Radiant Heat, and its Conversion thereby into Sound. 445 



elusion that the veritable radiators and absorbers are the 

 molecules of the vapours. Apart from all experiment, the 

 notion that vapours must act in this way conies commended 

 to us by the proved, and conceded, deportment of gases. It 

 would be unreasonable to admit that a compound gaseous 

 molecule is active, and at the same time to affirm that a 

 compound vaporous molecule is inert. 



This hypothesis of liquid films formed on the interior surface 

 of the experimental tube, and on the plates of rock-salt, becomes, 

 I think, more embarrassed as we proceed. It depends on the 

 unproved assumption that liquids possess powers of absorption 

 which are denied to their vapours. To water and brine, for 

 instance, Magnus largely concedes such powers, but not to 

 aqueous vapour. That the state of aggregation exerts an 

 influence is not denied; but that it is here the dominant factor 

 is open to doubt. To admit this would be to concede that 

 the seat of absorption is the molecule as a whole, to the 

 practical exclusion of the constituent atoms of the molecule. 

 For if the atoms exert any influence, the mere passage from 

 the liquid to the vaporous condition, which separates the 

 molecules from each other, but leaves them individually intact, 

 cannot destroy their powers of absorption. 



At an early stage of these researches the parallelism of 

 liquid and vaporous absorption forced itself upon my attention. 

 Thus my experiments on bisulphide-of-carbon vapour were 

 connected with the deportment of liquid bisulphide, as set 

 forth in Melloni's table. The vapours, moreover, of chloride 

 of sulphur and chloride of phosphorus, whose liquids stand in 

 Melloni's table next to bisulphide of carbon, were afterwards 

 proved by me to possess a diathermancy corresponding to 

 that of their liquids. After various references to this subject 

 in preceding memoirs, a portion of the Bakerian Lecture for 

 1864 was devoted to its examination. Liquid layers enclosed 

 between plates of transparent rock-salt were tested in regard 

 to their diathermancy; and, for the sake of control and veri- 

 fication, they were employed in five different thicknesses. 

 The vapours of these liquids were examined in quantities pro- 

 portional to the quantity of liquid, the same quality of heat 

 being applied both to liquids and vapours. By these experi- 

 ments it seemed to me placed beyond a doubt that the 

 liberation of the molecule from the liquid condition did not 

 destroy its absorbent power, the order of absorption being 

 proved to be precisely the same for liquids and their vapours.- 

 Ten different substances were shown at the time here referred 

 to to follow this rule. The list has since been extended; and 

 I am not acquainted with a single real exception to the rule. 



Phil. Mag. S. 5. Vol. 13. No. 83. June 1882. 2 L 



