136 SCIENCE PROGRESS 



optics ; thus it includes a dynamical aspect of that conception added on 

 to and of course consistent with the Huygenian or rather Young-Fresnelian 

 one. The electric and optical ray is implicitly in Maxwell's equations and 

 is only a corollary to them. But in any other kind of elastic transmission, 

 e.g. waves in an elastic-solid medium, a corresponding theory can be worked 

 out. I take it this idea is Poynting's main contribution, and it clarified 

 many things, especially electrical." Poynting appears not to have noticed 

 that the energy vector is indeterminate theoretically, yet although other 

 forms have been proposed, the balance of evidence seems to be in favour 

 of Poynting's vector, which is generally accepted as correct and has assumed 

 fundamental importance in electrical and optical theory. 



The third important series of papers by Poynting were concerned with 

 radiation, the pressure of light and related subjects. These papers are 

 collected together in Part III. Here again, Poynting's delicate manipulative 

 skill, which had been so well shown in his work on the mean density of the 

 earth, proved invaluable. In conjunction with Dr. Guy Barlow he suc- 

 ceeded in establishing the existence of the tangential force produced when 

 light is reflected from a surface on which there is some absorption. They 

 also proved the existence of a torque when light passes through a prism, 

 and further demonstrated the existence of the recoil from light of a surface 

 giving out radiation. Other papers were concerned with certain theoretical 

 aspects of radiation. The most important of these is a Phil. Trans, paper 

 of 1903 on " Radiation in the Solar System : its Effect on Temperature and 

 its Pressure on Small Bodies." In the first part of this paper the tem- 

 peratures of the planets are determined, assuming the fourth-power law of 

 radiation ; as regards Mars, Poynting arrived at the conclusion that the tem- 

 perature of Mars was so low that life as we know it was not possible on its 

 surface. This conclusion was contested by the late Prof. Lowell ; Poynting, 

 however, replied to the criticisms (as the present writer considers, success- 

 fully) in a later paper, in which he confirmed his previous results. In the 

 second part of the same paper he considers the effect produced by radiation 

 on the orbits of small bodies round the sun and shows that they will be 

 retarded in their motion by the reaction of their exchanges of radiation so 

 that they will ultimately fall into the sun. He also shows that the effect 

 of radiation on the particles of Saturn's rings might make them — if of suit- 

 able size — repel instead of attract one another. 



The succeeding parts of the volume, dealing with Light, Miscellaneous 

 Papers, Statistics, and Addresses and General Articles, are of lesser value 

 from the permanent scientific view-point, although several of the addresses 

 and general articles contain well-written and interesting accounts not only 

 of those branches of physics in which Poynting had particularly specialised, 

 but also of several related subjects. That there is a certain amount of repe- 

 tition and overlapping in these is, of course, inevitable. The editors gener- 

 ally decided, however, when in doubt as to whether or not to include a 

 paper, that it was wiser to include it. 



Poynting's writings were always clear, simple and lucid. Their publica- 

 tion in book form will be found of great value by students of the science of 

 physics, to which Poynting contributed so much that is of permanent value. 



H. S. J. 



MATHEMATICS 



Mathematics for Engineers : Part II. By W. N. Rose, B.Sc. (The Directly- 

 Useful Technical Series.) [Pp. xiv-f 419, with 142 figures.] (Lon- 

 don : Chapman & Hall, 1920. Price 13s. 6d. net.) 



The problem before the author of a text-book on the Calculus for engineers 

 is to reconcile the two necessities of a logical treatment of the subject and 

 the elimination of such purely theoretical matter as is unnecessary for the 



