Radiant Heat, and its Conversion thereby into Sound. 491 



from the gas-holder through an orifice of fixed dimensions in 

 a stream of the utmost possible constancy. It was then led 

 into a Sugg's regulator, whence it issued under an absolutely 

 constant pressure. The flame issued from a circular brass 

 burner with an aperture j\ of an inch in diameter. It was 

 carefully surrounded by a hoarding, the space within the 

 hoarding being packed with horsehair. Every precaution 

 in fact was taken to avoid the agitation of the air around the 

 flame. Proper care was also taken to secure the pile against 

 disturbance by air-currents. The air being first purified, by 

 passing it through caustic potash and sulphuric acid, was 

 rendered humid by carrying it over wet bibulous paper con- 

 tained in a suitable tube. It required some minutes to enter; 

 and it was therefore necessary, by prior patient observance of 

 the needle, to make sure that during this interval no change 

 occurred in the radiation save that effected by the humid air 

 itself. The humid air was removed from the experimental 

 tube, not by exhaustion (which always causes precipitation), 

 but by gently forcing, by means of a compressing-pump, dry- 

 air through the tube. When this was done, the needle, in all 

 the experiments above recorded, returned within a small frac- 

 tion of a degree to zero*. 



With the rough, wide experimental tube to which reference 

 has already been made, I, ten years ago, found the absorption 

 of a column of humid air 38 inches long to be 8 per cent, of 

 the total radiation from a flame of hydrogen. 



§ 6. Conservation of Molecular Action. 



If the absorption of radiant heat be the act of the consti- 

 tuent atoms of compound molecules, its amount depending 

 solely on the number of molecules encountered by the calorific 

 waves, then, whatever may be the changes of density which 

 gases and vapours undergo, so long as the number of molecules 

 remains the same, the absorption ought to continue constant. 

 Such constancy, should it be proved to exist, I name the " con- 

 servation of molecular action." The experiments now to be 

 described deal with this question. 



Besides the silvered experimental tube already described as 

 38 inches long, I had another constructed of the same diameter, 

 and with similar terminal apertures. Its length was 10*8 

 inches. The one tube was, therefore, 3*5 times the length of 

 the other. The shorter tube was constructed with the view 

 of proving that the absorptions I had recorded in previous 



* The temperature of the laboratory air during the foregoing experi- 

 ments was 60° Fahr. 



