the operculum.
The symj)lectic is well developed, entering fiir into the inferior quadrate. The
latter is a broad bone, large, in contact with the metapterygoid, which is
itself a thin plate, nearly attaining the pterotic. In Portheus, the pterygoid
is well developed as a Inoad plate extending to the inferior boundary of the
orbits. The palatine exhibits a marked peculiarity in the genera of this fam-
ily. It is a shortish bone, soon uniting postero-inferiorly with the ectoptery-
goid, but .supporting as its supero-anterior extremity a body comparable to
the head of a hammer. This malleolar body, as it may be called, is a short-
ened cylinder, with one extremity articulated to the prefrontal and the other
to the posterior superior of the maxillary facets. This gives the latter bone
a firnmess of support unusual among fishes. It also probably permits,
of .some movement of the maxillary in a horizontal plane, which, thougli
small, would have the effect of considerai)ly expanding the gape of the
mouth, tluis enabling tliese fishes to swallow large bodies, in the manner
of the Mosasauroids of the same sea and epoch.

The ectopterygoid is a large bone, and extends down on the front of the
inferior quadrate. Neither it nor the palatine supports teeth in any of the
known genera.

The sclerotica of (he eye is ossilied in Portheus and Ichthyodectes. This
cssificatiou does not cover tlie eye, is not a complete circle, and is unseg-
mented.

Little can be said respecting the hyoid apparatus in this family. Some
superior branchihyals, preserved in Portheus thaumas, are short flat rods.
Two long llat bones, in place between the dentaries of a P. lestrio, a[)pear to
be the distal ceratohyals. They terminate in some crushed basihyals, and are
covered with minute teeth en brosse on the inner faces and superior margin.s.

No specimen exiiibits the entire scapular a?-ch, but several preserve the
scapula witii adjacent parts; two, a Portheus and probably an Ichthyodectes,

186

display most of the elements in place, and several others exhibit the artic-
ulation of the pectoral fin. In the genera named, the clavicle is a wide bone
antero-posteriorly, and is connected with the epiotic by a strong osseous bar,
and probably with the apex of the parietal by another bar. The posterior
part of the arch between these connections is occupied by the stout scapula.
Its posterior face is principally occupied by three convex articular facets.
It sends a short laminar continuation downward behind the clavicle, and
turns inward above, with a massive body at right angles to its long axis.
This transverse portion is supported by the coracoid, which is a stout, flat
rod, narrower than the clavicle, and is apprcssed to the inner face of the
latter nearly to its distal end. Its posterior border is separated from the
clavicle by a deep groove, but the anterior margins are continuous. Above,
it incloses a large foramen with the exterior part of the scapula. It is not
now possible to state whether there is any prsecoracofd (Parker; '^ spangen-
s/wcA-," Gegenbaur) ; but the upper part of the bone here called coracoid
occupies the position of the prsecoracoid in some fishes, articulating with the
superior instead of the inferior extremity of the scapula, as is usual. Except
iu the elevated position of the scapula, the entire arrangement approaches
.that of the Siluroids ; but the inferior part of the scapular arch is not hori-
zontal as in those fishes, but vertically compressed.

The articular facets of the scapula are convex : the inferior and largest is
oblique ; the median and smallest is situated behind the axis of the others.
Behind the superior two, on the transverse part of the scapula, are two round
foss£E, in line, adapted for the reception of the condyles of two basilar bones.
The pec/oral Jin is composed of osseous rays, of which the first is much
the largest, forming a powerful defensive weapon. As the fossils are found,
a number of these rays usually lie in close apposition, edge to edge; but they
are not coosified, and in life probably diverged in the usual manner, extend-
ing the intervening membrane. Their component halves difier much in form,
and are easily separated ; and, as they often occur in this condition, inex-
perienced persons may be led to regard them as entire spines.

The femoral bones, or those supporting the ventral fins, are preserved in
specimens of Ichthijodectes and Portheus. They resemble those of many
physostomous fishes, but present a number of characteristic peculiarities.
Their posterior portion is massive and is expanded on the outer side to sup-
port the facets for the ventral rays. They are also expanded on the inner side,
and strongly united, in the case o^ Portheus, by suture. The anterior part of
these bones is thinner, and consists of two parts, — an inner rod, and an outer
lamina at the base of the rod. The rods, or styles, are directed (()rward, par-

187

allel to each other ; the lamina is turned outward and upward (see Plate XLV,
fig. 7, and cut, fig. 9). In Iclithijodectes anaides, the outer portion of the
lamina is extended backward as well as forward; in a species of Portkeus
(fig. d.), forward only. The face of attachment of the ventral fin is infero-
lateral, and in antero-posterior line with the anterior rod ; it is therefore
within the line of the external margin of the lateral alu of the femur. The
face is expanded in a vertical direction, and is suhround. Its manner of
articulation with the spines of the ventral fin presents a close analogy with
the corresponding articulation of the pectoral spines witii llie scapula, so far
as regards the first rays. Thus, there are two plane articular facets, — the one
superior, the other inferior. Between these is a prominent and narrow
transverse tuberosity, which bears an articular facet directed partially back-
ward. Behind this, instead of the fossse of the scapula, there is another
tuberosity, which is directed posteriorly. Thus, there are four facets, of
which the largest pair supports the diverging halves of the bhse of the first
ventral spine. The middle tuberosity is probably in contact with a portion of
the second spine; and the posterior tuberosity with the base of the tliird.

I have alluded to the -ventt'al rai/s as spines, and such arc tlie first three.
Whether there are additional spines is more than doubtful, as in Porlheus, at
least, there are no other articular surfaces than tho-se described. A p;iir of
ventral fins, found in relation with their lemoral supports, by Profigssor
Merrill, and .sketched on the spot before removal from the rock, embraces
only the three spines; the halves of the first being separated, as is oflen the
case in isolated specimens. Whether additional soft rays existed is also
improbable; but, if existing, they must have been very few, as the basis fitr
their support, such as exists in SalmonidcE and otiier physostomous fishes, is
absolutely wanting. It is therefore improbable that the internal basilar
bone was similar to that in the above recent fishes ; though from the analogy
of the pectoral fins, one would expect to find ventral basilars. There are
two small bones in the specimen described, which, according to the sketch,
occupied the position of basilars ; and such they prol^ably are. Each is an
irregular discoidal body, with one, a concave facet on one side.

The osseous condition of the ventral rays is a remarkable character it)
fishes of the general physostomous affinities of the S(iiirodo)ilU}ce. It is a
point of resemblance to the physoclystous or spinous fishes, for which the
structure of the superior walls of the skull in some measure prepares us.

A third kind of spinous ray was originally described by me as jiertaining
to the Portheus thaumas, and had been previously referred by Professor
Agassi/, to tlie genus Ptychodux (vide Poissons Fos.siles). This is coniposcd

188

of closely-appressed osseous rays of different widths, each of which is com-
posed of narrow, oblique segments. The sutures of these segments are of
different character in ahiiost every ray : being in the marginal ray en chevron ;
in others, step-like ; and, in others, dovetailed (see p. 200, and Plate XLIV,
fig. 4). I suppose this compound spine to belong to one or other of the
borders of the caudal fin.

The vertebrae in all the species certainly assignable to this group are,
where known, deeply two-grooved on each side, besides the pits for the inser-
tion of neurapophyses and pleurapophyses, except in the cervical region,
where the lateral grooves are wanting. There are no diapophyses. The
caudal vertebrae are rather numerous, but not so much so as in Amia, nor are
they so much recurved as in that genus.

Affinities of the Saurodontidce. — More perfect specimens received since
the description of the cranial structure on p. 183 was printed, render it
almost certain that the median bone of the superior cranial walls is a supra-
occipital; that the parietals are produced upward into an angle (epiotic, p.
183) on each side; and that the epiotics (opisthotic, p. 183) form the postero-
lateral angles of the skull. If this interpretation be true, there is no opis-
thotic bone. It is quite possible to interpret the superior cranial structure
of the Siluroids in the same way, while the arrangement is very different
from that seen in Salmonidce, Cijprinidce, Characinidce, and Esocidce, where
the opisthotic is present, and where the supraoccipital does not present any.
such anterior prolongation. The structure of the scapular arch, if we except
the position of the scapula, has much in common with that of the Siluroids;
while the two basilars and double articulation of the pectoral spine are
striking points of resemblance to the same group. As characters of more
typical Physostoini, we have the maxillary arcade of the mouth, the form and
position of tlie ventral fins, and the apparent absence of dorsal spines. The
form of the bones of the mouth, the presence of symplectic, subopercular,
and postfrontal bones, the basioccipital muscular tube, and the unmodified
anterior vertebrae, distinguish the Saurodontidce widely IVom the Nema-
tognathi, and ally them to the Isospondyli; and with the latter I have for the
present allowed them to remain. It must not be forgotten, however, that
the long supraoccipital and osseous vientral rays arc physoclystous charac-
ters, and that the form and position of the femora arc much nearer to those
of i?(?/o«e than to those of any jjliysostomous fish known tome. The three
persistent vertcbr.ie of the caudal fin rec&W Snlmo, while (he dorsals nearly
resemble those of the Clupeidce. The peculiar malleolus of the palatine bone
is closely imitated by the Ponwlobus iJS&udoharengus (alewife) ; but in that

189

pscudolmrengus (alewifc) ; but, in that species it articulates posteriorly with a
process of the ethmoid, instead of the prefontal. In the blucfish (Poiiiatomus
saltatrix) there is a nearer approach ; here, the malleolus articulates with tlic
prefontal and maxillary, but with the latter only by a squamosal joint. The
same species betrays a resemblance to. this family ia the insertion of the
teeth by roots in alveoli, but the roots are much shorter in the living genus.
They are also accompanied by a series of foramina on the inner side of the
dcntary, as in Saurocephalus.

