448 ]\Ir. W. Swan 07i a new Method of observing 



later, of which the words " Mr. Joule's conjecture " have called 

 forth his reclamation : — 



" A more convenient assumption has since been pointed to 

 by Mr. Joule's conjecture, that Carnot's function is equal to the 

 mechanical equivalent of the thermal unit divided by the tempe- 

 rature by the air thermometer from its zero of expansion ; an 

 assumption which experiments on the thermal effects of air 

 escaping through a porous plug, undertaken by him in conjunc- 

 tion with myself for the purpose of testing it (Phil. Mag. October 

 1852), have shown to be not rigorously, but very approximativcly 

 true." 



I remain, Gentlemen, 



Yours very faithfully, 



William Thomson. 



LVII. On a new Method of observing the Spectra of Stars. 

 % William Swan, F.R.S.E* 



ABOUT the beginning of March last, being engaged in ex- 

 periments which involved observations of the dark lines 

 in the solar spectrum, I attempted to observe the spectrum of 

 Sirius; but the light of the star, enfeebled by two reflexions 

 at the specula of a heliostat which formed part of my apparatus, 

 and accompanied by a blaze of gas-light from the street lamps, 

 was too faint to be visible. The failure of this experiment was 

 so obviously due to the unfavourable nature of the trial, that I 

 determined to renew it on the first opportunity. 



Although I was fully aware that Fraunhofer had observed 

 dark lines in the spectra of several stars, I did not suppose that 

 he had executed exact measurements, as no reference to such 

 observations is made in his celebrated paper on the dark lines in 

 the solar spectrum, where he describes the appearance of several 

 stellar spectra f. I therefore resolved, as soon as I had leisure, 

 to attempt a series of observations on the spectra of stars ; and 

 the recollection of a method of ascertaining refractive indices, 

 described by me in 1844, immediately suggested an arrangement 

 by which accurate results could be readily obtained J. 



I have, however, more recently found that Fraunhofer actually 

 measured the deviationsof the rays in star-light; and before de- 

 scribing my own, I will shortly explain his mode of observation §. 



* Communicated by the Author, 

 t Schumacher's Astronomische Abhandlungen, 1823. 

 + Edinb. New Phil. Joura., Jan. 1844. 



§ See his paper translated in Edinburgh Journal of Science, vol. viii. 

 p. 7, 1828. 



