564 SCIENCE PROGRESS 



with a Huxley or a Kelvin. He means, I suppose, a Hegel or a Spencer, for 

 Kelvin and Huxley were essentially specialists. But the value of the works of 

 philosophers such as these lies not so much in their systems as in their profound 

 knowledge and insight, in their grip of the knowledge of their time, in the fact 

 that they understood the principles of science more clearly than the men of 

 science themselves, and were sufficiently well acquainted with the details. Kant 

 was a physicist before he became a philosopher. Spencer would have achieved 

 eminence on biological work alone. It is not much use putting forward systems 

 unless one knows enough to know when one is not talking sense. 



Mr. Jamyn Brooks has written a work on psychology which, it seems, has 

 been well received. If his work in that department is unsound, he is covered by 

 the fact that the whole science is a little vague and shadowy. The reviewer 

 would suggest that it would be better if he concentrated on the psychological 

 side and if he did not attempt to deal with problems of natural science until he 

 has acquired a sound elementary knowledge of the sciences with which he deals. 

 Let us assume that the author's hypotheses are all true and valuable (to the 

 reviewer they scarcely appear so), it would still require the ability and the 

 knowledge of a Hegel or a Spencer to set them forth in detail. The present 

 volume provides no evidence that Mr. Brooks possesses either, and, if he does 

 not, the very existence of his book is a weapon in the hands of the " stodgy " 

 man of science who is impervious to new ideas. Who was it called van 't HofPs 

 chemistry in space the vapouring of an unsound mind ? The type always exists, 

 and a book like that of Mr. Brooks is so much grist to his mill. Those who think 

 they are the originators of new ideas may look at this volume and decide that, 

 after all, it is safer to leave it alone. 



H. S. S. 



Vectorial Mechanics. By L. Silberstein. [Pp. vi + 197.] (Macmillan <fcCo., 

 19 1 3. Price 7s. bd. net.) 



The work of Heaviside, which has demonstrated so clearly the power and relative 

 simplicity of vector methods for dealing with quantities essentially vectorial, is 

 bearing good fruit. There are already in Germany good textbooks, such as those 

 of Bucherer and Gans, which give an introduction to the methods of vector 

 analysis. In the book under review we welcome at last an English book in which 

 a systematic account is given of the vector analysis in use among the physicists of 

 the present day, and its applications to mechanics. Even to those with a good 

 knowledge of the subject it will prove very interesting, as there is a distinct 

 originality of treatment and a unity and sequence which make a strong appeal. 



The work is divided into six chapters. We have first a general introduction to 

 vector analysis, which is probably the best in any English textbook. In this the 

 vector and scalar products, the curl, divergence, and the important theorems con- 

 nected with them are exposed. There follow three chapters, under the headings 

 of General Principles, Special Principles, and Rigid Dynamic, dealing with the 

 application of the analysis to dynamics, those parts, such as the motion of a body 

 under no forces, which best lend themselves to vector methods being naturally 

 treated in the greatest detail. The final two chapters deal with the theory of 

 elasticity and hydrodynamics. There is a useful and instructive appendix, giving 

 the most important equations of the book, together with their Cartesian equiva- 

 lents. The linear vector operator, of which, outside Heaviside's work, it is hard to 

 find an account, is introduced in connection with moments of inertia, and developed 



