OF RADIANT HEAT THROUGH DIFFERENT BODIES. 57 
prisms of flint or crown glass, water, alcohol, or some other diaphanous 
body. It was exactly the same as if they pretended to be able to analyse 
solar light with a prism formed of coloured glass. 
Of the properties of the calorific rays immediately transmitted by different 
; bodies. 
The radiant heat which has passed through a plate of glass is trans- 
mitted in a greater proportion by a second plate of the same substance 
and the same thickness ; the rays issuing from the second will be trans- 
mitted in a still greater proportion by a third, and so through any number 
of successive screens. The losses sustained by the calorific rays in their 
passage through a succession of screens, as compared with the quantity 
incident on each plate, will therefore form a decreasing series. But the 
difference between every two terms of this series becomes less and less 
as the number of terms increases, so that there must be somewhere a 
limit beyond which the difference has a tendency to vanish. We may 
conclude therefore that the rays after they have passed through a cer- 
tain number of screens, will in their further transmission be subject to 
a loss reducible to a constant quantity as compared with the quantity of 
heat incident to each of the screens through which this further transmis- 
sion is made. 
The same phenomena may be traced in a continuous mass of diather- 
manous matter; that is to say, that if we imagine a piece of glass di- 
vided into several equal layers and measure the loss sustained by the 
radiant heat in its passage through each layer, the greater the distance 
of the layer from the surface at which the heat enters, the less will be 
the diminution suffered by the rays passing through that layer, and the 
losses have a tendency to become constant within a limit depending 
on the thickness of the layers. Some of these results we have already 
verified in the preceding memoir, and it is easy to establish their truth, 
in reference to the sources of heat employed in our present inquiry, by 
means of the numbers which represent the transmissions of the plates 
contained in the first table*. 
* Let us imagine the screen of 8™™ divided into seven layers having for their 
degrees of thickness the differences between two consecutive plates. (See the first 
table in this Memoir.) The quantities of heat incident on the layers when the 
radiation is from a Locatelli lamp are 
100, 77, 54, 46, 41, 37, 35, 33°5, 
and the quantities lost in the successive transmissions are 
: 23, 23, 8, 5, 4, 2, 1-5. 
Now the mean losses for the hundredth part of a millimetre of each screen will be 
23 23 8 5 4 2 1:5 
7’ 43’ 50’ 100’ 100’ 100’ 100 
or  3:286, 0-535, 0-160, 0-050, 0-020, 0-010, 0-007." 
Hence the losses sustained by the rays of the lamp in the first hundredth part 
