98 
vertebra, certainly through the cervicals and lumbars. The 
limbs were massive and heavy, the bones, like those of the 
rest of the skeleton, being nearly or quite solid. The fore- 
foot was larger than the hind-foot, its component bones 
being comparatively short and massive, with five well- 
developed digits, as in Proboscidians, but the carpal 
bones interlocked with the metacarpals as in Perisso- 
dactyls. The feet, as in the modern elephant, were planti- 
grade, and were doubtless covered below with a thick 
pad. 
We can picture these dull, heavy, slow-moving creatures 
haunting the forests and palm-jungles around the margin 
of a great lake. Into the quiet depths of that lake their 
carcases from time to time found their way, swept down 
perhaps by river-floods. Among their contemporaries 
were other forms whose remains have also been more or 
less abundantly preserved in the same deposits. Of these, 
two genera next in size to the Dinocerata were Perissodac- 
tyl ungulates somewhat larger than a tapir (Pal@osyops 
and Lzmnohyus). Another interesting form is Orohippus 
—a four-toed ancestor of the living horse, while additional 
varieties of the ungulate type were related, though 
distantly, to the tapir and rhinoceros (Colonoceras, 
Helaletes, Hyrachyus). Two remarkable genera (7Zz?/o- 
thertum, Stylinodon), nearly as large as a tapir, possessed 
characters resembling those of the ungulates, carnivores, 
and rodents, and have been embraced by the author in a 
new order called by him 77//odontia. Among the carni- 
vores there was one (Zz7zo/elzs) nearly as large as a lion, 
and another hardly less in size (Oveocyon), while Dromo- 
cyon was somewhat smaller and ZLznocyon about as 
large as a fox. There were likewise lemurs having some 
affinities with South American marmosets ; also repre- 
sentatives of the Marsupials, Insectivora, Chiroptera, and 
Rodentia, but no true Quadrumana or Edentates. Rep- 
tiles abounded, especially crocodiles, turtles, lizards, and 
serpents, while fishes of many kinds swam in the lake. 
The structure and history of the Deinocerata with 
their place and affinities in the animal kingdom are 
fully discussed in this important monograph. Like his 
previous work on toothed birds in the same series of 
memoirs, Prof. Marsh’s present volume is an admirably 
executed and exhaustive research. Every bone is care- 
fully worked out and drawn. Every available fragment 
of evidence is patiently collected, compared, and tabu- 
lated. Whatever may be disputable regarding the 
conclusions drawn, there can be no variety of opinion as 
to the actual data. No fewer than fifty-six lithographic 
plates, and nearly 200 woodcuts depict with singular 
fidelity every part of the skeleton of the Deinocerata as 
at present known. 
But Prof. Marsh is much more than a comparative 
anatomist. It is not enough for him to describe the 
bones he has unearthed, and to point out their analogies 
in the living world. He is instinctively an evolutionist, 
and every extinct animal seems to propound to him the 
problem of its ancestry and its descendants. One of the 
most suggestive chapters in his present memoir is devoted 
to the genealogy of ungulate animals, and the place of 
the Deinocerata among them. He believes that from 
some primitive form, of generalised type, probably small 
in size, resembling generally an insectivore, and going 
back at least as far as Permian time, all the mammalian 
NATURE 
[| ¥une 4, 1885 
tribes have descended. Such a genealogical mammal, 
belonging to Prof. Huxley’s group of Hyfotheria, would 
possess all the general characters of the subsequently 
developed mammalian orders. But special characters, 
acquired in adaptation to conditions of environment, 
would be developed in the course of time, and would lead 
to the establishment of different modified types. The 
general characters would thus alone be a safe guide in 
tracing a community of ancestry, while those of a special 
kind need not necessarily indicate affinity, but may have 
independently arisen from the influence of the same sur- 
roundings in groups already quite distinct from each 
other. In the Cretaceous system, a well-marked group of 
mammals is found which is represented now by the 
living Hyrax, along what appears to have been 
the main stem of ungulate descent. From this 
stem, after the remarkable waning of reptilian life 
at the close of the Mesozoic ages, there diverged, in 
Cretaceous times, a branch which terminated in Cory- 
phodon—a tapir-like form which, both in America and 
in Europe, probably quite equalled if it did not surpass 
in size and power any of the representatives of the fading 
reptilian types of an older creation. Another branch 
which may have been given off about the same time 
reached its full development in the Deinocerata, which 
were certainly the monarchs of the region where they 
lived. But nothing is more striking in the history of 
these and the other colossal mammals than the rapidity 
with which they appear and disappear from the scene. 
Dinoceras and its allies, so far as the evidence yet goes, 
appear to have been restricted to the middle part of the 
Eocene period. Their remains are not found in the 
earlier deposits of that period, and cease to occur before 
we reach the upper parts of the series. The cause of this 
speedy extinction is to be sought, according to Prof. 
Marsh, in the small brain of the animals, their highly 
specialised characters, and huge bulk, whereby they were 
unfitted for adapting themselves with sufficient rapidity 
to new conditions ; and a change of surroundings brought 
about their extinction. But this is a point on which the 
geologist may not unnaturally claim to be heard when he 
demands some evidence of such change of surroundings. 
Had the supposed geological vicissitudes been sufficiently 
serious to cause the extinction of a whole tribe or sub- 
order of large mammals, they might have been expected 
to have left sonie palpable evidence of their passage in a 
corresponding change in the nature of the deposits accu- 
mulated in the lakes. But there is certainly nothing in 
the nature or succession of these deposits to suggest that 
any important modifications of topography or climate 
took place during the time when they were being de- 
posited. On the contrary, they seem to point to pro- 
tracted uniformity in the conditions of sedimentation. 
They afford no indication whatever that the successive 
appearance of Coryphodon, Dinoceras, and Diplacodon 
was accompanied, far less was determined by, any essential 
change of physical conditions. That such change actually 
took place is of course quite conceivable, but when it is 
demanded as an essential factor in mammalian evolution, 
some admissible proof may very fairly be demanded. 
Like Prof. Marsh’s previous memoir on “Odontor- 
nithes,” the present volume may be regarded as a model 
monograph. It is complete without being overloaded, 