Six genera are enumerated below as belonging to this fi\mily, of which
one, Erisichthc, Cope, is placed in it provisionally :

Synopsis of genera.

I. Jaws without foramina on the inner face below the alveolar
margin :
a. Teeth cylindric :

Teeth of unequal lengths ; some of them greatly

developed Porthcus.

Teeth of equal lengths - Ichthyodcclcs.

aa. Teeth compressed, knife-liUe :

Teeth of unequal lengths; some of the anterior ■

greatly develojied Erisichthc.

Teeth equal. Daptinus.

II. Dentary bones pierced by foramina l)clow the alveolar
border :

Teeth with subcylindric crowns Saurodon.

Teeth witli short, compressed crowns Saurocephalus.

There are some other forms to be referred to this family, whose cliarac-
ters are not yet fully determined. Thus, Hypsocbn, Agass., from the European
chalk, is related to the two genera first named above, but, as left by its
author in the " Poissons fassiles," includes apparently two generic forms.
The first figured and described has the mandibular teeth of equal length.
In the second, they are unequal, as in Portheus, to which genus this specimen
ought, perhaps, to be referred. Both are physostomous fishes, and not related
to the Sphyranidce, wliere authors have generally placed them. Retaining
the name Hxjpsodon for the genus with equal mandibular teeth, its relations
to Ichthrjodecles remain lo l)e determined liy liutlun- study of the H. Icvesiensis.

190

The view of tlie superior walls of the cranium given by Professor Agassiz
presents characters quite distinct from what I have observed in PortJieus.

A species oi Ichthyodectes, from the chalk of Sussex, England, is figured,
but not described, by Dixon in the Geology of Sussex.

A number of forms, erroneously placed by Agassiz and Dixon in the
genus Saurocephalus, have been referred by Leidy to a genus he calls Proto-
sphyrmia,^ with two species, P. ferox and P. striata. The latter much
resembles a Saurocep/nilus, having equal teeth; while the former prol)ably
includes several species, and probaldy genera. The teeth first referred to it
resemble those of P. striata; while others resemble those of Portheus. An
examination of the figures of the mandibles of some of these in Dixon's work,
shows that the large antl small teeth occupy ditTercnt areas, separated by
grooves, in a manner quite distinct from anything seen in Portheus; but it
can scarcely be regarded as typical o[' Protos2)hyr(sna, which name, moreover,
has never been accompanied by the necessary description. -

Dr. Leidy applied the name Xiphactinus to a genus indicated by a spine
in some degree like those regarded above as ventrals of Saiirodontidfe.
Whether it belongs to any of the genera above enumerated, or, if so, to which
of them, is a question which can only be settled by future investigation.

The history of the definition of this family may be found in the follow-
ing references: Proceedings of the American Philosophical Society, 1870, p.
529; Hayden's Survey of Wyoming, etc., 1871, p. 414; Proceedings of the
American Philosophical Society, February, 1872.

PORTHEUS, Cope.

Proceedings oi' the Aincricaii Philosophical Society, 1871. p. 173; 1. c, 1872, Febriiarv.

Teeth sulicylindric, without serrate or cutting edges, occupying the pre-
maxillary, maxillary, and dentary bones. Sizes irregular; the premaxillary,
median maxillary, and anterior dentary teeth much enlarged. No foramina
on inner face of jaws. Teeth on the premaxillary reduced in number. Oper-
cular and preopercular bones very thin. Cranial bones not sculptured.

The fishes of this genus were rapacious, and, so far as known, of large
size. They constitute the most formidal)le type of physostomous fishes
known. Five species are known to the writer from the Niobrara Cretaceous
of Kansas. These are represented by numerous fragments of many individ-

' Trausactioiis of the American Philosophical Society, 1856.

191

uals, which inchide large portions of the cranium, two almost entirely com-
plete. Others embrace jaws, and one a large part of the vertebral column,
with segmented caudal rays. In one, these rays were found with the cutting
pectoral ray above described, while the simple flat ventral rays occur with
ceveral specimens. In none have any traces of symmetrical spinous rays
been found, nor strong interneurals capable of supporting such. In none of
the more perfect specimens with crania have the segmented rays been found;
but the fossil of P. thaumas, where they occur, is represented by a vertebral
column and its appendages, which do not differ appreciably from those of P.
molossus.

In the cranium of this genus, there is a well-marked supraorbital rim.
Each opisthotic forms a prominent angle, directed posteriorly on each side of
the exoccipital. The parasphenoid is a stout and narrow bone, deeply emar-
ginate behind for the passage of the muscular canal. It has a transverse
expansion in front of the base of the prootic, which rests on a backward con-
tinuation of the same. Tliis expansion is pierced behind by two round fora-
mina. The shaft is abruptly contracted in front of the expansion, and is tri-
gonal in section. The prefrontal extends downward and forward, and carries
inferior and anterior articular faces for the maxillary; the latter vertically
transverse. The postero-inferior portion of the ethmoid bears on its posterior
extremity a concave articular face, which opposes that of the vomer. The
floor of the brain-case in front of the prootics is supported by a vertical style,
which is bifurcate above, and rests on the parasphenoid below.

There are large thin supernumerary bones attached to the upper side
of the distal half of the maxillary, as in various clupeoid genera. There is a
chain of thin suborbital bones. The crest of the vertex /is compressed, and,
in one, at least, of the species, greatly elevated and overlapped by a super-
ficial thin bone, wiiich forms a laminar extension of the crest in front of it.
There is neither enamel nor sculpture on the cranium in any of the known
sj^ecies.

Of the teeth in general, it may be added that their pulji-cavity is rather
large at the base but rapidly diminishes in the crown. The mode of succes-
sion is by direct displacement from below. The young crown rises into the
pulp-cavity, and destroys the vitality of the crown wliile tlic root is absorbed

192

Niinieroiis empty alveoli are to be found in all the jaws of this genus, in
which examination will often detect the apex of the crown of the young
tooth.

All the bones of the palatine arch are present in this genus. The
ectopterygoid is curved and concave on its lower border; the pterygoid is,
on the other hand, thickened and concave on its upper border, where it is
also strongly beveled outward, forming the inferior internal boundary of the
orbit. The metapterygoid is a large, flat, and thin bone, joining the greater
part of the superior border of the quadrate. It does not inclose a foramen
with the other bones of the arch as is seen in Characinid fishes. These
bones are all in place on a block, which also contains pectoral spines o{ Por-
theus. On another block of the same specimen, the bones of the cheek are
exhibited in connection with the quadrate. One is a greatly-expanded, thin
bone, with the middle portion of its surface coarsely pitted. It is adjoined
by a much smaller laminar bone of an irregular, semi-discoid form, witli the
middle of the convex side with three obtuse processes separated by fissures.
The corresponding bone in P. molossus is figured in Plate XL, fig. 9. Tiie
interoperculum is a subparallelogrammic bone, with an open sigmoid artic-
ular surfice at one end. In P. imlossus, there are two short articular faces
on the other end. There is, therefore, no doubt, a suboperculum, though I
have not identified it.

The proximal part of the
hyomandibular of another large
specimen displays a continuous
narrow articular surface for the
pterotic. The posterior portion
is a protuberant condyle ; the
middle jmrt narrower and con-
cave; the anterior part truncate,
and j)resenting forward. Not fur
below the proximal end, on the
2^osterior border, is a condyle for
the operculum. It is sessile and

Fig. 9.— Femoral bones ami ventral fin of a fPoW/ifiis: "ot Very protuberant, and is a
right-hand upper figure, from above ; lower figure, from vertical OVal in Outline. This
the right side ; left-hand figure, from below. , , , i i i

bone closely resembles the cor-
responding one o{ Ichthyodectes anaides.

An incomplete skeleton of another Portheus includes pectoral arch and
spines, branchial arches, etc. The arches are extended posteriorly, and the
inner surfaces of the branchihyal bones are covered with minute teeth en
brosse, and support a few tuberosities. The branchial fringe is preserved,
consifting of long and slender processes.

193

Tlie spine supported by the scapular arch in Portheus and Ichthyodectes^
is a defensive weapon. Proxinially, it presents a concave articular surface for
the sca])ula, with a sliort hook-like projection bounding one end. A specimen
in relation, but somewhat mutilated, exhibits a flat, discoidal basilar Ikmic,
which is probably applied to one of the scapular facets. Two rod-like
basilars are visible, and two round condyles, projecting from the mass at the
base of the fin-rays, fit into the scapular cotyli. The principal spine is flat
and curved ; the convex edge trenchant beyond the middle. Tlie posterior
edge is obtuse but narrow, and exhibits a slight groove on one side medially.
Proximally, there is a shallow rabbet, whose floor is transversely rugose.
Several layers of the tissue of the spine beyond the l)asal portion are deli-
cately longitudinally striate. The distal half is broken away. Length of
fragment, 1 toot; width, 1.5 inches; thickness at middle, 5 lines. The
largest pectoral spine of Portheus in my possession measures 2.75 inches
in width.^

The vertebrae in this genus are rather short, but not so much so as in
sharks. In P. thaumas, nearly eighty dorsals and caudals were preserved ;
those without lateral grooves, or cervicals (the term not appropriate), are not
numerous. There are not more than three vertebrae entering the caudal
fin ; a fact which is difficult to determine, owing to the concealment of the
terminal centra by bases of radii. There are seven liaemapophyses in the
support, all flat except the first, which is like those anterior to it. The sec-
ond is articulated freely to its centrum, and is wider than the others. Its
condyle is characteristic, being double, and with a foramen between it and
the produced extremity of the posterior margin of the bone. It is slightly
se|)arated distally from the third, but the remainder are in close contact.
The radii of the superior lobe of the caudal fin extend at least as far down as
near the end of the third haemal spine from below. The structure of these
parts in the P. molossus is as in the P. thmanas, so far as preserved.

