1879.] of a solution of mercuric iodide. . 259. 



iodide as a convenient' liquid for testing specific gravities, and men- 

 tions that he has obtained it with a specific gravity upwards of three 

 times that of water. Being in search of a substance which would 

 give a considerable dispersion of the red end of the spectrum I 

 thought I would try the dispersive power of this solution ; and 

 though its. power of dispersing the red is not nearly so great as its 

 power of dispersing the green, its dispersive power is still very 

 remarkable throughout the range of colours which it transmits. 

 My assistant, Mr Robinson, has prepared the solution for me, but 

 though he has obtained it of sp. gr. 3 "01 3 at a temperature of ■ 

 17" C. since the observations described below were made, the 

 liquid of which I have examined the dispersion is of sp. gr. " 

 2 77 at 18" C. It is of a pale- yellow colour, and seems to 

 absorb completely the violet and indigo light. On observing 

 the solar spectrum through it there is no perceptible diminution 

 of light until nearly half way between F and G ; the absorption 

 begins there and thence increases very rapidly. At the red end 

 A could be seen, so there is not much absorption of visible rays of 

 low refrangibility. 



Filling a hollow prism of 59" 11' with the liquid, I obtained the 

 following indices of refraction : 



A B CD E F 



1-628 1-637 1-641 1-654 1673 1693 



Taking E as the standard ray, I get for the dispersive power 



(C2 — ^) between 



A—B B—D D—E E—F 



•01263, -02555 '02822 -02896 



Comparing these figures with those given in Watts' Dic- 

 tionary for dense flint-glass and for carbon disulphide, we have 

 refractive indices 



The dispersive power of the mercury solution is therefore very 

 nearly three times that of flint-glass for green light and more 

 than double for orange light : and it is more than one-half greater 

 than that of carbon disulphide for green light, and nearly one-third 



