Fune 2, 1870| 
NATURE 
8I 
is that of a great nation of common descent, not divided 
by one of the subjects into the arbitrary “ provinces” 
of a kingdom, or “classes” of an army, but falling 
naturally into “alliances” which are those of blood, 
“families” which depend on common parentage, and 
“orders” which are hereditary. Now it is a great point 
gained to know that, probably, at least, such natural and 
convenient language is no metaphor at all, but strictly and 
literally true. 
Prof. Rolleston follows Gegenbaur in elevating Echixo- 
dermata to the rank of a primary division, and its various 
orders to classes ; and also in including under the head 
Vermes, not only the Rotifera and Helminthes, which, as 
“Scolecida,” form, with Echinodermata, Prof. Huxley’s 
Annuloid group; but the Azzzzlata as well, which have 
been associated with the Arthropod classes by all 
other naturalists since Cuvier (see the long and valuable 
note on pp. 152-157). Onthe other hand, 7wmzcata and 
Polyzoa are not also placed in the same heterogeneous 
crowd of “ Wiirmer,” but are retained in their probably 
more natural position among the Mollusks. Indeed, the 
classes included by Dr. Rolleston under Vertebrata, 
Arthropoda, Mollusca, Coelenterata, and Protozoa, are 
almost precisely the same as those recognised by Prof. 
Huxley, and generally admitted in this country. The 
only variationis in re-admitting Radiolaria among Rhizo- 
pods, and making “Ctenophorae” (why not Cteno- 
phora ?) a separate class, instead of an order of Anthozoa. 
The “characteristics of Vertebrata” are given with the 
fulness and accuracy which mark the author’s work. 
The eight closely-printed pages devoted to this section 
may be advantageously compared with the very short, 
but masterly account of the same group in Prof. Huxley’s 
‘Introduction to the Classification of Animals.” Indeed, 
we would advise all students to read thoroughly the latter 
work before beginning the one under review, which would 
then admirably fill up the details of the London professor’s 
sketches. At p. xxxv., it is implied that the kidney of 
Amphibia and most fishes answers to the permanent one 
of other vertebrates ; its homology with the Wolffian body 
is, however, rightly stated at p. Ixiv. We would suggest 
that the terms “outer” and “inner,” are better than 
“uppermost” and “ lowermost,” to denote the serous and 
mucous layers of the germinal membrane ; and at p. xl. 
“ventral” should replace “ anterior,”—a term which con- 
fuses by mixing up the relations of human anatomy with 
those*of general zootomy. We are not surprised that 
Dr. Rolleston refuses to accept Hamatocrya as a natural 
group of Vertebrata ; he follows Prof. Huxley’s division into 
Branchiata and Abranchiata, and of the latter into Mam- 
malia and Sauropsida. Anammiota is a better alternative 
name for Branchiata than Anallantoidea ; not only for 
euphony, but because an allantois is certainly present to 
some extent in certain Ichthyopsida, while its develop- 
ment and importance in mammals is far less than in birds 
and reptiles. In describing the characters of Mammalia, 
we notice that the author endorses Prof. Huxley’s revised 
opinion that the malleus, not the incus, represents the 
quadrate bone, although the opposite is stated at p. 25 of 
the second part ; but he appears to regard the marsupial 
bones as part of the pelvis, and not as mere ossifications of 
the internal pillar of the abdominal ring, The orders of 
the vertebrate classes are adopted from those given in 
the “Introduction to Classification” before referred to. 
That most of these will receive general assent there 
can be little doubt; but with respect to the placental 
classification of Mammalia, we venture to suggest some 
objections. In the first place the structure of the uterus 
and placenta is not strictly an embryonic character, but 
belongs to the parent organs of generation, which, in other 
classes, are found to be of minor importance in classifica- 
tion. ‘Then it is very difficult to get accurate information 
as to the condition of the placenta—opportunities for ob- 
serving parturition are, of course, much less frequent than 
for studying almost any other process—so that it will 
be long before we have any facts as to whole groups of 
animals, ¢,g., the Sirenia. Moreover, if here opportunity is 
fleeting, judgment would seem to be often difficult : thus 
the placentation of so common a mammal as the rat was 
completely misunderstood by an eminent naturalist, until 
his account was corrected by Prof. Rolleston’s own dissec- 
tions. Compared with the skeleton, the teeth, or even 
the brain, the placenta is a far less available criterion, 
and far more liable to misinterpretation. But even if this 
were not the case, a grave objection to the placental 
classification remains in the fact that it necessarily ex- 
cludes all fossil forms, the study of which has been so 
well used as a help in tracing the affinities of living 
animals, and by none more than Prof. Huxley himself. 
Lastly, judging this system by the test of concomitant 
variation, it is surely sufficiently condemned by com- 
pelling together animals so different as Primates and 
Rodentia, Orycteropus and Simia, while it separates 
Hyrax from Rodents and Perissodactyla, to unite it with 
Carnivora. The better plan we conceive is that followed 
by Prof. Flower in his recent lectures, to place the several 
orders in as natural juxtaposition as may be, and to put 
placental characters on the same level only as those 
afforded by the brain or the extremities. 
In the careful description of the Mollusca which follows, 
Dr. Rolleston does not divide the Gasteropoda into two 
classes ; nor does he admit the constancy of the primary 
flexures of the intestine, which Professor Huxley has 
made an important criterion of all the Molluscous series. 
(See pp. 58, 68, and 235.) In the Asczdza this question of 
the intestinal flexure depends upon the view taken 
of their great branchial sac. Dr. Rolleston does not admit 
Prof. Huxley’s theory of its being morphologically a pha- 
rynx, but with Mr. Hancock regards it “as homologous not 
with a dilated pharynx, but with the branchial cavity, and 
the inhalent aperture to represent not the mouth, but the 
inhalent syphon of the Lamellibranchiata” (p. 69). If this 
view be accepted, it unites the Molluscoidea more closely 
to the Mollusca proper, and is an additional argument 
against their association with Vermes. The remarkable 
observations of Kowalewsky and Kupffer on the resem- 
blances to vertebrate structures in the larva of Phalusia 
are duly noted (pp. ci. ciii.) Dr. Rolleston says of the 
Arthropoda that they “have frequently been classed 
together with more or fewer of the Vermes in one sub- 
kingdom, that of the ‘Annulosa ;’ and whilst by such 
highly organised forms as the marine Polycheta an ap- 
proximation appears to be made to certain of the less 
specialised of the Crustacea ; or even of the Myriopoda, 
or the larvee of insects, amongst the air-breathing Arthro- 
poda ; the microscopic Rotifera connect the Vermes, to 