An outhne-restoration of this genus is given on Plate LV.

The species of this genus may be distinguished as follows:

a. Two preuiaxillary teetli :

Maxillary arch thin, deep, with narrow anterior coiulylo ; large
maxillary teeth Jive ; third uiaudibular tooth large, behind a

cross-groove P. molossus.

Maxillary large teeth three ; third mandibular small, without cross-
groove in front of it P. thaumas.

aa. Three to five premaxilhiry teeth:

Maxillary arch stout, deep, with heavy anterior condyle ; larger

teeth five P. Icstrio.

Maxillary arcli thick and shallow ; larger teeth five P. mudgei.

aaa. Pieniaxillaiy teeth unknown :

Maxillary bono deeply concave ; small i'. arciuttus.

• See Plato XLII, fig. 4.

194

PORTHEUS MOLOSSUS, Cope.
Proceedings of the American Philosophical Society for 1871, p. 173.

Represented by four individuals : one from Fox Canon, near Fort Wal-
lace, with complete cranium, and many vertebrae and radii ; a second from
another part of the same, with large part of cranium ; and a third and fourth
from Lower Butte Creek bluffs, both with fragmeuts of cranium and other
portions. In the first specimen, the jaws are perfect and dentition complete.

The premaxillary is vertically oval, convex externally, nearly flat within,
and more than half underlaid by an anterior lamina of the maxillary. The
anterior or median margin is regularly convex, and exhibits no surface or
suture for union with the bone of the opposite side. Its posterior margin
extends obliquely backward to beneath the superior articular condyle of
the maxillary, and has a ragged edge, though the suture is squamosal. Its
superior margin is deeply inflected in front of the condyle, and then convex
and thickened. The anterior margin is thick and rugose witli tubercular
exostoses. There are but two teeth, which are very large, and directed
obliquely forward ; the first is two-thirds the diameter of the second.

The maxillary is a large laminiform bone, with the upper margin consid-
erably thickened proximally, but much thinned distally. It is abruptly con-
tracted at the distal two-thirds its length, apparently for the attachment of a
supernumerary bone. The extremity is curved saber-shape upward, and has
an acute toothless edge. The teeth are four small, five large, and eighteen
small. These teeth, except the largest, have cylindric bases; the crowns
(and bases of the latter) are slightly compressed or oval ; they are straight
and regular, and lean backward. The middle one of the five is largest, being
six times as long as the small ones, but little more than half as long as the
large premaxillary or mandibular. The surface of the maxillary is rugose
with small tubercles on its lower half, and has shallow grooves for nutritious
vessels running downward and forward.

The mandibular rami are short and deep, and have but little mutual
attachment at the symphysis. They are not incurved at that point, and were
bound by ligament only. There is no coronoid bone, and the articular is dis-
tinct. It is short, of a rather irregular wedge-shape, and supports half the
cotylus, above which it sends a short acuminate process. The angular has a
prominent angle, like half an ellipse, somewhat contracted at the l)asc; below

195

it is a rough, prominent, muscular insertion. The l)one extends in a loiiij
sword-shaped process, on the inside of the ramus, to l)eyond its middle;
externally, it is soon covered by the thin truncate edge of" the dentary. This
element is very large. From the angular it rises steeply to a coronoid process,
which lias a slight outwardly-twisted eminence, and then follows a gently
concave line to the symphysis. The teeth are as follows : Two large — a
transverse groove ; tliree large; four very small; nine medium; and two very
small — total, twenty. These teeth have straight cylindric-conic crowns, with
cementinn without stri^ or facets. The larger are a litllc ((impressed.

Measurements of the jaws ciud teeth.

M.

Length of the premaxillary boue ou the alveolar border 0. 04.')

Di^pth of the premaxillary bono on the alveolar border 0. 09:J

Thiclciiess ou the alveolar margin 0. 01(>

Length of the crown of the second tooth 0. ()4i)

Diameter of the crown of the same at the base 0. 014

Length of the maxilhiry bone from the premaxillary 0.270

Depth of the maxillary bone at the condyle 0. OSO

Depth of the maxillary bone at the middle 0. 046

Length of the crown of the third large tooth 0. 028

Diameter of the crown at the base 0.011

Length of the crown at the second .small tooth from the large 0. 006

Diameter of the crown at the base 0.004

Length of the ramus mandibuli 0.350

Length of the angle 0.040

Length of the angular bone exteriorly 0. 080

Depth at the coronoid process 0. 112

Depth at the fourth tooth 0.080

Length of the crown of the first tooth 0. 038

Diameter of the crown at the base 0.011

Length of the crown of the fourth tooth 0. 055

Diameter of the crown at the base 0. 016

The opercular bones are thin ; the operculum broad ; the preoperculum
rather narrow. The latter is without armature, and has some shallow
grooves radiating toward the circumference. Length of bone vertically,
0'".245 ; radius from inner curve, 0"'.09.

Tiie vertebrce display deep lateral grooves ; articular faces smooth. Length
of centrum, 0'".028 ; diameter, O^.OrtS. The fan-shaped ha3mal spine, or
second of the caudal fin, is like that of P. thauinns, but smaller. The last
caudals contract in size very rapidly ; the cup of the penultimate, or last, is
transverse diamond-shaped.

The fragments of the saber-shaped veuiral apine display several layers of
parallel striate dense bone; and the edge is tubercularly dentate, and one side
is much more rugose than the other. At the base, one side is (lat, the other
convex ; and there is a transversely rugo.se band near one edge.

190

The scales are thin ami cycloid, and, though largo, arc not reniarkaljly so
for tlic size of the fish. They are not readily preserved.

Measurements of the cranium.

M.

Length from the augle of the opisthotic to the anterior extremity of the ethmoid 0. ;'0

Length from the angle of the opisthotie to the front of the proiitic 0. 11

Length from the postfrontal to the prefrontal across the orhit 0.11

Length from the occipital condyle to the transverse process of the parasphenoid 0. 117

Length i'roni the occipital condyle to the bottom of the parasphenoid emargination 0.0.").')

Length of the parietal hone on the outer suture 0.07

Width of the parietal boue at the middle 0. 014

Width of the parietal bone to the edge of the pterotic 0.07

Width of the frontal at the middle of the orbit 0.04

Width of the parasphenoid at the middle orbit 0.03

Length of tile inferior cpiadrate 0. 10

Length of the condyle of the inferior quadrate 0.03

Length of the symplectic 0.064

This cranium is figured on Plate xxxix.

The gape of the mouth of the Fortheus wolossus extended the whole
length of the cranium proper, and far beyond the orbits, since the maxillary
reaches to opposite the occipital condyle. The orbits were large. The lower
jaw was deep, and gave the countenance that bull -dog expression from which
it derives its name. The body was stout and moderately elongate.

A comjjlete cranium of u Portlieus, probably the P. molossus, for which
I am indebted to my friend Professor Merrill, of Lincoln University, at
Topeka, Kans., furnishes several points of interest previously unknown. The
moiitli is nearly vertical, somewhat as in Osmeroides, while the vertex is sur-
mounted by an elevated crest. Hence, the superior and inferior facial out-
lines meet at a right angle at the muzzle. The eye is small, and there is a
suborbital chain of laminiform bones. The elevation of the skull is 16.75
inches, while the length is only 12.75 inches. Cut 8, page 184, represents
this specimen.

PORTHEUS THAUMAS, Cope.

SuuroccphaTua thaumas, Cope, Proceedings of the American Philosoiihical Society, 1870, November ;
Hayden's Survey of Wyoming, &c., 1871, p. 418.

This large species rests on a specimen without cranium, originally pro-
cured by Prof. B. F. Mudge. The parts preserved are not distinguishable
from the corresponding ones in two individuals obtained by myself in West-
ern Kansas, which include the greater portions of tlie jaws and suspensorial
apparatus. These indicate larger animals than those of P. molossus, and
one of the most powerful of the physostomous fishes, rivaling in this respect
many of the saurians, which were its contemporaries.

197

The distinguishing features of the species have been already pointed out.

The premaxillary is an ol)liquely oval or subpcntagonal bone ; the suture,
with the maxillary, is not toothed, and the anterior or free edge is smooth, not
tubercular, as in two specimens of P. molossus. There are but two teeth, of
which the anterior is immense, and the second little more than half its diam-
eter. The maxillary is stout, and supports in front four very small teeth;
then three very large, of which the median is largest. The teeth recom-
mence very small and are closely placed in the same line ; but, as the extremity
of the maxillary is lost, the number cannot be stated.

