Prof, Tyndall on Lunar Radiation. f 471 



The galvanometer was a differential one ; that is to say, two 

 wires ran side by side round the astatic needle of the instrument. 

 The ends of one of these wires were connected with the pile on 

 the roof, the ends of the second wire were connected with a 

 second pile, which was turned towards a vessel kept at a constant 

 temperature by boiling water. The direction of the current 

 caused by the heat below was opposed to that generated by the 

 cold above ; one of them in a great measure neutralized the other, 

 and the needle was thus compelled to take up its place among 

 the lower degrees of the scale. 



I then ascended to the roof, fixed my pile at the proper angle, 

 and directed it off the moon ; I descended and observed the 

 galvanometer; the needle oscillated between 10° and 20°, its 

 mean position being therefore 15°. 



I reascended and turned the pile on the moon ; on descending 

 I found the needle oscillating between 35° and 45°, the mean 

 position being 40°. 



The ascending and descending was repeated six times, and the 

 following results were obtained : — 



Mean deflection. 

 Off the moon. On the moon. 



15 



40 



27 



40 



33 



40 



These numbers all show cold, the deflection being such as 

 would be produced by the cooling of the face of the pile pre- 

 sented to the heavens ; and the result is that the chilling was hi 

 all cases greatest when the pile was directed towards the moon. 



The explanation given of this result in my letter to Sir John 

 Herschel, I think, deals with a true cause. One hot body may, 

 I think, be chilled by the presence of another in virtue of an action 

 on the intervening medium. But whether the cause is sufficient 

 may admit of question. It would not be sufficient if the height 

 „of our atmosphere were restricted to the limits which many assign 

 to it, But if I understood the Astronomer Royal aright at 

 Manchester, there is some reason for supposing the atmosphere 

 to extend immeasurably beyond those limits. But then its ex- 

 treme tenuity at great distances would probably be urged against 

 the possibility of its producing any sensible effect. Tenuity in 

 the abstract, however, hardly furnishes a sufficient argument. 

 In a very few weeks I shall have occasion to show that the action 

 of a stratum of vapour three feet thick, and possessing a tenuity 

 which amounts only to a fraction of that assigned to our atmo- 

 sphere at a height of eighty miles, is capable of accurate niea- 



