490 Prof. Tyndall on the Action of Free Molecule* on 



that its heat, as coming from aqueous vapour, is absorbed 

 with special energy by aqueous vapour, is also a strong re- 

 commendation. In the Bakerian Lecture for 1864 I have 

 illustrated this point. It was then shown that when a platinum 

 spiral, rendered incandescent by a vol Laic current, had 5*8 

 per cent, of its heat absorbed by undried air, a hydrogen- 

 flame had from 17 to 20 per cent, of its heat absorbed. The 

 mere plunging of a platinum spiral into the flame caused the 

 absorption to foil from 17 per cent, to 8' 6 per cent. Hence 

 my reason for choosing a hydrogen-flame in the present 

 instance. 



Dry air and humid air being caused to occupy the experi- 

 mental tube in succession, both of them were compared with 

 the radiation through the tube when very perfectly exhausted 

 by a Bianchi's pump. The following results were obtained : — 



Deflection. Absorption per 100. 



Vacuum 0°'0 0*0 



Dry air 0°'0 0-0 



Humid air .... 8°-0 11'7 



Vacuum 0°'0 0-0 



Dry air 0°'0 0-0 



Humid air .... 7°-6 11-2 



The deflection through a vacuum, otherwise the total heat, 

 was 47°, which, according to my calibration-table, is equivalent 

 to 68 units. 



A fresh supply of hydrogen was here introduced into the 

 gas-holder. The total heat being 46° or 65 units, the follow- 

 results were obtained : — 



Deflection. Absorption per 100. 



Dry air 0°-0 0-0 



Humid air ... . 6°'2 9-5 



Dry air 0°-0 0-0 



Humid air .... 6°'7 10'3 



Mean of the four determinations with humid air, 10 - 7. 



A column of humid air 38 inches long absorbs, according 

 to these experiments, 10*7 per cent, of the radiation from a 

 hydrogen-flame. I have been criticised for estimating the 

 absorption of the earth's rays within 10 feet of the earth's 

 surface at 10 per cent. This estimate I consider a moderate 

 one; and the foregoing experiments prove it to be so. 



It would be an error to suppose that determinations like 

 these are easily made. They require the most scrupulous care 

 for their successful accomplishment. The hydrogen issued 



