130 Notice of some Recent 



suppose that this takes place. Then the molecules, which 

 form the first layer of -0154 thickness, will form a hind of 

 sieve, retaining all that is not completely transmissible by 

 the rock salt ; and the quantity of heat lost in the transit 

 by one or other plate, that is to say, 1 — 0-923, or -077 will 

 be the sum of the rays absorbed or retained, and of the 

 rays reflected, to the two surfaces. When, therefore, 

 radiating heat is received upon one of the plates, the thinnest 

 for example, and when they are transmitted by the other, 

 the supposed absorption, or ejmration, will take place in 

 the first, and no more will arrive at the second than the 

 rays completely transmissible by the substance of which it 

 is composed, with the exception of the quantity lost in the 

 two reflexions ; so that the loss experienced by these rays, 

 in passing through the second plate, must necessarily be 

 less than -077. But experiment shows that on this passage, 

 there is exactly 0*923 of heat transmitted, and -077 of heat 

 lost. Hence, no absorption has in reality taken place in 

 the first transmission, and the quantity -077 expresses only 

 the loss produced by the reflexion of the radiating heat to 

 the first and second surfaces of each plate. We can readily 

 determine the special value of either of two reflexions. If 

 Ave term R the reflexion for unity of incident heat, then 

 1 — R will be the quantity which will penetrate into the 

 interior of the plate, and R (1 — R) the reflexion which the 

 latter will undergo on the posterior surface ; for, as the salt 

 possesses no absorbing power, the whole quantity 1 — R 

 arrives at the second surface, and is reflected in the pro- 

 portion of R : 1. Now, the sum of the two reflexions, 

 added to the quantity transmitted, 0-923, ought to produce 

 the quantity of incident heat, which we suppose equal to 

 unity. 



We have then the equation, 



R + R(l-R) + 0-923 = 1. 

 from which we deduce, 



R = 1 +_ VO 7 ^ = 1 + 0-9607. 



The first sign of the radicle, as it leads to an absurd 

 result, should be rejected. The reflexion at the anterior 

 surface of the plate will then be 1—0-9607=0-0393; and 

 such will also be the proportion of the second reflexion, 

 relatively to the quantity of heat, which arrives at the 



