Determinations of the Absorption, of Heat by Liquids. 211 



ments, which it would be tedious to record, have established this 

 fact. 



Although the comparison between the rock-salt and glass cells 

 was attended with so uniform a result, yet it is open to the ob- 

 jection that the walls of the two kinds of cells were of different 

 thicknesses ; for, from its friability, it is necessary to use rather 

 thick plates of rock-salt. A stricter comparison was therefore 

 instituted. Two plates of glass of exactly the same thickness as 

 the plates of rock-salt, 0*3 of an inch, were cut from a slab of 

 the so-called patent plate — a glass perfectly colourless when 

 looked at edgewise. Two thinner plates, O'l of an inch, of the 

 same glass were also procured to compare with the thick plates. 

 The same moveable cell was employed for both the rock-salt and 

 the glass, so that in both cases the diameter of the cell and the 

 position which determined the depth of the liquid layer were the 

 same. The source of heat was the same as in the former experi- 

 ments, and nothing intervened between it and the cell. The 

 conditions of the experiment being thus exact, I sought at the 

 same time to determine more precisely the difference between 

 the deportment of bisulphide and bichloride of carbon. The 

 results of these experiments are given in the following Table : — 



Source : — bright-red platinum spiral. 

 Layer of liquid — of an inch thick. 



Free radiation through Transmission. 



I. Empty cell, thin glass plates . . . 100*0 

 Filled with bisulphide of carbon . . 103*5 

 Filled with bichloride of carbon . . 104*0 



II. Empty cell, thick glass plates . . . 1000 

 Filled with bisulphide of carbon . . 109*5 

 Filled with bichloride of carbon . . 111*2 



III. Empty cell, rock-salt plates .... 100*0 

 Filled with bisulphide of carbon . . 97*5 

 Filled with bichloride of carbon . . 101*5 



Here, the liquid layer remaining the same, the thick glass 

 plates, it will be seen, instead of reversing the action, actually 

 increase the transmission 6 per cent, over and above that which 

 occurs when thin glass plates are used. The objection alluded 

 to is thus not only removed, but the above result afforded a clue 

 by which a solution of the enigma was finally obtained. The 

 bichloride of carbon, it will also be observed, under all circum- 

 stances surpasses the bisulphide in augmenting the transmission. 



An interesting result occurred in the course of the experiment 

 with the glass cells. In place of the transparent bisulphide 

 and bichloride of carbon, I tried the same liquids saturated with 



P2 



