Determinations of the Absorption of Heat by Liquids. 



209 



by increasing the thickness of a diathermic liquid. With chlo- 

 roform the addition of )( 1 )0 of an inch (J of a millimetre) to the 

 thickness of the liquid film changed the increased transmission 

 of 4J per cent., that it previously gave, to an absorption of 8 per 

 cent. In order to follow this change more closely, and to observe 

 where the reversal takes place, 1 made a series of experiments 

 with bichloride of carbon, gradually augmenting the thickness of 

 the liquid layer until absorption became marked. These results 

 I state at once in a tabular form. For the sake of clearness and 

 brevity I have omitted the deflections, which gradually became 

 less till zero was crossed, when they then increased on the side 

 indicating absorption. Taking the transmission through the 

 empty cell as equal to 100, the figures in the second column 

 show the transmission by the layers of liquid of the thickness 

 given in the first. 



Radiation through Bichloride of Carbon enclosed between 

 Rock-salt plates. 



Source : — platinum spiral at a low white heat. 



Thickness of liquid layer Transmisssion 



in parts of an inch. per 100. 



0-02 112-4 



0-03 107-0 



0-04 105-3 



0-11 100-8 



0-14 100-0 



0-27 94-0 



The singular fact is here brought out, that at a thickness of 

 •14 of an inch the transmission through the bichloride was the 

 same as through the empty cell. Eighteen months after this 

 some of these determinations were repeated, and no appreciable 

 difference from the above Table was discovered. 



In connexion with past and future determinations of the ab- 

 sorption of heat by liquids, I need hardly point out the influence 

 of the fact here established, and the conclusions to which I shall 

 allude in the sequel. 



The foregoing experiments were made with a cell having rock- 

 salt ends, and they show that absorption becomes manifest after 

 a certain thickness in a diathermic liquid is obtained. This is 

 not the case, however, when glass instead of rock-salt cells are 

 employed to contain the same liquids: here the anomaly is still 

 more striking. 



The first experiment in the next series was made with a cell 

 having plane and pavallel glass ends 1*2 inch diameter. Repre- 

 senting by 100 the transmission through the empty glass cell, 



Phil Mag. S.4. Vol. 36. No, 242. Sept. 1868, P 



