218 



Mr. W. N. Hartley. 



was photographed more strongly than the rest, the slit being wider, 

 so that any feeble lines not easily seen nnder the usual conditions 

 might be made apparent. Two spectra were photographed, bat one 

 was rather too weak to justify conclusions drawn from its appearance. 



Presence of the Water-vapour Spectrum. 



The second spectrum was sufficiently dense, and its examination 

 disclosed the curious fact that it contained two groups of the water- 

 vapour lines (Liveing and Dewar, ' Phil. Trans.,' vol. 179, p. 27). 

 The strongest group in the carbon monoxide spectrum belonged to 

 Section VI, wave-lengths 3428, 2472, 3548, and 3933, the last line 

 corresponds in position with the solar line K. It is stated that the 

 lines in this section are weaker than those in Section V of the oxy- 

 hydrogen flame, which lie between 3063 and 3203*5. Now, although 

 these latter lines were photographed in the carbon monoxide they 

 were of much feebler intensity than those in Section VI. 



The lines "were identified by comparing them with those from the 

 flame of the oxyhydrogen blow-pipe photographed with the same 

 instrument, with which they were found to coincide. That both are 

 obtained by the same process of combustion, but by gases from 

 different sources and with groups of lines with different relative 

 intensities, was thus established. 



There were no other lines visible, only a continuous band of rays. 



Possible Origin of the Water-vapour Lines in the Carbon Monoxide 



Spectrum. 



As to the possible origin of these two groups of lines, it was at 

 once conceivable that the water- vapour present in the carbon mon- 

 oxide was the cause of their appearance. 



Another spectrum was then photographed, using on this occasion 

 a four-prism instrument, but there is, unfortunately, no means of 

 obtaining on one plate both groups of lines including those more 

 refrangible than 3502. 



The gas was dried by passing it through pumice soaked in oil of 

 vitriol. A continuous spectrum was visible, with very slight indica- 

 tions of some of the less refrangible of the water-vapour lines. A 

 further trial was made of the dried gas, which in this case was burnt 

 from the interior or oxygen jet of the Deville blow-pipe, the oxygen 

 passing through the platinum nozzle of the outer tube. A steady 

 brilliant flame burnt for fifty minutes. In this instance the two 

 groups of the water- vapour lines were as distinct as before, with a 

 continuous band of diffused rays, but no other lines or bands were 

 seen. It may be remarked that damp weather prevailed during these 