The dentary is similar in form to that of the P. molofsus, but has rather
more numerous teeth. Counting from the front, there are two large, one
rather small; two large, and eighteen small and medium following; the
smallest from third to ninth, inclusive. None of the crowns are preserved, but
the alveoli are round, or nearly so. The large tooth of the premaxillary, if
proportioned as in P. molossus, must have projected 0™.0755, or three inches,
above the alveolus ; the fourth mandibular was but little smaller.

Measurements of the jaws.

Length of the premaxillary 0.075

Depth of the premaxillary 0.09

Depth of the maxillary at the condyle 0. 08

Thickness of the maxillary jusfc hehind the condyle 0.025

Length of the deutary O.'Ji

Depth of the deutary at the symphysis 0. 08

The various portions o( cranial bones preserved are much like those of P.
molossus, but stouter. The hyomandibular is nearly perfect; it is thin, but
has a convex rib extending to its acuminate extremity at the postero-inferior
angle of the metapterygoid and the superior extremity of the symplectic.
The preoperculum is attached by a thickened grooved margin, and is not
overlapped by the hyomandibular. It extends in a curved form round toward
the condyle of the inferior quadrate. Three elongate bones, closely appressed,
I suspect to be part of this bone, with the interoperculum and superior cera-
tohyal adherent. The last is rather narrow, and with smooth distal articular
surface, without suture. The superior branchihyals are a little like phalanges
oi' Mosasaurus in form, being subsimilar and expanded at the ends, and much
flattened. The parasphenoid is similar to that of P. molossus. The position
oi' liu; hyomandibular is vertical to the axis of the l)asioccipital, the siiprrior
part directed forward.

198

Mcastiremciits.

II.

Length of tbo basioccipital to tbo cuil of tbo muscular foranicu 0.077

Length of the hyomanilibular -. 0. 2G0

Lcusth of tbo inferior quadrate (obliquct) 0. 11:5

Leugtb of tbo condyle of tbo inferior quadrate 0. 03G

Length of tbo preoperculum jireservcd 0.305

A portion of one of the flat unsegmented or ventral spines preserved
exhibits an irregular rabbet on each edge of one side; width, 0"\042. The
sclerotic bones are as already described.

A second specimen is still stouter in proportions, as the following meas-
urements show :

Measurements.

M.

Diameter of tbo maxillary condyle 0.034

Diameter of the maxilla above, bobiad the condyle 0.033

Length of the angle of the jaw (exteriorly) 0. 050

Diameter of tbo paraspbenoid at the middle of the prootio 0. 03

Diameter of the dorsal vertebra (crushed) 0. 0G7

The diameter of the vertebra must be corrected by a little reduction.

The largest fish-vertebrae I obtained may be here mentioned. They are
peculiar in having numerous concentric grooves on the articular faces, as in
Ischi/rhiza. They are otherwise as in this genus. Length, 0".04 ; diameter,
0".062.

. A peculiarity of dentition is observable in the two specimens first
described, and in less degree in P. molossvs. A considerable number of
alveolas support no functional teeth (though included in our enumeration),
but arc occupied at some point by successional teeth. In some cases, the
mouth of the alveolus appears to be narrowed by ossification, even where the
tip of the young tooth is in sight; in one case, so tar developed as to close up
to the projecting apex. In other cases, the orifice is entirely stopped by the
ossification, which presents the appearance of a scar with radiating lines of
pores.

The first specimen was discovered in a denuded area among the lower
bluffs of Butte Creek. The flat cranial and jaw-bone occupied the summit
of a cone of twenty or more feet in height, a relic of the ancient blue lime-
stone strata spared from the surrounding denudation. The flat bones had
shed oflT the water, which, running off on all sides, had formed the cone.

Tii(! second specimen ca IVom Fossil Spring Caiion, near the remains of

Platecarpus curtirostris.

109

This species is also represented by wholes or parts of from sevent}' to
eighty vertebrae, with numerous neural and liaemal spines and fin-radii, and,
perhaps, some ribs. There are no teeth nor cranial fraa:mcnts. The l)iilk
of the vertebrae is double that of those of Daptimis phhhotomus.

The vertebras present the usual two inferior, two lateral, and two supe-
rior grooves — the last for the neural arch. There are no cervical vertebrae;
for these characters show them all to be dorsals and caudals. The suture for
the ueurapophyses forms a regular angulatc convexity projecting downward.
The arch is not closed above anteriorly, and is expanded laterally, while the
spine is directed very obliquely backward. The concavities of the articular
extremities are equal in the dorsals ; but, in the caudals, one surface is much
more deeply concave than the other, one being funnel-shaped, and the other
nearly plane in a few.

A number of consecutive vertebrae which represent the posterior portion
of the caudal series are preserved One of these is fortunately the very
extremity; and they demonstrate the tail to have been vertebrated or hetero-
cercal, after the manner of Salmo. On the anterior three of the series, the
lateral grooves have disappeared from the centra ; the neural canal is very
small, and the spines are very massive and curved backward, but much less
so tlian in the more posterior parts of the column; they are flattened, wider
than deep, and in close contact with each other, except the anterior of the
three, which presents a narrowed edge forward. The ha^mapophyses are
Ihin, and suturally united to the centrum by a flat gomphosis. The terminal
series eml)r;ices six vertebrae, which have a minute or obsolete nrural canal,
l)ut haemal canal distinct, but apparently interrupted. The haemal arches are
united to tlie centra by a rather smooth suture. The general direction of these
vertebrae forms a light upward curve. The haemal spines are flat and laminar,
and their margins in contact; they decrease in width and length to the end
of the series. The neural spine lies obliquely backward, and has a narrowed
anterior ridge, but stout shaft.

The anterior haemal spine in place exhibits a subglobular base, like an
articulation, and its shaft is wider than those posterior to it. It is a subtri-
angular flat bone, with neck and subglobular extremity, which applies very
well to a concavity l)etween the anterior pair of pleurapophyscs, but does not
in that position preserve contact witli the aiitc'ior margin of the succeeding
spine. One margin of the enigmatical bone is thin and divergent; tlie other

200

expanded laterallj' and straight. The latter gives olFa transverse prominence,
like half a globular knob, before reaching the extremity. Just within the
latter are two lai-ge foramina, which are connected with the extremity by a
groove on eacl^^side, which meet in a notch where the thin edge passes into
the knol).

Both sides of the neural and hasmal spines are concealed in this species
and in the Ichthyodectes jyrognathus by numerous parallel osseous rods, which
are somewhat angulate in section. They lie along the centra of the anterior
series of caudal vertebrae, but are not to be found on vertebrse of any other
part of the column. Numerous loose and fragmentary rods of the same char-
acter accompany the loose and attached caudal vertebrae, and all of them
according to Professor Mudge, belong to the "posterior swimming organ" of
this animal. There is also a collection of these rods from the anterior region
of the body, which Professor Mudge thought occupied the position of an
anterior limb. They do not, any of them, present a segmentation such as
would be exhibited by the cartilaginous radii of caudal and pectoral fins, and
their nature might have remained doubtful but for the explanation furnished
l)y the anterior compound ray or spine of the posterior, probably caudal, fin.
Tliis ray, as in the case of the pectoral spine and first anal rays of some exist-
ing siluroid and loricariid fishes, is composed of a number of parallel rods
closely united. These are in their distal portions remarkably and beauti-
fully segmented, of which a very simple form has been figured by Kner, as
existing in the pectoral spine of the siluroid genus Pangasius. This seg-
mentation becomes more obscure proxiraally, and finally disappears alto-
gether,- leaving the spine and rods homogeneous. This portion of them is
quite identical with the rods found in tlie positions of fins already described,
and I therefore regard these as fin-radii of the attenuated form presented
l)y cartilaginous rays of most fishes, but ossified sufficiently to destroy the
segmentation. They are thus in the condition of the anterior rays of the
dorsal fin of some of the large CatostomidcR, or marginal caudal rays of some
Ghnracinidce, where they are proximally homogeneous and bony, distally seg-
mented and cartilaginous.

The segmentation above alluded to presents the following cliai'aclers
The spine consists of four principal parallel rods, of which the external on
each side thins, the one to an obtuse, the other to a thin edge. The more
obtuse edge presents a groove on one side, which is occupied by a very slen-

201

(ler rod, and a sliallow raljbet along the Hat edge is occupied by a slender flat
rod. Of the tour principal rods, the two median are the most slender, and
llie flat maririiial the widest. Of the two median, that next the latter is tlu;
wider. Tlie stout marginal, or protiabl}' anterior, rod is segmented en chev-
ron, tlie angle directed disfally and lying near the free margin. The suture of
the resulting segments is entirely straight, except, when returning, it approaches
tlie margin, where it suddenly turns to the margin at right angles to it. The
next rod is segmented without chevron obliquely backward and inward ;
where it leaves and reaches the margins, it is at right angles to them, and the
margin projects obtusely at those points. Between the ends, the suture is very
irregular and jagged, sending processes forward and backward. The segmen-
tation of the next rod is similar, but more regularly serrate; distally, it becomes
as irregular as in the last. The transverse marginal termini of the sutures are
serrate in both. The inner and widest rod presents a still more regularly
serrate oblique suture with the truncate extremities; but, owing to the width
of the rod, the near approximation of the sutures continues for a longer
distance. When broken, the suture appears steplike.

Measurements.

M.

