REVIEWS 541 



ASTRONOMY 



Stellar Movements and the Structure of the Universe. By Prof. A. S. 

 Eddington, M.A, F.R.S. [Pp. xii + 266.] (London : Macmillan & Co. 

 Price 6j.\ net.) 



In this book, which forms the eighth in Macmillan's excellent series of Science 

 Monographs, the author gives an account of the many recent advances in 

 our knowledge of the sidereal universe. There are many causes which have 

 contributed to the rapid advance within the past few years of our knowledge of 

 stellar statistics, upon which any discussion of the structure of the stellar universe 

 must necessarily be based. A most important factor has been the application of 

 photography to astronomy ; the development of and great advance in photometric 

 methods has led to the accurate determination of the photographic magnitudes of 

 many stars, giving valuable data upon which to base star-counts ; it has provided 

 a method by which the proper motions of stars may be accurately derived from 

 photographs taken with the comparatively short interval of ten or twelve years ; 

 it has enabled stellar parallaxes to be determined with much more precision 

 and with far less trouble than by visual measurements, and, by so doing, has 

 greatly increased our knowledge of the distances of individual stars ; it has 

 enabled stars to be easily classified according to the type of their spectra ; and, 

 with the aid of the spectrograph, has been the means of bringing about a very 

 rapid increase in our knowledge of the radial velocities of stars — not a single 

 radial velocity being known with accuracy twenty-five years ago. Also there has 

 been a great increase of our knowledge of the proper motions of stars, due merely 

 to lapse of time, and these furnish the best data for the determination of the 

 average distance of groups of stars. 



After a brief account of the data upon which any theory of the structure of the 

 stellar universe must be based — a knowledge of the position in space, motion, 

 luminosity, and spectral type of each star — the author passes on to discuss in some 

 detail the nearest stars, of which our knowledge is unusually full, their parallaxes 

 and in many cases their radial motions being known ; this is followed by a chapter 

 on moving clusters, the discovery of which has given a means of calculating, 

 within small limits of error, the parallaxes of a large number of stars. The 

 motions of the stars in general are next considered, and it is explained how, by 

 means of a statistical discussion of the proper motions and of the radial velocities 

 of a large number of stars distributed over the sky, the direction and magnitude 

 of the solar velocity may be calculated. The investigation of the distribution of 

 the proper motions showed that the stellar motions are not distributed at random, 

 but that there exists a systematic irregularity, capable of explanation by supposing 

 — as announced at the British Association meeting in 1904 by Kapteyn— that the 

 stars tend to move in two favoured directions. A clear account is given of the 

 methods by which the directions of the two star-streams relative to the sun, their 

 relative velocities and distances, and the distribution of the stars between them have 

 been determined. Schwarzschild's alternative explanation of the phenomenon — 

 that there is one direction in which the stars have greater mobility than in 

 perpendicular directions— is compared with the two-stream theory, and the 

 mathematical theory of each is briefly and lucidly given. 



The next chapter contains an account of the phenomena associated with the 

 spectral type. Although the average proper motions of the stars, when grouped 

 according to type, in the order which is supposed to be that of stellar evolution, 

 pass through a maximum with type F stars, and then decrease, it was announced 



