82 SCIENCE. 
The second half of this volume is given up toa 
discussion of equilibrium in systems containing 
one solution phase and at least one solid phase. 
Volume IY. is devoted to what Duhem calls 
‘double mixtures’ and to general equilibrium 
in heterogeneous systems. By ‘double mix- 
tures’ Duhem means two component systems 
containing at least two phases of variable com- 
position. Under this head come fractional dis- 
tillation, critical states of mixtures, liquefac- 
tion of mixed gases and systems containing two 
liquid phases. A great deal of space is devoted 
to a consideration of the alleged law that the 
vapor-pressure of a dineric system is the same 
as that of the more volatile component. The 
volume closes with a general discussion of the 
phase rule, in the course of which it is pointed 
out that the classification followed throughout 
the four volumes hasbeen based on the phase 
rule and that all good classifications must be so 
based. This is very satisfactory, but it would 
have been more satisfactory if we had been 
told this at the beginning of the first volume 
instead of at the end of the fourth. One great 
fault in all of Duhem’s writings is his refusal 
to tell the reader what is to be proved. The 
result is that the reasons for the single steps do 
not become clear until the second reading. 
From the Baconian point of view it is very 
pretty to marshal the facts in a splendid array 
and then to point out the general law of which 
they are special illustrations, but Bacon is not 
famous as a successful writer of text-books. It 
would have been very much simpler to have de- 
duced the phase rule and then to have pointed 
out the way in which it should be applied. As 
far as the qualitative equilibria are concerned, 
this is also the historical method. Gibbs de- 
duced the phase rule as a general theorem, and 
Roozeboom has, since then, shown its valueasa 
guide. 
These four volumes of Duhem’s constitute a 
monumental work and will be of immense ser- 
vice. On the other hand, it would easily be 
possible to overestimate their value. What we 
have is an exhaustive study of chemical equilib- 
rium put into mathematical form and expressed 
in terms of the thermodynamic potential. This 
application of mathematics to chemistry is un- 
fortunately more ornamental than useful. There 
[N. S. Voz. X. No. 238. 
are myriads of formulas, but very few can be 
applied to any concrete case. The book is 
really only amathematical outline in which 
formulas are indicated. The equations contain 
unknown functions. To the experimental 
theorist the book is a joy and a sorrow, a joy 
because it points out so much and a sorrow be- 
cause it always stops short of becoming practi- 
cal. One of the most striking features about 
Helmholtz was the fact that he cast his theoret- 
ical speculations into such a form that they 
could be tested quantitatively. Duhem has 
never done this in physical chemistry. He has 
done brilliant work, but his theory has always 
been qualitative and not quantitative theory. 
If anyone doubts this he has only to read the 
four volumes of the Mécanique chimique and he 
will be convinced. 
WILDER D. BANCROFT. 
Le céramique ancienne et moderne. Par E. 
GUIGNET et EDOUARD GARNIER. Paris, 
Felix Alcan, 108 Boulevard Saint-Germain. 
1899. 8vo. 69 figs. Pp. 311. 
This volume is No. 90 of the series ‘ Biblio- 
théque Scientifique Internationale,’ edited by 
M. Em. Alglave. Its authorship is in collab- 
oration by MM. E. Guignet and Edouard Gar- 
nier. The work consists of two grand divisions, 
the first, by the Director of the Dyeing and 
Coloring Department of the Gobelin and Beau- 
vais Tapestry Manufactories, relates to the 
fabrication of ceramics; the second, by the 
Conservateur of the Museum of the Pottery 
and Porcelain Manufactories at Sevres, is on the 
history of Ceramics. 
The first part deals with the material of 
which pottery and porcelain is made, describes 
it at length, shows the differences between the 
different products, gives by analysis the compo- 
nent parts of the various materials required for 
these products, and describes their mode of 
treatment and preparation for use. It presents 
by elaborate definitions the different kinds of 
ceramics, and shows principally by chapters, the 
differences between pottery, faience and porce- 
lain. It represents by description and diagram 
the machinery used in the treatment of the ma- 
terial, in the fabrication and forming of the 
objects, and the ovens in which they are baked. 
