VARIABLE STARS, 111 



spectrum, nor the positions of the lines in the solar clnomosphere, nor 

 the positions of the lines in the spectra of any of the bright-lines stars; 

 they do occupy the positions of the lines in the nebuljc, and the spec- 

 trum resembles nebultv spectra as closely as well known nebulae 

 spectra resemble each other." This is said in answer to some high 

 authorities who find it difficult to admit so surprising a fact as such a 

 transformation. It certainly looks as if we had to do with a reversion 

 of the ordinary couise of stellar development; instead of the forma- 

 tion of a star by the slow condensation of a nebula, we seem to have 

 the sudden formation of a nebula by the violent explosion of a star. 

 But sjjeculation on so small a basis of fact is hardly sound. 



I must not close without a reference to the important, though unpre- 

 tending, catalogue of variable stars just published by Mr. Chandler, 

 of Cambridge, who, smce the death of Schonfeld, may be regarded as 

 at the head of the department of variable stars, so far at least as relates 

 to the mathematical aspects of the subject. The first step to any real 

 mastery of it must be an accurate collection of facts, such as this cata- 

 logue embodies. It gives a list of all the stars which are certainly 

 known at present to be variable, 200 in number (02 of them naked-eye 

 stars), and adds a subsidiary list of 100 more which are more or less 

 strongly suspected. Of the 200, about 35 are known to be irregular 

 and unpredictable in their variations, and about as many more are 

 still in doubt as regards the periodicity of their changes. The remain- 

 ing 190 are clearly periodic, and Mr. Chandler gives for each of them 

 the formula by means of which its variations can be predicted, thus 

 embodying in a few modest figures the result of an enormous amount 

 of skillful labor. Ten of the variables belong to the Algol class, about 

 25 to that of Beta Lynv, and the rest resemble Omieron Ccti more or 

 less perfectly. It shouhl be added also that the list of variables is 

 constantly growing — chiefiy, of course, among the telescopic stars. 



