REVIEWS 151 



to be given at length in several contributions from the Mount Wilson Observatory 

 which will appear in the course of a few months, criticism of the results at present 

 would be inadvisable even if possible. These results are of such fundamental 

 importance that more detailed reference will be made to them when full par- 

 ticulars are available. It may be stated, however, that they revolutionise our 

 conception of our stellar universe. Various lines of argument lead Shapley almost 

 naturally to the conclusion that the globular clusters are all members of our 

 galactic organisation, and that they outline its extent and arrangement. The 

 immense distances at which many globular clusters are situated have been 

 referred to under (1). Adopting this view of the stellar system, it follows that all 

 known sidereal objects — including globular clusters and such structures as the 

 Magellanic Clouds — are parts of one enormous unit whose volume is more than 

 one hundred thousand times that commonly assigned to the stellar universe. 

 Various lines of argument had appeared to lead to the conclusion that the spiral 

 nebulas were separate stellar universes, and an analogy between their structure 

 and the structure of the nearer part of our galactic system was pointed out. But 

 such estimates of their distance as have been made (see Science PROGRESS, 

 p. 551, April 1918) are of the order of distance deduced for the globular clusters. 

 If, then, the latter are members of our universe, so also must the nebulas be. It 

 appears, therefore, as though the " island universe " theory has been seriously 

 shattered by Shapley. A more definite conclusion must be withheld until his 

 detailed investigations have been published. 



H. S. J. 



PHYSICS 



The Theory of Electricity. By G. H. Livens, M. A. [Pp. 717 + vi.] (Cambridge 

 University Press, 1910. Price 30s. net.) 



This is a general textbook on the mathematical aspects of modern electrical 

 theory. It is based mainly on the original Faraday- Maxwell theory. This theory 

 has, of course, been generalised to meet the growing body of experimental 

 knowledge and to embrace the concept of the electron and the electro-dynamics 

 of moving systems. In this generalisation the author follows the views of Sir 

 J. Larmor as expounded in his lectures and papers and his book, /Ether and 

 Matter, on the ground that although "this form of the theory has been almost 

 entirely abandoned in recent accounts of the subject, it remains the only one 

 which appears to be completely satisfactory from the point of view of mathematical 

 and physical consistency, and in its generality it is unapproached by any other 

 form." 



The book forms a very comprehensive account of the subject, the author's 

 aim being to emphasise the dynamical aspects of electrical and magnetic 

 phenomena, and to give a rigorous formulation of underlying physical principles 

 as well as translate them into mathematical theory. There is a mathematical 

 introduction giving the essential parts of vector analysis, as required later, the 

 theorems of Green and Stokes being expounded for moving circuits as well as 

 stationary ones, and a section inserted giving a short account of Poincard's 

 discussion of potential integrals as regards convergence and possibility of dif- 

 ferentiation. The main part of the book is divided into two parts on the 

 stereotyped lines of books on Mechanics, Chapters I. — VIII. treating the "Statics 

 and Kinetics" of electrical and magnetic phenomena and Chapters IX. — XV. 

 dealing with " Dynamics." The treatment in the first part follows well-recognised 



