160 [June 19, 



very considerable modification. Notwithstanding the great length 

 of the chemical spectra obtained by the aid of the bisulphide, not 

 more than one-sixth or one-seventh of the true extent of the spectrum 

 produced by the electric spark between various metals is procured, 

 as may be shown by comparing the spectrum with one of the same 

 metal furnished by the use of a lens and prism of rock-crystal. 



Rock-crystal, however, possesses but a comparatively small refrac- 

 tive and dispersive power, whilst it almost always affords some trace 

 of double refraction in one portion or other of the spectrum procured 

 by its means. 



In searching for some singly refracting medium which should 

 possess sufficient refractive and dispersive power to enable it to be 

 used advantageously in the construction of lenses and prisms suitable 

 for this inquiry, the author was led to examine the photographic 

 absorption of a variety of colourless substances which appeared per- 

 fectly transparent to the luminous rays. The experiments detailed 

 in the first portion of the present paper refer to this absorbent 

 action of various media upon the chemical rays of the spectrum ; 

 whilst the second portion of the paper is devoted to a description of 

 the electric spectra of some of the more important elementary bodies, 

 and the effect of varying the gaseous media in which the sparks 

 producing these spectra are made to originate. 



1. The Photographic Transparency of Bodies. In the experi- 

 ments upon the absorbent action of the different media, the source 

 of light employed was the electric spark obtained between two 

 metallic wires (generally of fine silver), connected with the terminals 

 of the secondary wires of a ten-inch induction-coil. The light, after 

 passing through a narrow vertical slit, either before or after traversing 

 a stratum of the material the chemical transparency or diactinic 

 quality of which was to be tested, was allowed to fall upon a quartz 

 prism placed at the angle of minimum deviation for the mean of the 

 refracted rays. Immediately behind this was a lens of rock-crystal, 

 and behind this, at a suitable distance, the spectrum was received 

 upon a collodion-film coated with iodide of silver; this was sup- 

 ported in the frame of a camera, and after an exposure, generally 

 lasting for five minutes, the image was developed by means of pyro- 

 gallic acid, and fixed with cyanide of potassium. 



The general results of these experiments were as follows : 



