292 
discussion of certain maximum and minimum properties 
leads to considerations of the changes which take place 
in an evsemble of system both when left to itself and 
when subjected to external influence, also of the results 
obtained by bringing two canonically distributed e7- 
sembles within influence of each other. The general 
conclusion is that there exist in statistical mechanics 
processes strictly analogous to many of those occurring 
in thermodynamics. Thus equations may be formulated 
closely resembling those which represent the irreversible 
heat-changes between two bodies of unequal temperature. 
When it comes to choosing a pair of conjugate variables 
to represent temperature and entropy, it is found that 
these are not uniquely determined, but that several 
systems are possible, a fact previously brought out, 
indeed, by von Helmholtz in his “ Statics of Monocyclic 
Systems.” 
The last chapter deals with evsemdbles analogous to 
mixtures of different kinds of molecules, and these the 
author calls grands ensembles. They differ from the 
petits ensembles previously considered in the fact that 
they contain particles or systems of different kinds which 
may be present in different numbers. 
Prof. Gibbs’s work is not very easy to read, and it 
hardly seems appropriate to apply the title “elementary ” 
tu it ; but the difficulties are no doubt inherent in the 
subject. It does much to elucidate the conditions under 
which a body composed of molecules obeying the equa- 
tions of rational dynamics presents to beings of compara- 
tive dimensions similar to those of the human race 
attributes which may be summed up in the single word 
“temperature.” G. H. BRYAN. 
AN ATTEMPT AT ORIGINALITY IN THE 
TEACHING OF ZOOLOGY. 
A Course in Invertebrate Zoology. By Henry Sherring 
Pratt, Ph.D. Pp. xii +210. (Boston, U.S.A., and 
London: Ginn and Co., the Athenzeum Press, 1902.) 
R. PRATT’S book, defined on its title-page as 
a guide to the dissection and comparative study 
of invertebrate animals, is the latest of the many 
novelties which aim at effecting an improvement 
on the world-famous Huxleyean system, to which 
acknowledgment is made. The author sets _ out 
with the intention of enabling the student to study 
the larger groups as a whole, instead of detached 
types of different groups, as he claims is now generally 
done. In order to achieve this end, he deals in 174 pp. 
with no fewer than thirty-four representative animals, and 
the headlines of some of his chapters even bear the 
names of two alternative genera, for which a single de- 
scription is made to suffice. Although this gives an 
average of little more than five pages for each animal, 
it must be admitted that, so far as they go, the de- 
scriptions and instructions, of necessity of a very 
elementary form, are lucid and correct. 
Without going into further detail concerning the body 
of the book, it may be said that the essence of its 
novelty lies rather in the appendix and its associated 
classificatory scheme. This leads off with a copy of 
Claus’s 1887 system, in which, as an all-conspicuous 
NO. 1708, VOL. 66] 
NATURE 
[JuLy 24, 1902 
feature, the Sponges were classed as Ccelenterates, the 
Enteropneusta as Echinoderms. Then follows a short, 
but withal a useful, historical sketch of the growth of 
classificatory systems, from Cuvier to Hatschek, whose 
scheme of 1888 is given in tabular form, with a succeed- 
ing list of “short definitions” which are supposed to 
be ez suite, and of which it is remarked that while not 
exhaustive they ‘“‘are intended but simply to characterise 
the various groups in the fewest possible words.” The 
first great subdivision is into subkingdoms (Protozoa 
and Metazoa), divisions follow, then types, classes, and 
orders. When, however, on comparison, one finds that 
while the table provides for five types (Spongiaria, 
Cnidaria, Trochozoa, Echinodermata, and Chordata), 
the three first-named are for the definitions num- 
bered in order, and the two last-named are numbered 
five and six, one is led to seek for number four. The 
search is vain, since table and definitions do not agree. 
Most of the descriptions, moreover, in their would-be 
conciseness, are inadequate. And when with this it is 
said that, under type Trochozoa, defined as “ Meta- 
gastrozoans whose common descent and relationship 
are shown by their possession in some form of a tro- 
chophore larva or of an embryonic form allied to it,” 
there are included as subtypes Vermes, Articulata, and 
Mollusca, further comment becomes unnecessary, except 
to give it as our opinion that whatever the future of the 
zoological training of the young, it will not develop on 
these lines. 
The above analysis might be extended with even 
humorous results; but whatever the good points the 
book may possess, failure appears to us certain in the 
attempt todo too much. The would-be new departure 
is foreign to the best traditions of the Huxleyean system. 
In the later development of this, the thorough mastery 
of type-structure has come tc be regarded as an alphabet, 
by which the student learns to read, and the broadest 
possible survey of the structural limitations of the several 
groups of which the types are members, as a reading 
lesson to follow, under the special guidance of the 
teacher. 
OUR BOOK SHELF. 
Slide Rule Notes. By Lieut.-Colonel H. C. Dunlop, 
R.F.A., Professor of Artillery at the Ordnance College, 
and C. S. Jackson, M.A. (Barrister-at-Law), Instructor 
in Mathematics R.M. Academy, late Scholar Trinity 
College, Cambridge. Pp. 66. (London: Simpkin, 
Marshall, Hamilton, Kent and Co., Ltd., 1901.) 
THE slide rule is one of those things which can be less 
readily explained in writing than verbally. A few words 
explaining the principle so as to develop the slide rule 
sense is all that is required to put anyone of reasonable 
quickness in the way of becoming an adept. On the 
other hand, the full exposition of the logarithmic theory 
of the mode of setting for each class of operation, which 
is essential where the art is to be taught from a book, 
makes the thing seem so complicated and difficult to 
remember, that many who would find no difficulty in 
being taught by the first method might well give it up in 
despair at the very outset when taught only by the second 
method. However, it is not given to everyone to be 
able to find an adept with a power of exposition, and so 
the book becomes a necessity. 