Length of a iiagmeut of the (?) caudal spiue 0.25

Width of the fragment at the iiroxinial fracture 0.06

Greatest thickness at the i)n)xiuial fracture 0.013

Width of the posterior rod at the proximal fracture 0. 0245

Ijenj|;tU of six distal caudal vcrteb:!0 0. 1(1

Width of tlie hicmal spine of the secoud of the series 0.024

Vertical <lianietcr of tlie centrum of the first of the series 0.025

Leugth of the ueural spine and centrum of the anterior caudal 0. 108

Transverse diameter of the neural spine of the anterior caudal at the base 0.0235

Antero-posterior diameter of the four anterior caudal neural spines in coutact 0. 069

Li'iigth of the centrum of a dors.al 0.04

Vertical diameter of a dorsal 0. 0015

Transverse diameter of a dorsal (crushed) 0.041

These remains were found in place by Prof B. F. Mudge; he states tiiat
their extent was eight feet. As they embrace no cervical vertebne nor por-
tions of cranium, two feet are probably to be added, giving a total of near ten
feet for the length of this fish. It was discovered at a point on llie bank of
the Solomon or NepahoUa River, in Kansas, one hundred and sixty miles
fiom its |)uinf ofjuiidioii with the Kansas River.

POUTHICUS 1.ESTRIO, CopC.

Represented l»y a porlinii of the cranium, including both niandil)iil;ir

rami, and the maxillary and premaxillary bones of one side, all witli denlilion
2fi n

202

nearly complete, of" one individual; by the tooth-bearing bones, palatine arch,
muzzle, and sclerotic bones, of a second ; by the tooth-bearing bones, witli fin
rays, of a third; and l)y many cranial bones, with vertebrae, of a fourth.
Three of these indivicUials had reached a larger size tlian those of the P.
molossus which have come under my observation,' and represent the largest
species of the Kansas Cretaceous.

It differs from the P. molossus m the possession of three, and sometimes
four, premaxillary teeth. Three is the usual number, but one specimen
exhibits a minute fourth, which is present and still larger in another jaw.
As P. molossus may in like manner present a variation in the possession of a
third minute tooth, it is necessary to note otiier differences. While the pre-
maxillary and anterior portions of the maxillary have the same transverse depth
in the specimens of the two species, these elements are notably thinner and
lighter in P. molossus, so that the anterior condyle of tlie maxillary is much
narrower and smaller in it than in the P. lestrio. The depth of the maxillary
distal to its jjosterior condyle is also proportionately less in the P. ?nolossus.
The measurements of corresponding parts in these species and the P. mudaei

may be compared as follows :

Measuremfnts.

p. moluntius. 1\ lestrio. P. mudjci.

Depth of the premaxillary 0™. 0.)4 0"'.094 0"'. 070

Depth of the maxillary bebinfl the posterior condyle .000 .065 .040

Width of the auterior coud vie of the maxillary... .011 .016 ?

Width of the maxillary behind the posterior condyle . Olli .018 .020

In the specimen first enumerated, the first premaxillary tooth is very
large, the two others of moderate size. These are followctl by a long rugose
diastema before the maxillary teeth begin ; these are, one small, five large ;
twenty-two small, and eleven or twelve very small. Tlie mandibular teeth
are, one very large, one immense, one small, one medium, four small, eight
large, and two small — total, eighteen. The teeth are all sin.i[)]y round or oval
in section, and the external, probably cementum, layer is smooth. The max-
illary bone has two large superior proximal condyles, separated by a space;
the anterior is the narrower, and is directly behind the premaxillary condyle.
The anterior margin of the latter bone is very rugose.

3Ica.iityemcnfs.

M.

Li!M};lli of the tooth-line of the npper jaw 0. 300

Length of the tooth-liue of the premaxillary 0.0,58

Depth of the premaxillarj« (axial) 0.09.5

' See Proceedings of the American Philosophical Society (on the fishes of the Kansas Cretaceous),
J.muary, 167-2.

203

M

Uiiptli of the maxillary at the secoud coudylo 09.")

De|)th of the maxillary at the twentieth tooth 0. OIm

Diameter of the base of the lirst premaxillary tooth 0. OIH

Diameter of the bases of live of the large maxillaries 0. 0.')2

Length of the deutary above 0. 'J70

Depth at the second large tooth '. 0.095

Depth at the last tooth 0. lO.",

Diameter of the base of the first large tooth 0. Oi:i

Diameter of the base of the second large tooth 0. O'JO

Length of the crown of the secoud large tooth 0. 050

Length of the crown of the sixth tooth from behind 0. 025

If of the same proportions as the P. molossus, the vertical diameter
of the head of this fish would be twcuty-three inches.

The type-specimen of P. lestrio was procured by Prof. B. F. Mudg<',
of Manhattan, Kans., during his survey of 1872, near the Solomon River.
Professor Merrill's specimen was obtained from the same region.

PORTHEUS MUDGEI, Cope

Represented by portions of the jaws, with vertebrae, of a single
individual, discovered by Prof B. F. Mudge in Trego County, Kansas.
The prominent character is seen in the possession of four subequal teetii
in the premaxillary bone, which, therefore, presents a relatively long alveolar
border for their accommodation. The bone is also more massive than in
the other species, and is peculiarly thick on the free anterior edge. The
maxillary bone presents a similar character, and shows this fish to have
•been the most robust species of the genus. The width of the superior
border of the maxillary is greater than in the others, although the vertical
extent of the boue is considerably less (see the measurements under
P. lestrio). There are five or six subequal large teeth behind an edentulous
space on the maxillary bone, while those on the posterior part of it are
small. The specimen is smaller than is usual in other species of the genus.

Measurements.

M.

Width of the premaxillary bone 0. 054

Elevation of the premaxillary bone above the first tooth 0. OGO

Thickness of the premaxillary bone at the middle 0. Oil

Depth of the maxillary at the condyle 0.057

Depth of the maxillary at the fifth large tooth 0.040

Length of the bases of the five large teeth 0. 036

Length of an anterior vertebra 0. 019

Diameter, vertical 0. 0:il

Diameter, transverse 0. 0;{1

Occasionally, the P. lestrio exhibits one, or even two, minute additional
premaxillary teeth, but it exhibits l)iit three large teeth in contradistinctiou
to the four large ones of tlio P. mudgni

204

PORTHEIJS AKCUATUS, Cl)])e.

The smallest species of tlie genus, equaling Ichthyodedes ctenodon in
size, is repi'esented by incomplete maxillary and palatine bones, and perhaps
by accoiiipanying vertebrae and other pieces. Apart from its small ^ize,
this species may be known l)y the compressed and concave alveolar border
behind and below the posterior maxillary condyle, and the very small size
of the teeth which protrude from its subacute edge. The superior border
behind this above-mentioned condyle is oblique, and its anterior border an
acute edge. The interior face of the maxillary is convex ; the exterior
plane, and anteriorly dotted with radiating lines of pore-like impressions.

This species .is less nearly related to the precediug species of Portheus

than they are to each other.

Measurements.

M.

Depth of the luasillary behind the second condyle 0. (125

Thickness of the maxillary behind the second condyle 0. 004

Width of the second condyle 0. 007

Length of the articnlar segment of the palatine .- 0.010

From the vellow chalk of the Solomon River. Kansas. Found by
Prof. B. F. Mudge.

OBSERVATIONS ON THE PECTORAL AND VENTRAL SPINES OF THE GENUS

PORTHEUS.

As already stated, the fin-rays of this genus are readily separated into
their constituent halves. The su]>erior elements of the pectoral fin are
laminiform and concave on the side in contact with the inferior halves,
which they somewhat exceed in width. The inferior halves are massive, and
exhibit a strong superior rabbet on the posterior margin for an overlapping
border of the second ray. On the anterior margin is a more shallow rabbet,
which soon disappears, which is covered by the superior lamina. The
posterior rabbet also disappears at a point varying with the species.
The superior lamina bears the cotylus and adjacent hook, which embrace
the suj)erior scapular facet; the inferior half supports the facets which
correspond to the inferior two of the scapula. The front of the pectoral
spine is sharp-edged, forming a thin blade, hardened by a deposit of dense
bone, which is transversely roughened. The blade is the edge of the supe-
rior lamina, which extends beyond the equally acute border of the inferior
half, the latter fitting closely to the concave inferior face of the former.
Both faces of the spine are covei'ed with a dense layer of bone, which is
marked with delicate longitudinal grooves; and, when the superficial layers
are broken away, the deeper ones are found to be grooved in the same

205

manner. This i)ectur;il si»iiic is a formidablo weapon, measuring between
two and three feet in length and two inches in width in a Portheus thaumas
whose mandibular ramus is a foot in length.

The ventral rays considerably resemble the pectoral but are relatively
weaker. The second and third are much smaller than the first, have a
lonsriludinal rib on the side of one of the angles, and are scarcely flattened.
The first is much flattened, and exhibits a sharp anterior edge, formed by
the projecting border of the thinner half. The capitula of all the rays are
strongly curved to a beak-like apex, bearing a tuberosity on the convexity in
the two smaller ones. They differ from those of the anterior pectoral spines
in the conve.xity of their articular facets, as they do not embrace the facets
of the femur as do the pectorals those of the scapula. Like the latter, the
dense superficial layer of bone is often finely striate-grooved.

ICHTHYODECTES, Cope.

Proceedings of the American Pbilosopbical Society, 1870, November; Hayden's Geological Survey of

Wyoming, etc., 1871, p. 4'21.

