594 



SCIENCE. 



[N. S. Vol. XVIII. No. 462. 



in a quite regular way, iiid that among the 

 objects of one and the same system the ele- 

 ments of the light changes, especially the 

 length of period, amplitude of variation and 

 form of light-curve, show certain common 

 characteristics, impels one to the acceptance 

 of a common, or at least similar, cause for the 

 changes of light. To be sure, up to the pres- 

 ent time, no entirely satisfactory explanation 

 has been found; but precisely this enigmatical 

 character of the phenomenon, which, without 

 doubt, stands in a certain relation to the 

 stage of development of the cluster concerned, 

 increases the interest in itself, and incites to 

 further research on this new subject. 



The credit for the discovery of the many 

 variables in the star clusters belongs to Pro- 

 fessor Bailey, who is in charge of the Arequipa 

 Station of the Harvard Astronomical Observa- 

 tory. This credit is much enhanced from the 

 fact that the discoverer has taken upon him- 

 self the enormous task of studying the light 

 changes of this increasing number of vari- 

 ables, and of determining their elements. The 

 first fruit of this undertaking is the present 

 volume, which is especially concerned with 

 the cluster <u Centaiiri. The unwearied in- 

 dustry and great skill which the author, and 

 his colaborer. Miss Leland, have shown in the 

 measurement and study of the rich materials 

 of observation, will call forth special acknowl- 

 edgment and admiration, and one may well 

 congratulate the Harvard Observatory on a 

 publication which commands a prominent 

 place among the valuable works of this insti- 

 tution. 



This work bears eloquent witness also to the 

 high importance which the application of 

 photography has won in the development of 

 astronomy at the present time. In general 

 only by photographic means was the solution 

 of the above problems possible. It is scarcely 

 conceivable that in the densely crowded star 

 clusters any such valuable results could have 

 been obtained by direct estimates of brightness 

 or measures in the telescope. Even the identi- 

 fication of the individual objects would pre- 

 sent the greatest difiiculties, and the precision 

 of the observations would be injured by the 

 troublesome nearness of the other stars, quite 



aside from the great expenditure of time 

 which the execution of the work would 

 require. On the other hand, a single photo- 

 graphic plate renders possible the determina- 

 tion of the brightness of all the variables in 

 the cluster. By means of reseaux, by the use 

 of enlargements, and the selection of special 

 portions of the photographs, the discovery of 

 the various variables is notably simplified. 

 The danger of confusion can thus be wholly 

 eliminated, and the first comparison of bright- 

 ness can be checked by another determination, 

 and improved at will by the independent esti- 

 mates of another observer. 



In the discussion of this comprehensive 

 volume one naturally can not enter into all the 

 details; and in what follows only the most 

 important points in regard to the reduction 

 of the observations and the most weighty 

 conclusions are discussed somewhat carefully. 

 In the first chapter the author gives, first 

 of all, a brief historical sketch in regard to 

 the discovery of variables in the different 

 clusters. The surprising phenomenon was 

 first observed in the two clusters Messier 3 

 and Messier 5. In the first, Pickering, in the 

 year 1889, had discovered a star near the 

 center of the cluster, which he at first re- 

 garded as a nova, but which later proved to 

 be an ordinary variable. In the second, 

 Packer, in 1890, found by direct observations 

 the variability of two stars, which was veri- 

 fied at the Harvard Observatory by means 

 of photographic plates. Also in the same year 

 Common made the discovery of several vari- 

 able stars in this cluster. An examination 

 by Mrs. Fleming and Professor Pickering of 

 photographs made at Arequipa led in August, 

 1893, to the discovery of two variables in tu 

 Centauri, and soon after followed the finding 

 of six variables in 47 Tucanae by Professor 

 Bailey and Mrs. Fleming. In the year 1895, 

 after Pickering and Bailey had succeeded in 

 establishing the variability of 26 stars in u> 

 Centauri, a systematic investigation of the 

 densest star clusters was undertaken at the 

 Arequipa Observatory, and this search brought 

 to light a surprisingly large number of vari- 

 ables. A summary of all the objects discov- 

 ered up to 1898 was given by Pickering in the 



