REVIEWS 567 



Stellar Motions — with special reference to motions determined by the Spectro- 

 graph. By William Wallace Campbell, Sc.D., LL.D., Director of 

 the Lick Observatory, Univ. of California. [328 pp., 8vo, 14 figures and 

 34 tables in text.] (Henry Frowde, University Press, Oxford, 1913. 

 Price 17s.) 



Science, in all its branches, has during the last century advanced with enormous 

 rapidity. What to one generation seems impossible may, to the succeeding 

 generation, be merely a commonplace. Thus it was but about seventy-five years 

 ago that Auguste Comte wrote that "we shall never be able to study the chemical 

 composition of the celestial bodies ; . . . our positive knowledge with regard to 

 them will necessarily be limited to their geometrical and mechanical phenomena. 

 It will be impossible, by any means, to include investigations of their physical, 

 chemical (and other) properties." So much for the dogmatism of the philosopher, 

 for within twenty-five years from the time of writing the fundamental principles 

 of spectroscopy were formulated by Kirchoff. Sixty-five years ago the possibility 

 of determining the radial motions of the stars was undreamt of; but with the 

 enunciation by Doppler in 1842 of the effect of the motion of the source upon 

 the wave-length of the emitted disturbance, and with its application to optical 

 problems by Fizeau in 1848, the way was paved for the solution of the problem. 

 Little progress was made, however, for some time. Visual methods of determi- 

 nation are difficult, even in the hands of skilled observers, and liable to errors 

 which may greatly exceed the velocity to be found : thus the star "a Cassiopeia?" 

 which, according to the best modern determinations, has a radial velocity towards 

 the sun of 3'a. ± C15 km. per sec. was found by visual observations at Greenwich 

 in 1885 and 1887 to have velocities of 90 km. and 58 km. per sec. respectively, 

 away from the sun. Twenty-five years ago the radial motion of not a single star 

 was known, even approximately. The rapid development of this branch of 

 astronomy during recent years has resulted from the introduction of photographic 

 methods. Even with this great advance, the displacements of the spectral lines 

 to be measured are so small and systematic, and other errors may so easily enter, 

 that great precautions have to be taken if an accurate result is to be obtained : 

 in 1890 the radial velocity of the star cited above was determined photographically 

 at Potsdam as 15*2 km. per sec. towards the sun, almost four times as large as 

 the correct value. It is to the improvements in the spectrograph and in the 

 methods of measurement introduced by Dr. Campbell, that the present degree 

 of accuracy is largely due, and it is to the untiring labours of him and his 

 colleagues at the Lick Observatory, and at its southern dependance at Cerro 

 San Cristobal, Santiago, Chile, that we owe almost the whole of our knowledge 

 of the radial velocities of stars ; it is, therefore, very fitting that an account of the 

 theory and methods used, and discussions of some of the problems which have 

 arisen in connection with the results obtained, should be for the first time collected 

 together into one volume by Dr. Campbell himself. 



The eight chapters of this volume formed the Silliman lectures delivered by 

 the author in Yale University in 1910. To bring the volume up to date, a series 

 of footnotes have been added incorporating the chief results obtained since their 

 delivery. An important feature of the book is the valuable series of tables, 

 thirty-four in number, which illustrate the text, and contain a mine of information. 

 Each chapter of the book is essentially complete in itself, and more or less 

 independent of the others. The first two are introductory, and include a brief 

 account of the development of the subject, and a description of the D. O. Mills 



37 



