398 



NA TURE 



[February 21, 1895 



its application has been able, wiih the aid of the large (unds 

 that he has at his disposal, and the magnificent help which he 

 has accumulaled round him, to obtain practically the spectra of 

 all the stars down to the fidh or sixth magni'uHe in both hemi- 

 spheres. In a few years' time we shall be able to work on the 

 spectra of all the stars in both hemispheres, just as well as we 

 can at present deal » ith their magnitudes and positions by the 

 star charts. 



An instrument of this kind (Figs. 8 and 9), having an aperture 

 of only 6 inches, has been in use at Kensington for some lime, 

 and some of the results which have been obtained by its aid are 

 shown in Figs. 3, 10, and II. They are absolutely untouched 

 photographs. 



Without going into minute differences, we can, and, if 

 we are wise, we shall deal first of all with the larger differences 

 presented bythe various classes of stars which people space. Here 

 we have photographs of stars of different classes (Fig. 3). Vou 

 will understand from these photographs how perfectly justified 

 Rutherfurd and otheis have been in attempting to classify the 



Fic. 9.— Details of objective prism. 



Stan by means of their spectra. In Sirius we get one class of 

 stars, distinguished by the development of certain lines, which 

 are due to the absorption of hydrogen. In a Cygni the hydrogen 

 gas is represented quite distinctly, but the absorption there with 

 regard to certain lines is much more developed than in such 

 a star as Siriiu. In Arcturus the absorption of the hydrogen is 

 .almost hidden in an enormous mass of lines. Here again we 

 have another class, and it is not too early to remark that 

 Arcturus in its spectrum exactly resembles our own sun. Thus 

 we can «ay the sun is like Arcturus, not like a I lerculis, a Cygni, 

 and so on. -.-•^ 



Fig. 10 is an exemplification of the kind of result which is now 

 being obtained, and ihe kind of work which can now lie carried on 

 with regard totheminutcstruclureof thcscspec'ra. Onci.sthe star 

 Arcturus, and the other a star in the cnnsicllationol ihe Swan. 

 Vou 5ee at once that, if it is a quesliun of attempting to deter- 

 mine whether stars are like each other, or whether they are un- 

 like each other, and Ihe points in which ihcy differ, with the 

 resources of modern science at Ihe present moment — small 



NO. I32I, VOL. 51I 





gar.' 





-\ 