Teeth equal, suljcylindric, in a single row, sunk in deep alveoli.
Premaxillaries short. No foramina at the bases of the teeth on the inner
alveolar walls. Vertebrae deeply grooved laterally.

The species of this genus are, so for as known, smaller than those
of the last, and, as their remains are more perishable than those, form less
striking objects among the fossils of Kansas. They are, nevertheless very
abundant, especially in species, five of which are now described.

The general structure in detail is much like that of Portheus. The
maxillary l)one is not contracted distally for a supernumerary bone, as in
Portheus. The quadrate is similar, and the symplectic has a wide exposure
on its outer foce in /. amides. An entire anterior spine of the pectoral fin
of/, anaides is preserved, the halves partially separated (see Plate XLV, fig. 8).
The superior half is wider than the inferior, and projects beyond it, forming
the trenchant anterior border, which is roughened by a deposit of dense
osseous material. The inferior half is but little thicker, and has an acute
posterior border; its surface is delicately striate-grooved. The spine; widens
distally, and thins out to an oblique, irregular edge, and was doubtless con-
tinued as cartilage. A ventral spine accompanies the Itones of /. prognafhvs
and /. mu/ilde»taftts, which is of a more robust form than the above-mentioned
pectoral, but not so wide.

The vertebrae are deeply longitudinally grooved, as in Portheus, with the
exception of a few of the anterior.

206

111 a series of vertebrse similar to those of" this genus, those included in
the hasis of the caiulal fin are not more than three in number.
The species are distinguished as follows :

Premaxillary^ teeth 5, second most prominent ; maxillary
not concave ; dentary with 30 teeth, and biconvex
alveolar border, with obtuse extremity /. anaides.

Premaxillaries ? ; maxillary straight, large, with 40 teeth;

dentary straight, not produced at end ; teeth 26 /. ctenodon.

Premaxillaries 5, first most prominent; maxillary concave,
narrow ; teeth small ; dentary with a hook at apex,
teeth 25 - /. hamatus.

Premaxillaries 7, first most prominent, compressed ; smaller. /. prognalhus.

Premaxillaries 12, second most prominent; the bone mucli

narrowed above, smaller /. multidentaius.

The English species of this genus is figured by Dixon in the Geology
of Sussex, PI. xxxii, Figs. 9 and 9*. I can find no letter-press nor name
relating to it, and cannot determine its specific characters from the fragment-
ary character of the piece of mandible figured.

ICHTHYODECTES ANAIDES, Cope.

Indicated by two individuals: one with both dentary bones and teeth,
with vertebrse; the other with many portions of cranium, fin -rays, vertebras,
and other elements more or less separated. The latter were all taken from
the upper face of a spur of a limestone-bluff, elevated about five feet from
the ground-level, where they were denuded and exposed as on a table.

It is the largest species of the genus, and the anterior premaxillary teeth
are larger than«the posterior. The premaxillary hones are oblique ovoids, very
convex on the external face, thinning laterally and above. The superior mar-
gin presents a thickening bearing an articular surface, while behind it is
an open gutter-like inflection. The large teeth are quite cylindrical. Both
these bones are preserved. But part of the right maxillary remains. It is
thickened above in front of tlie condyle, and is regularly convex at that point.
The teeth are small, tiiere being 10.5 in an inch. The margin is not
concave.

The mandibular rami are preserved almost entire. They are short and
deep, and have a short angular process, which is relatively shorter than in

207

Portheua. The margin rises steeply to the dentary, whieli presents a nar-
rowed rectangle behind. The alveolar margin has two convexities, witli a
depression between; the syniphyseal angle is not prominenl. The lower
posterior angle of the dentary is quite prominent for muscular insertion. Tlie
crowns of the teeth are cylindric, slightly curved inward. The dentary bones
of the second specimen coincide with these in all respects.

Thirty-three vertebra are preserved, all deeply two-groovcid on the sides.
The ribs are articulated by a sigmoid surface to a l)road, short element of a
sigmoid form, which is inserted in the lateral groove of the inferior face, or
articulated by gomphosis.

The ventral spines already noticed are quite flat, witliout serrate edge,
but with some rugosities near the edge on one side only. There are no grooves
on the upper side, but the dense bone is delicately striate.

Measurements.

M.

Length of the prcnuixillnry 0. 033

Depth of the premaxillary 0.045

Depth of the maxillary at the condyle 0.037

Tliickness of the maxiUaiy just behind the condyle 0.012

Length of the mandibular ramus 0. 172

Length of the angular process 0. 014

Depth at the corouoid process 0. 0.")8

Depth at the symphysis 0. 04 1

Length of eight vertebnu 0.212

Width of the articular face 0. 0;;0

Width of rib 0.004

Width of the ventral spine .at the middle 0. 025

Length of the ventral spine (fragment) 0. l.'io

Length of the condyle of t he inferior quadrate 0. 020

The scales associated with this species were tJiiii and cycloid, and diffi-
cult to preserve.

From near t!ie Smoky Hill Uiver, Kansas.

IciITHYODECTES CTENODON, Cope.

Proceedings of the Ainri icaii P]iili)sopliical Society, 1870, November ; H.ayden's Geological Survey of Wy-
oming, &c., 1871, p. 421, part.

Found l)r Profiessor Mudge on the North Fork of the Smoky Hill River;
common in many other localities.

This species is established on one complete maxillary bone, and three-
fourths of the other, a large part of the dentary bone, witli tlu! entire dental
series, and numerous portions of cranial bones. These, according to Profes-
sor Mudge, were found together, and, to all appearance, belong to the same
aniiuiil.

208

rpi

he ilriiuil characters differ from those of Saurocepfialus, as aljove
pointed out, and resemble more those of Saurodon leantts. The crowns of the
teeth are more exserted and slender. The inner face of the crown is more
convex than the outer; but there is no angle separating the two aspects.
The apex is moderately acute, and directed a little inward, owing to a slight
convexity of the external face. Enamel smooth. The alveoli are very close
together, and are probably only separated in their deeper portions. There
are forty-two teeth and alveoli in the maxillary bone. The superior condyle
is low, and its anterior border falls opposite to the last tooth, or the indented
surface wliich wns occupied by the premaxillary bone. The more proximal
part of the maxillary curves inward and backward behind the position of the
premaxillary, more than in S. inognathus. Tne maxillary is a rather thin and
narrow bone, with a broad obtuse and thinned extremity. Its superior mar-
gin is marked with one or more acute ridges, which look as though it had a
contact with a large preorbital bone. The alveolar border is nearly straight.
The dentary bone is remai'kable for its straightness and laminar charac-
ter, and for the depth of the symphysis. The length of the latter is preserved,
wliile posteriorly to it the lower margin of the dentary is broken away. The
alveolar margin is slightly concave, and unites with the symphyseal at an angle
of 65°. There are twenty-six teeth and alveolae, which grow a little larger
to the posterior extremity of the series ; anteriorly, the alveoli are confluent
externally, but, posteriorly, the septa are frequently complete, though thin.
In neither this bone nor the maxillary are to be found the foramina along the
bases of the teeth, characteristic of Saurocephalus or Saurodon, as pointed
out by Harlan and Hays.

Measurements.

M.

Ltfngtb of the maxillary bone 0. 158

Depth at the condyle .' 0.031

Depth at the extremity 0.022

Length of the crown of a tooth , 0. OOGl

Diameter of the crown at the base 0. 00:>S

Length of the alveolar border of the dentary 0. 106

Depth of the symphyseal border of the dentary 0.047

Length of the oper^nlar condyle 0. OlS

Several osseous ventral rays accompany the cranial bones ; they are prob-
ably interior in position, and are much more slender tlian the ventral spines
observed in /. annides and /. prognathus. Tliey are subquadrate in section,
not sculptured nor enameled; there is a shallow groove on the side.

209

ICIITIIYOUKCTKS HAMATUS, Cope.

Represented by a considerable nnmbcr of nnnains of an indivi<liial I'roin
the blue Cretaceous sliale, near Russell Spring, on the Smoky Hill River.

The characters which distinguish this species from /. anaides are numer-
ous; hut they are less marked when compared with those of /. ctenodon, partly
because the premaxillary bones of the latter have not been preserved. In
the first place, the dentary bones of the two arc; of equal length, and su})port
the same number of teeth, while the maxillary of L hamatus is shorter, and
supports more teeth; it is concave at the proximal part of the tooth-line, but
is straight in the corresponding part of /. ctenodon The end of the dentary
is furnished with a strong obtuse process or hook directed upward and for-
ward, not seen in /. ctowdon. The maxillary behind the premaxillary is in
this species thickened, and with two articular surfaces; the proximal looking
outward, the distal inward, and separated by an ol)lifpie ridge from the con-
dyle. In /. ctenodon, there is but one smootii surface, gradually narrowing
with the thinning of the bone from the condyle.

The premaxillary is less extended antero-superiorly than in the species
already described, but supports, as in it, an articular face. There is no groove
l)ehind it, as in /. anaides and Portlieus. It disj)lays a surface for osseous
articulation to near its extremity on the inner side ; while below it, and on
the (external face, near the basis of the first and second teeth, the surface is
rugose. Maxillary teeth, forty-three. The dentary supports twenty-five.
Tlic anterior hook is obtuse, and rises abruptly to above the apices of the
crowns of the teeth. It is knol)be(l above, and supports a lootli not larger
than (he others.

All the cranial bones preserved are not sculptured.

