CELESTIAL CHEMISTRY. 265 



part which is covered up. If we open the fan still more, the conceal- 

 ed parts will come into view. This corresponds with the additional 

 dispersion that we must give to a spectrum that has been imperfectly 

 spread out, in order to sec the entire pattern it can show. 



To spread the light out wider, a multiplicity of prisms may be em- 

 ployed. A spectroscope constructed for Mr. Gassiot of London, 

 the finest instrument yet finished of its kind, has nine prisms, about 

 2-2 inches long, and two high, and was originally supplied with tele- 

 scopes of two inches aperture, an:l two feet focal length. Two fresh 

 telescopes have recently been made for this instrument, of 2^ inches 

 aperture, and three feet focal length. These show many additional 

 lines. With this instrument a very careful survey and mapping of the 

 spectrum is being carried on, and the results will, no doubt, be the 

 most precise and complete information concerning the number and 

 position of the dark lines that has yet been obtained. The size of the 

 Gassiot spectroscope enables it to give a bright image with a small 

 allowance of light; but certain inquiries in which Mr. Huggins is en- 



fagcd need a further extension of power in this direction, and Mr. 

 rowning is constructing for him a monster instrument, with about 

 half the number of prisms on Mr. Gassiofs instrument; but these 

 prisms fully double the size of Mr. Gassiot : s, and furnished with tele- 

 scopes of four feet focal length, and proportionally large aperture. 

 With this splendid and costly apparatus, which has quite novel and 

 special means for increasing the dispersion when deemed necessary, 

 it will be possible to obtain great separation with a minimum loss of 

 light, so that the spectra of very feebly luminous bodies may be made 

 out. 



In ttyese two great spectroscopes the extra separation is produced 

 by the multiplication of glass prisms, and when these are of tine qual- 

 ity and exquisitely wrought, the definition is the sharpest that can be 

 obtained. 



Different substances vary in their refractive and dispersive powers. 

 Sir J. Herschel observes, " in general, high refractive is accom- 

 panied by high dispersive power; but exceptions are endless, espe- 

 cially among the precious stones, of which the diamond affords a strik- 

 ing instance. 11 



If the two powers had always gone together, we could have had no 

 achromatic lenses ; but happily by selecting different kinds of glass, 

 and setting them to opposite work, we can correct all (or nearly all) 

 the error arising from dispersion, and only neutralize a portion of the 

 refraction. AVhen greater dispersive power is required in a prism 

 than ordinary glass exerts, another kind of glass may be chosen, or 

 recourse may be had to a liquid like bisulphide of carbon, which gives 

 a very wide spectrum. A liquid has the disadvantage that it cannot 

 be maintained in the prismatic form except by putting it in a vessel 

 of the required shape. Hollow prisms are accordingly made of thin 

 glass, and filled with the fluid required. Mr. Browning has made a 

 spectroscope of this kind for Mr. Gassiot. It consists of eleven 

 prisms, with telescopes 2^ inches aperture and three feet focal length, 

 and it is able to separate the principal soda lines to the extent of 3' 

 &", the nine glass prisms only separating them 1 ; . The definition is 

 excellent while the fluid is of uniform temperature and density ; but 



23 



