872 
forces,” of ‘‘ eviction,” and of “resilience.” In his 
discussions of meteorological phenomena these principles 
recur repeatedly. We meet with an example of the 
application of the principle of resilience on the first 
page of the first lecture, when we read this characteristic 
remark concerning the conditions for the formation 
of orographic rainfall; “‘ But when you come to think 
of it, the explanation requires that the air on the 
windward side has to be made to flow up-hill, and no 
fluid which technically must be called heavy, as it 
is affected by gravity, even if it is as light as air, 
flows up-hill without protest. It prefers to go round, 
and will exhaust all the possibilities of doing so before 
submitting to be driven over.” This principle of 
resilience should be remembered not least by mathe- 
maticians who will work out the theory of atmospheric 
movements. In theoretical hydrodynamics we generally 
assume the equation s=f(p), density as a function 
of the pressure. This equation leads to that state 
of “unlimited miscibility ’’ which excludes resilience 
and would make it possible for the air to flow up-hill 
without protest. But the true equation s=/(p, @) 
permits the air to take a stable stratification, a per- 
mission of which it makes a most extensive use ; 
with the consequence that we have laws of motion 
very different from those of the idealised fluid, in 
which s=f(p). I can scarcely be wrong when I say 
that this equation has for more than a century acted 
as a barrier which has prevented the representatives 
of theoretical hydrodynamics from taking up meteoro- 
logical problems with success. 
The principle of ‘balanced forces’? merits great 
attention, not only in qualitative discussions but also 
perhaps still more by the attempts to work out mathe- 
matical theories of atmospheric motions. The author 
gives no mathematical formulation of the principle. 
But if I have understood him rightly, I should call 
it rather very good advice than a principle. The 
most obvious way of developing atmospheric move- 
ments might seem to be this: first to consider the 
state of equilibrium, and then to examine the con- 
sequences of a disturbance of it, in the case before us 
of a disturbance of thermal origin. But on account 
of the rotation of the earth there is a very long and 
difficult way from the state of equilibrium relative 
to the earth to the ultimately resulting motion. The 
primary tendency is the production of a direct flow 
from the cold to the warm areas. But this tendency 
is almost completely checked by the effect of the 
earth’s rotation. Instead of the direct flow from cold 
to warm areas, we get a circulation cyclonic round 
the warm and anticyclonic round the cold areas. 
Only a small residual leakage is left, conveying very 
gradually air from the cold to the warm areas. For 
NO. 2800, VOL. 111] 
NATURE 

[JUNE 30, 1923 
this leakage the process of “eviction” plays an 
important part. 
Now Sir Napier Shaw’s advice is to shorten this 
long development, which it is very difficult to give 
in a satisfactory form, and tg start with that state 
of steady motion relative to the earth which is char- 
acterised by the “gradient wind.” This gradient 
wind, by which pressure gradient and deflecting force 
balance each other, gives immediately the cyclonic 
circulation round the warm areas and the anticyclonic 
circulation round the cold. Then, in the second 
approximation, we have to add to this pure circulation 
the further disturbances, as those connected with — 
convection and that particular form of convection for 
which Sir Napier Shaw has introduced the word eviction, 
No doubt his advice will be followed more and more, 
both in elementary treatises and especially in mathe- 
matical theories, for which this may prove to be the 
only practicable way. 
I cannot finish this notice without mentioning Sir 
Napier Shaw’s brilliant style of writing, the many ade- 
quate expressions and striking comparisons by which 
he succeeds in making the subject clear and ensures 
that the reader does not forget the main points. He 
is a master of finding the right words, and is not less 
a master of illustrating the text by characteristic 
diagrams. I only regret that he has not found place 
in the book for that really historical diagram by which 
he formulated his protest against the old cyclone 
theory and gave the main structure of the new one, 
replacing the fine logarithmic spirals running 
asymptotically to a centre in the old model simply 
by three sets of straight lines, of which two sets meet 
each other at right angles. V. BJERKNES. 

Philosophy for Men of Science. 
Scientific Thought. By Prof. C. D. Broad. (Inter- 
national Library of Psychology, Philosophy, and 
Scientific Method.) Pp. 555. (London: Kegan 
Paul and Co., Ltd.; New York: Harcourt, Brace 
and Co., Inc., 1923.) 16s. net. 
S men of science are usually impatient, if not con- 
temptuous, of philosophical discussion, Prof. 
Broad may be thought rash to address a philosophical 
work specifically to them, particularly as he is occupied 
in discussing the notions of space, time, matter, and 
motion, about which the man in the laboratory con- 
siders himself better qualified to speak than the philo- 
sopher, The author, however, brings to his task both 
a knowledge of mathematics and physics and an appre- 
ciation of the efforts of philosophers in the “ peculiarly 
obstinate attempt to think clearly,’ which constitutes 
their chief task. Moreover, unlike many philosophers 
—— 
i i age ne 