Portions of the thin Hat ventral spines display the delicalely-grooved
striation already observed, while the trenchant edge is Imrderedon one side
l)y raised longitudinal stritB. The other side is minutely pitted.

The vertehrm arc anterior, and without lateral grooves. Thj-ee of them

arc 0"'.06 in length ; an undistorted one is a little wider llian ('cep, and llie

cup is 0'°.026 across.

Mcaaiirements.

M.

I^cngtli of tlio ramus mandilmli 0. 174

Depth at the Bj'in])lijsi3 0.055

Depth of the pr(Miiaxillary (ohliqiic) 0. U-IU

Lciif;th of tho preiuaxillary (oUliqnc) 0. OW

27 c

210

M.

Doi>th of tho maxillary at tlio condyle ^ 0. ((27

Depth of tbo maxillary behind the condyle fl. 02:i

Depth of the maxillary near the middle (I. 0:i7

Width of the ventral spine 0. O.U

This S2)ecies and the two preceding were not very unlike in size ; the
two following are smaller.

ICHTHYODKCTES PROGNATHUS, (JopC.

Proceedings of tho American Philosophical Society, 1870, November {Sauiocephahis) ; Hayden's Geologi-
cal Survey of Wyoming, &c., 1871, p. 417.

In this species, the premaxillary is more rhomboid in outline than in the
others, and is less convex externally. Of its more numerous teeth, the first
is not larger than the last, differing thus from all others of the genus; it is in
line with the nearly straight anterior margin of the bone, and is more com-
pressed than in the other species. The surface of the l)one is peculiar in a
minute sculpture of impressed lines, or lines oC punctae. There is a very
small articular surface on the superior extremity.

This species is represented by a premaxillary and attached proximal
j)ortion of the maxillary l)oiies of the right side, and by a large number of ver-
tebra; and other bones. These portions were associated in the collections
placed in my hands by Professor Mudge, and relate to each other in size, as
do those of the preceding species and the Ichthyodectes ctenodon.

The premaxillary is characterized by its great depth as compared witli
its length, and by the shortness of its union with the maxillary. The pala-
tine condyle of the maxillary reaches a point above the middle of the alveolar
margin of the premaxillary. The latter contains alveolae of seven tectli, the
anterior of which only presents a perfect crown. It is elongate, compressed,
equilateral, smooth, and acute. Its direction is even more obliquely forward
than the anterior outline of the bone, which tself makes an angle of 50°
with the alveolar border.

The vertebrae consist of cervicals, dorsals, and caudals, to the number
of about sixty, most of which are supposed to have been derived from the
same animal. The grooves are as in D. phlebotoinus ; there being two below,
two on each side, and two above. The latter receives the bases of the neu-
rapopliyses, which are in many cases preserved. The inferior pair of grooves
becomes more widely .separated as we approach the cervical series, leaving
an inl'erior plane, which is longitudinally striate-grooved. This plane widens

211

till the grooves bounding it disap|)ear. The inl'erior lateral groove l)ecomcs
widened into a pit, which some of the specimens show to have been occupied
by a pluglike parapophysis, as in Elops, etc., or a rib-head of similar form.
The neurapophysial articular grooves become pits anteriorly, and these
only of all the grooves, remain on the anterior two vertebrae in the collection.
Some of the posterior caudals pi'eserve large portions of the neural arches
and spines. They form an oblique zigzag suture with the bodies, consisting
of two right angles on each, one projecting upward anteriorly, another down-
ward behind. The neural spines are very wide and massive, and in close
contact antero-posteriorly ; these probably support the caudal fin. They are
deeply and elegantly grooved from the basis upward. The centra exhibit no
lateral grooves.

An unsymmetrical ventral fin-ray accompanied these remains, and, from
its mineralization, color, size, and sculpture, probably belongs with them. The
anterior margin is thinned, and with obtuse denticulations; the posterior trun-
cate. The section is lenticular, with a deep rabbet on one side of the posterior
edge; section at the base circular, apex lost. The sculpture consists of fine,
longitudinal, raised striae, which bifurcate and send numerous similar ridges
to the teeth of the anterior margin. This ray difiers from the corresponding
one of /. anakles in its greater relative thickness, its anterior rugosity, and
peculiar sculpture.

Measurements.

M.

Long diameter of tbe spine 0. 0^45

Basal diameter ot the spine 0. 019

Lengtli of two ceivicals (not distorted) 0. 033

Diameter of tlie anterior 0. 021

Length of a dorsal 0. 016

Length of a caudal 0. 014

Width of the nenral spine ef the caudal at the base 0. 01'2

Length of the alveolar margin of the premaxillary 0.022

Length of the anterior margin of the premaxillary 0. 020

Depth from the condyle of the maxillary 0.026

Length of the crown of the premaxillary tooth- 0. 0042

Diameter of the crown of t he premaxillary tooth 0. 002

A fragment of a large Hat bone exhibits very delicate radiating grooves,
whicii are marked by spaced impressed dots.

From the North Fork of the Smoky River, Kansas, six miles south of
the town of Sheridan. Prof B. F. Mudge.

This species was about two-thirds the size of the species last described.

212

ICHTHYODECTES MULTIDENTATUS, Cope.

This species was first described from a preniaxillary aud part of a max-
illary bone, from tlie Smoky Hill River ; a second and more perfectly-pre-
served skeleton, obtained l)y Prof Merrill from tlie headwaters of the
Solomon River, adds much to our knowledge of it. The first-mentioned
specimen was described as follows:

" In this fish, we have the cimwex premaxillary of the larger species, with
more numerous (twelve) teeth than in any other of the genus. Those of
largest size are the first three, the last being small. The second and third
are about equally prominent, and more so than the first. The bone is much
contracted above ; there being an excavation on the anterior border, and con-
traction from behind. The superior edge is thin, and without trace of artic-
ular surface. Alveolar edge somewhat rugose. The maxillary is both
narrow and thin, but is only partially preserved. It bears five teeth on
O^.Ol. One of these, with complete crown, displays a longitudinal angle on
the antero-interior face. Length of the premaxilku-y, 0".039 ; depth of the
premaxillary (oblique), O-.O'ZS ; length of its tooth-line, C^.Q-ZS."

The second specimen is represented by a cranium with maxillary bone
and scapular arch, vertebrae and ventral spines, etc. The specific characters
are well exhibited in the dentition. There are thirty teeth and eighteen
empty alveoli in the maxillary bone; and the anterior apex of the same,
which is l)roken off, probably supported tour additional ones. The crowns
have a subround section, and differ from those of other species in being
marked with shallow sulci and longitudinal angles and ridges. The most
prominent ridges are, one on the middle of the external face, and one on the
outer side of the anterior face, but there may be two or three on the outer
face and on a few posterior teeth ; the outer face is smooth. The fragment
with tooth, above described as typical, belongs to the distal part of the jaw.
There are shallow grooves at the bases of the crowns of most of the teeth.

The cranium has an elevated crest, which stands on the narrow median
bone which I suppose to be supraoccipital. On each side, and just behind
the crest, the parietal bone rises as a posteriorly-directed process The
epiotic also forms a strong latero-posterior angle. The pterotic extends roof-
like on each side; its anterior part abruptly depressed below the posterior.
The postf rontal forms a short transverse process, which is preceded by a con-
cave excavation of its free border. The vertical position of the malleolar
process of the prefrontal shows that this species possessed the subvertical
mouth already ascribed to the Portheus molossus.

The scapula is much like that of Portheus, and in the specimen stands
on a level with the orbit. The ventral spine is of relatively large [)ropor-
tions, and its superficial dense layer is marked with rows of delicate, im-
pressed punctse, which are sometimes confluent into grooves. The first
cervical vertebra is unmodified aud grooveless ; the dorsal vertebrae are
grooved. Length of cranium, 0".154; diameter of first vertebra, U™.020;
depth of maxillary at middle, U'".()23 ; width of ventral spine, 0".020 ; length
of articular face of scapula, 0"'.031 ; width of clavicle below scapula, 0"'.U4U.

2KT

DAPTINUS, Cope.

Tliis gcims was proposed for Saurocephalus phlehotomus, Cope. TIk; (oiiii
of the crowns ol' 1 he tcelli is {\n\{ ot' Saurocejfhaliis and Erisichlhc; l)ii( if
differs from tlie lattcjr in llicir perfect ecpiality of size. It differs from the
fomier in the position of the nutritious foramina of the inner si(h; of the
dcntary hone; for, instead of forming an isohited series, as in tha( genus, tliey
only appear as notches on the inner margins of the alveoli.

The mode of articulation of the premaxillary and maxillary hones wilh
each other and with the prefrontals is similar to (hat already descrihcd in
the genus Portheus. The maxillary underlaps the premaxillary on its inner
face, forming a very extensive squamosal suture, and pres(!nts the two superior
condyles, the anterior for the ethmoid, the posterior for the palatine hones.
The palatine presents the malleolar segment which connects the maxillary
ami prefrontal articular faces, and then continues downward and hackward,
as a vertical lamina, without teeth on its free margin. Immediately in front
of eacli of these hones is a triangular element, compressed, with an acute
apex upward and truncate base downward, the posterior border in contact
with the anterior edge of the palatine. The vomer is toothless.

The anterior vertebrae present pits tor the neurapophyses, l>ut no others,
as far as the third. On the fourth and subsequent c(;ntra, tlicrc is a deep
lateral pit, and a smaller one above and behind it, near, and a little posterior
to the neurapophysial pit.

