394 Celestial Chemistry. 



of the constitution of those distant bodies, the stars, and the 

 light which is thus thrown upon their structure is conclusive 

 as to their being of the same nature as our own sun ; a result 

 which analogy had previously indicated, but which had 

 not been supported by any positive evidence. It might be 

 supposed that their distance offered insuperable obstacles to 

 such an inquiry, but spectrum analysis knows no such limits, 

 and as long as we can obtain light of an incandescent sub- 

 stance from a suitable source, it matters not whether it exists 

 within a few inches of the spectroscope, or at a distance 

 of unnumbered millions of miles, the result being equally 

 certain. 



The difficulties of these observations from other causes are, 

 however, very great, arising principally from the extremely 

 few occasions when telescopic observations with a large instru- 

 ment can be carried on in this country, on account of tho 

 frequent atmospheric changes. The amount of work done in 

 the past two years, although at first it appears small in quan- 

 tity, is in reality quite remarkable, and bears testimony in a 

 high degree to the devotion and patience of the observers, 

 as, in consequence of the few fine nights available when a star 

 is in a good position, the mapping of a single star completely 

 would occupy several years. The diagrams in Mr. Huggins^s 

 article of June, 1863, show the result of the cursory examina- 

 tion of three first-class stars, but since that time the apparatus 

 has been improved, and a large number of fine lines have been 

 observed in addition to those previously seen. In the spectra 

 of all the brighter stars that have been examined, the dark 

 lines appear to be as numerous and as fine as in the solar 

 spectrum. The great breadth of the lines shown in the former 

 diagram of Sirius, and which band-like appearance was so 

 marked as to specially distinguish it from the other stars of 

 the diagram, has to a very great extent disappeared, and 

 though these lines are still strong, they now appear, as com- 

 pared with the strongest of the solar lines, by no means so 

 abnormally broad as to require this star, and some similar ones, 

 to be placed in a class apart. No stars sufficiently bright to 

 be observed are without lines, and therefore star differs from 

 star only in the arrangement of the lines, and consequently in 

 the elementary substances present ; but all the stars are con- 

 structed on one and the same plan. The number of fixed stars 

 observed amounts to nearly fifty, but the observations of three 

 or four occupied most attention, and two only have been 

 mapped with any degree of completeness. The coloured plate 

 accompanying this ai-ticlo gives a representation of these 

 two spectra, namely, of Aldebaran and of Betelgeux (a Ononis), 

 with a portion of the solar spectrum for a scale of comparison. 



