414 Prof. Magnus on the Influence of the Adhesion of 



fluenceof the Absorption of Heat on the Formation of Dew"*, 

 which contains a comparison between the radiating-powerof dryair 

 and of that saturated with aqueous vapour. In this it was un- 

 doubtedly found tbat,so long as the vapours present in airwere per- 

 fectly transparent (that is, were vapours in the proper sense), there 

 was no appreciable difference between the absorptive power of 

 dry air and of air saturated with moisture, but that a difference, 

 both as regards absorption and radiation, was set up when the 

 vapours began to deposit in the form of mist. Prof. Tyndall 

 thereupon again defended his statementf, while a third physicist 

 afterwards appeared as an arbiter in the matter. 



Professor Wild of Berne has made comprehensive experiments 

 by TyndalPs methods and found his statements thoroughly con- 

 firmed. Whoever reads Professor Wild's paper J will lay it down 

 with the impression (as could not indeed but have been ex- 

 pected from so sound a physicist) that the statements are 

 completely reliable. I, at least, confess that this was the case 

 with myself — and that I could not see how those results could 

 be reconciled with my own observations, which I knew were 

 made with the greatest care, and without any preconceived opi- 

 nion. I determined, therefore, once more to take up the sub- 

 ject, and first of all to repeat Professor Wild's experiments. 

 This was easy, because Prof. Wild has accurately given the di- 

 mensions of his apparatus, — a gift which I recognize with great 

 thankfulness ; for, as will afterwards be seen, very much depends 

 on these dimensions. The arrangement of the apparatus was as 

 follows : — 



On each side of the thermopile, provided with its cones, a brass 

 tube, 60 centims. long and 6 centims. in diameter, was fixed 

 horizontally in such a manner that the axes of both tubes were 

 in the same line as that of the thermopile. In the prolongation 

 of this line, and at right angles to it, there was placed in front 

 of each tube a blackened metal cube, in which water was kept 

 boiling by a lamp. Each of these cubes, which served as sources 

 of heat, was at a distance of 10 centims. from the end of its 

 tube, and the other ends of the tubes were at the same distance 

 from the cones of the thermopile. Between one of these tubes 

 and its cube a screen, moveable by means of a screw, was placed, 

 by which part of the rays of this cube could be caught ; so that 

 the radiation and heating of the pile on both sides was always 

 equal and the galvanometer at rest. At distances of 15 centims. 

 from each end the brass tubes were provided with lateral apertures 



* PoggendorfFs Annalen, vol. cxxvii. p. 613. Phil. Mag. S. 4. vol. xxxii. 

 p. 111. 



f Phil. Mag. S. 4. vol. xxxii. p. 118. 



% Ibid. p. 241. Pogg. Ann. vol. cxxix. p. 57. 



