Lord Rayleigh's Investigations in Optics. 53 



involving, as it would, larger telescopes, and, unless a loss of 

 brightness were accepted, higher prisms. 



Another point requiring attention in the discussion of the 

 question of large versus small prisms is the relative difficulty 

 of securing the necessary accuracy of surfaces in the two 

 cases ; but it can hardly be treated satisfactorily a priori. It 

 would, however, seem that small prisms have the advantage 

 in this respect also. 



So far I have supposed the prisms to be simple. Of late 

 years compound prisms of flint and crown have come largely 

 into fashion ; and quite recently M. Thollon has revived the 

 use of compound prisms of bisulphide of carbon. Mr. Christie 

 considers that compound prisms are, on the whole, to be pre- 

 ferred ; but I cannot help thinking that, except where direct 

 vision is important, their advantages have been overrated. 

 The difference of indices in the case of crown glass for the 

 rays D and B is "0035, which is quite sufficiently high to come 

 into comparison with the corresponding numbers for flint and 

 bisulphide of carbon. In a direct-vision prism the thickness 

 of the crown is greater than that of the flint ; so that the loss 

 of efficiency is even more serious than a mere comparison of 

 the dispersions would suggest. See (6) § 4. 



The principle of the compound prism is carried to its limit 

 by employing media of equal refracting-power for the part of 

 the spectrum under examination. For this purpose I chose 

 bisulphide of carbon and flint glass. With Chance's " dense 

 flint" the refractions are the same, and the difference of dis- 

 persions relative to D and B is *0056, i. e. about as great as 

 the difference for "double-extra-dense flint" and crown. A 

 dozen glass prisms of 90° were cemented in a straight line on 

 a strip of glass and immersed in bisulphide of carbon. The 

 liquid is contained in a tube closed at its ends with glass plates. 

 The " thickness " of each prism is 1 inch and the height about 

 | inch, so that the total thickness is 12 inches. The character 

 of the glass is such that at ordinary temperatures the red rays 

 pass without deviation. To observe other parts of the spec- 

 trum with advantage, it is necessary to mix a little ether with 

 the bisulphide of carbon in order to lower its power. A 

 similar result could be obtained by slightly warming the tube; 

 but this course is not to be recommended. 



The instrument, which was made by Hilger, gave excellent 

 results at times, but showed the usual caprice of bisulphide of 

 carbon. With the refracting edges vertical, the definition 

 was usually good for twenty or thirty seconds after shaking 

 up, but would often rapidly deteriorate afterwards. Although 

 care was taken not to touch the tube with the hands, this 



