Intelligence and Miscellaneous Articles. 387 



certain preparations of sulphide of calcium, I found some much 

 more sensitive than others for infra-red rays. These substances 

 have permitted me to determine directly the wave-lengths for 

 the more brilliant lines of several incandescent metallic vapours 

 (potassium, sodium, cadmium), by using the diffraction-spectra pro- 

 duced by a very beautiful metal grating of Mr. Rutherfurd's, which 

 was kindly lent me by M. Mascart. 



Eor other metals, the spectra were obtained by means of a bi- 

 sulphide-of-carbon prism. The positions of the lines were referred 

 to those of the potassium and sodium lines, and to the positions 

 of the bands and characteristic lines in the infra-red region of 

 the solar spectrum, as well as the absorption-bands of samarium ; 

 the wave-length for each line was then determined by interpolation 

 with great accuracy. 



The wave-lengths for the lines and principal bands in the infra- 

 red of the solar spectrum have been again determined by means of 

 the grating, which had been placed at my service ; and the more 

 sensitive substances in my possession have permitted me to extend 

 these measurements much further than I was enabled to do in my 

 previous researches *. I propose to return shortly to this work, 

 which has led me to rectify several numbers relating to wave- 

 lengths at the less-refrangible end of the spectrum ; I will only 

 give here the wave-lengths of the principal lines used for reference : 

 the band indicated by A" in my former memoir is made up of 

 two bands, whose wave-lengths are from 0*00115 to 0*00119 

 millim., and from 0*001132 to 0*001142 millim.; the large band 

 A'" extends from 0*001351 to about 0*001400 millim., and the 

 extreme band A iT from 0*001800 to about 0*001900 millim. The 

 wave-lengths of the characteristic absorption-bands of samarium, 

 directlv determined, are : from 0*001079 to 0*001097 millim. for 

 the first, and from 0*001239 to 0*001267 for the second. 



The metals were volatilized in the voltaic arc. The intensity 

 was such, that I could make use of a very narrow slit, and 

 observe, on the phosphorescent substance, interesting details. In 

 this case, it is necessary that the image of the line be brought to 

 an exact focus on the phosphorescent screen ; for, unless this be 

 so, the results are not observable when the slit is narrow. I have 

 not as yet been able to invent a suitable arrangement for giving 

 a complete and definite image of the whole of the infra-red spectrum; 

 but, by repeated trials, it.was possible to obtain sufficiently distinct 

 images of the groups of lines which are to be examined. I am 

 occupied, however, at the present moment with an improvement, 

 which will admit of still greater delicacy. 



The following table contains the wave-lengths for the most 

 intense lines, bands, or groups of lines w 7 hich characterize the 

 spectra of several metallic vapours. The numbers are expressed in 

 millionths of a millimeti-e, and are generally exact to one or two 

 millionths of a millimetre. 



* Annales de Chimie et de Physique, 5 e serie, t. xxx. p. 5. 



