1864.] 



President's Address. 



501 



Anniversary I noticed briefly that he and Dr. Miller were examining 

 the spectra of the fixed stars ; their results have appeared in our Trans- 

 actions for this year, and can scarcely be estimated too highly. Be- 

 fore them Fraunhofer, and recently Donati, Secchi and some others had 

 examined a few stellar spectra with more or less success ; but with in- 

 ferior apparatus, and far less extensively. Theii' superiority is mainly 

 due to their referring the spectral lines to no mere instrumental scale 

 (which, fi-'om the inevitable variations of the spectroscope-prisms, must 

 be liable to error), but to standard spectra of known elements which are 

 formed in juxtaposition with that to be examined, and to which its lines 

 can be compared with extreme precision. As the stars are but points 

 in a good telescope, it is necessary to expand these images into lines 

 that the details of their spectra may be perceptible. This weakens their 

 light, already feeble ; and therefore the telescope to which the spectro- 

 scope is applied must be of considerable power ; in this instance it has 

 8 inches aperture. They have examined more than fifty stars, and find 

 all to have a constitution analogous to that of our own sun, and like 

 it to show the presence of several terrestrial elements. 



The lines due to sodium seem universal ; so also those of magnesium 

 and iron. Hydi'ogen is also frequent ; in two stars it is wanting. The 

 absence of nitrogen from all is very singular, unless it be present under 

 conditions which alter its relation to light. It is to be regretted that 

 the spectra were not compared with nickel, chromium, and the other 

 elements that have been found in aerolites ; this, however, we hope 

 will be done for those elements, and indeed for all the other terrestrial 

 ones. The explanation given of the diflerent colours of the stars is 

 doubtless the true one ; it arises from the existence of different bands 

 of absorption in particular parts of the spectrum. Thus in the beau- 

 tiful star /3 Cygni, these bands are crowded in the blue end of the 

 spectrum of the orange star, and in the red end of that of the blue 

 star. Possibly further investigation of elements may enable us to 

 conjecture why in dichromic double stars the small one is always blue 

 or green. In the planets a closer conformity to the spectrum of the 

 sun, by whose light they shine, might be expected — and is found. In 

 the moon there are no lines whatever due to an atmosphere, nor in 

 Venus. In Jupiter and Saturn, besides the solar lines, there are some 

 identical with those produced by our own atmosphere, and one which 

 must be due to the presence of some peculiar gas or vapour. Mars is 

 still more peculiar; but the appearances give no countenance to the 

 notion that his red colour is due to the chemical composition of his soil. 

 IVIr. Huggins has gone yet further and attacked the nebulae, though 

 their extreme faintness might seem to elude all the power of the spectro- 

 scope. The result is at once a strong proof of the soundness of Bacon's 

 precept in doubtful cases, " Fiat eccperimentwn,'^ and a demolition of 

 more than one accepted hypothesis. The notion of the elder Herschel, 