The teeth in Daptinus arc not a little like tliose of the existing genus
Fomatoimts, which includes the bluefish.

Daptinus pulebotomus. Cope.

Represented by all the tooth-bearing elements of three individuals, from-
distinct localities, with portions of crania and vertebra'. These show that
the jaws are long and slender, and that the teeth are closely set, and with
the roots l)ut little compressed, while the crowns are very much so. Eight
alveolae may be counted in fourteen millimeters. The dentary bones have a
vertical and transverse truncation at the symphysis, and are thickened so as
to afford strong attachment for something ; the absence of the usual oI)li(juity
in the one or the other direction is noteworthy. The teeth continue to the
symphysis, but of slightly reduced size. The internal groove of the dentary

214

is deep, and continued to near the symphysis. The cotylus of the articuhir
bone is j)rescnved largely backward.

The prcinaxillary is a large oval plato, gently convex on the outer side,
and of greater vertical than transverse extent. It differs from that of Porlheus
in lacking the condyle on the superior border. In the best-preserved speci-
men, a large piece is broken from its posterior alveolar border ; in that which
remains, nine alveoli remain.

In two of the specimens, the anterior part of the cranium is preserved.
The anterior part of the ethmoid is broad and convex above, and rapidly
contracts to an acute, flat apex. Laterally, it overhangs the fossa in whicli
the anterior maxillary condyle enters, atul furnishes a flat surface for the
posterior condyle of the same bone. The vomer is excavated laterally in
front, .so as to be somewhat cross-shaped, the apex being rather produced.
It is edentulous.

Measurements.

M.

No. 1, depth of the iirom.axillary 0.054

No. 1, depth of the maxillary at the middle 0.025

No. 1, depth of the triaugiilar bone of the palate 0. 038

No. 1, depth of the aympbysis of the mandibuli 0.026

No. 2, length of three anterior vertebra; « 0.040

No. 2, diameter of ouc anterior vertebra - 0. 010

No. 2, depth of the dentary 0.041

No. 2, depth of the .symphysis 0. 02G

Fo. 2, width of the ethmoid aliovo 0. 0105

No. 2, width of the craiiinm at the i>refroutal facets 0.022

No. 2, length of the muzzle to the prefrontal facets 0.023

A third specimen consists of some vertebrae and portions of the cranium;
the latter including the dentary, maxillary, part of the premaxillary, the
palatine and vomerine bone;?, compressed into a mass by pressure, the
separate pieces preserving nearly their normal relations. From the, latter,
the following characters may be derived :

Palatine bones toothless; teeth of both maxillary and dentary with
compressed crowns, wiiich are longer than wide at base, and closely placed;
those of the dentary twice as large as those of the maxillary. Maxillary bone
proximally deep ; dentary shallower; the maxillary with elongate suture with
the premaxillary behind.

The teeth are equilateral, without intermarginr.l groove or barb, and with
smooth enamel- surface, or only minutely striate under the microscope.
A series of larger foramina extends along the alveolar margin of the maxil

215

lary and dciitary bones, one loranicn 1«) each tootli. The alveolae are confluent

as they approach this margin.

There arc three vertebrae, which present two pairs of deep longitudina

grooves, viz, two on each side, two on the interior, and two on the superior

face of the bone ; the last receives the basal articulation of the ha3niapophyscs.

The centra arc crushed. Their nieasureinents, witJi thos(^ of the jaws, are as

follows :

ifcavurcments.

M.

Length of the cciitruui 0.025

LoDg (liatnctcr (crusbcd) 0. 035

Short diameter (crushed) 0.0175

Depth of the maxillary bono .'vntoriorly 0.031

Depth of fho dentary bone anteriorly i 0.015

Length of thu crown of the inferior tooth 0. OOG

Length of the crown of the superior tooth 0. 004G

Number of inferior tooth in O^.OI, :! ; number of aupcrior teeth in O'^.Ol, 4.5.

The vertebrae arc about as largo as those of a fully-grown "drum-fish,"
Pogonias.

From the yellow chalk of the Upper Cretaceous of Kansas, found on the
Solomon or Nepaholla River, Kansas, at a point one hundred and sixty miles
above its mouth, and in Trego and Rooks Counties, by Prof. ]J. F. Mudge,
jn'ofessor of natural science in the State Agricultural College of Kansas.

SAUROCEPHALUS, Harlan.

Leidy has pointed out the mode of implantation of the teeth in the
typical species of this genus. The mode of succession of the teeth has not
yet been indicated, but is well displayed in a specimen of the jaw of 6". ara-
pahovms. Cope. It is known from Harlan's description that a large Ibramen
issues on the inner wall of the jaw, opposite each root. The fractured ends
of the specimen exhibit the course of the canal which issues at this foramen.
It turns abruptly downward between the inner wall of the jaw and the fang
of the functional tooth, and not far from the foramen. Its course is inter-
rupted by the crown of the successional tooth. This is situated obliquely as
regards the long axis of the jaw.

It is thus plain that the successional ap|)earance of teeth is different in this
genus from what I have described in Portlieus and Iclithyodecles. In them,
the foramen is wanting, and the young crown rises within the pulp-cavity of
the functional teetli, as in the Crocodili/i. In this genus, on the other liand.

216

it is deveioped outside of the pulp-cavity and fang of the old loot li, and takes its
place, as in many Lacertilia and in the Tijllwnomorpha, by exciting the absorp-
tion of the latter. The conic form of these fimgs in Saurocej)luilus is ap|iro-
priate to such a succession, and their great length seems to preclude the
nutrition of the young tooth from their bases. The use of the foramina on
the inner face of the jaw is thus made apparent, viz, the nutrition of the
successional teeth from without. I cannot trace the canal below the crown
of the young tooth to the base of the pulp-cavity of the old tooth; and there
are canals in the jaw below the latter, one of which probably carried the
dental artery.

Species of this genus are less abundant in the part of Kansas examined
by me than those of the preceding genera. Two only have been observed up
to the present time, as follows : .

Saukocepiialus LANCiFORMis, Harlan, /. c.

Medical aud Physical Researches, 362 ; Leldy, Tiausactiona of the American Philosophical Society,
185G, plate. — Saurodoii lanci/ormis, Hays, Transactions of the American Philosophical Society,
1830, 47C.

Established on a right superior maxillary bone, from a locality near the
Missouri River. It differs Irom that of the other species in having a very
elongate superior suture with the premaxillary bone, and in the very short
dental crowns, which are as wide as deep. The largest species ; known from
the jaw only.

Saurocephalus arapaiiovius, Cope.

Established on a portion of a maxillary bone with a part of a suture, per-
haps for attachment to a supernumerary maxillary. The size of the species
is nearly that of S. lanci/ormis, and the crowns of the teeth are rather short,
as in that species, and less elongate than in D. 2'hleboio?iius. The teeth are
very closely set, and the fangs are separated by very narrow septa. The
crowns are expanded so that the edges overlap in some cases. The form of
these is much compressed ; width about equal to height ; the edges convex
and acute. The enamel is smooth and without facets. The I'oots are without
the facets shown by Leidy to exist in S. lanciformis, and appear to be longer
than in that species, exceeding the length of the crown nearly four times.
None are, however, perfectly enough exposed for complete measurement. As
usual, there is a large foramen opposite each fang, below the inner alveolar

217

margin; ami, between the latter and the scries ol' ioramina, the surface is
slightly cpuvex and minutely rugose.

Measurements.

M

|)i'|illi oCllir 1)1111(1 0. Olir)

Thickiioss iit Mill niHiwr'fiiiiul O.OOOri

Total Iciisth of ;i tooth (f) • ". 0"-i

Length of a crown 0. 004;j

Wi'Uliofacrowii .■ 0.0030

Number of tooth in ono inch, 8.

The size of this fish was probably about equal to that of the Ichtlujodcctes
anaides above described.

The type-specimen was found loose on a bluff of blue shaly limegtonc,
fifteen miles south of Fort Wallace. Kansas.

ERISICHTHE, Cope.

Proceedings of tho Academy of Philadelphia, 1W2, p. 260.

In this genus, the teeth are implanted in deep sockets as in other Sauro-
dontidce, and the subalveolar line of foramina seen in Saurocephalus is wanting.
The crowns of the teeth are compressed and knife-like, as in Daptinus; l)ut
those of the anterior parts of the dentary and maxillary Iiones are greatly
enlarged. Maxillary bone short, and rapidly tapering to a narrow edentulous
extremity. Greater part of the dentary with a rugose band on the inner
side of the teeth ; its distal portion with a row of small compressed teeth,
separating the large teeth into two areas.

While this genus agrees with Portheus and Ichthijodectes in the absence
of nutritious dental foramina on the inner face of the dentary bone, and
especially with Portheus in the irregular sizes of the teeth, the crowns are
compressed and knife-like, and closely similar to those of Saurocephalus.
But the form of the maxillary is so different from anything known among
Saurodonlidce as to render it probable that the genus pertains to another
family-division.

The Portheus angulatus. Cope, from North Carolina, perhaps belongs
to the genus Erisichthe, differing from E. nitida in its greater size and less
degree of compression of the crowns of the large fangs.

EuiSICHTHK NITIDA, Cope.

Represented l)y numerous portions of a cranium with a fragment of a
pectoral ray, discovered l)y Prof B. F. Mudgc near the .SoIoukmi River,
28 c

318

Kansas. Tlic, ray \s of (he compound character already described as belong-
ing to otlier genera of