AvuGuSsT 11, 1899. ] 
cation, or that the new systems proposed 
are better than the Dewey system, already 
used in the bibliographies of several sciences. 
In physiology, for example, there are some 
800 classes into which it is expected that 
8,500 entries per annum will be sorted. 
Cards would need to be duplicated many 
times to be placed in all the sub-divisions to 
which an article may refer. The schedules 
of classifications are open to many criticisms. 
We recommend that the classification of 
each science be referred to committees who 
shall especially consider the classifications 
in existing bibliographies. 
J. McKeen Carrety, 
H. F. Ossorn, 
R. S. Woopwarp. 
REPORTS ON THE SEPARATE SCHEDULES. 
A. MATHEMATICS. 
1. ‘A’ DESIGNATES in one place Mathe- 
“ matics, in another ‘ Pure Mathematics.’ No 
provision is made for Mechanics. 
2. Under Bibliography a list and de- 
scription of the various mathematical jour- 
nals, especially of the bibliographical jour- 
nals might well be included. 
3. Under Arithmetic a sub-division 
might well be devoted to continuity, count- 
ability, ete. (Jordan, Canton, Stoltz). 
4. Under Algebra and Theory of Equa- 
tion, 1250, Klein’s theory of the reduction 
of the solution of equation to the theory of 
linear groups seems to have been over- 
looked. In fact, this number (1250) rep- 
resents far too much, and ought to be 
divided. 
5. Under Groups a sub-division might 
be given to congruence groups. No. 2010 
includes too much. 
6. Under Calculus belong mean value 
and probability. The latter is unprovided 
for anywhere. 
7. 5210, the title is ambiguous. 
8. Under Analytical Plane Geometry 
there should be a sub-division on co- 
SCIENCE. 
167 
ordinate system, projection, metrical geom- 
etry, etc. 
F. N. Cone. 
B. ASTRONOMY. 
(No schedule submitted. ) 
B. PHYSICS. 
THE very greatest care should be used in 
the details of the classification. 
Everything of interest should appear 
once, and only once, and then in its natural 
association. 
Upon this point rests the working value 
of the lists. A class should be sub-divided 
in proportion to its natural sub-divisions 
and not to the amount of literature covered. 
Primary Divisions. 
Bibliography should have a section un- 
confused by others. 
Dynamies should not be separated from 
the theories of matter andether. Confusion 
is sure to arise when ‘heat’ is separated 
from ‘thermal effects’ and from ‘ invisible 
radiations.’ 
I. Bibliography. 
II. Dynamics of solids, liquids and gases, includ- 
ing vibrations and wave motion. 
III. Heat. Including temperature, specific and 
latent heat. 
IV. Radiant energy. Including radiant heat, 
light and ultraviolet. 
V. Electrochemistry and electrolysis. 
VI. Electricity and magnetism. The title electro- 
magnetism is misleading ; it applies now to a small 
section of the subject. 
It may be argued that any classification 
will be artificial and each person must learn 
the classification of his section. But the 
value of the lists in libraries will lie almost 
entirely in the opportunity they offer an in- 
vestigator to look over the literature in re- 
lated divisions. 
For example, the physicist cannot be ex- 
pected to keep posted in the classifications 
of mathematics, mechanics, chemistry, 
crystallography, psychology, and possibly 
