364 



Mr. J. Joly. 



the surface ; in other words, the formation of mist by radiation from 

 the steam to the cold substance. 



Now it is assured that on the replacement of air by steam a film 

 of water is instantaneously deposited on the surface of the substance, 

 and, assuming an abundant supply of steam, condensation then pro- 

 ceeds as fast as heat is transferred by conductivity and diffusion 

 across this film to the cold surface of the solid beneath. Whi'e then 

 we must assume that the inner surface of the film is appreciably at 

 the temperature of this cold surface, it is probable that we may also 

 assume the outer surface of the film to be throughout the experiment 

 nearly at the temperature of the steam. There can be but little 

 difference, for fall in temperature is made good with great rapidity 

 by fresh condensation. There is here in the first place a favourable 

 condition, as surface radiation can, under such circumstances, be but 

 insignificant in amount, and radiation from points in the surrounding 

 steam can only be supposed as affecting points within the film or, in 

 some cases, within the substance. Only in this way can we suppose 

 the formation of precipitation elsewhere than at the very surface of 

 the film by actual contact. But again, the opacity of water to radia- 

 tion from aqueous vapour will here afford protection. This opacity, 

 it will be remembered, has been shown by Tyndall and others to be 

 very great. Thus I find that Tyndall records among his experiments 

 that a layer of water but 0*07 of an inch in thickness transmitted but 

 1*1 per cent, of the radiation from a hydrogen flame. It may be 

 supposed that the more perfect accord between the vibration periods 

 of radiator and absorber obtaining in the calorimeter will ensure 

 greater, if not perfect, opacity. 



The contact precipitation, consequent on the loss of vis viva of the 

 impinging molecules — the film transmitting the energy to the sub- 

 stance — insures that there will be many molecules entering and few 

 leaving, and hence extension of free path or diminution of pressure in 

 directions normal to the surface of the substance. There is then super- 

 added as a protective element a converging drift of steam upon the sub- 

 stance, tending to restore to it what precipitation may possibly be due 

 to radiation. Under perfect conditions of steam supply — could we 

 suppose, for example, the substance, in a chamber which could be 

 supplied with steam from so many points that we might substitute 

 the idea of an unlimited region of steam suddenly brought around 

 it— this drift, I think, excludes almost the possibility of error supposing 

 radiation. In attainable conditions, where cross draughts will prevail 

 to some extent, there would still evidently be protection involved in 

 the very fact of condensation. 



There are, then, three circumstances tending to nullify this error ; 

 the high temperature of the external surface, the adiathermanous 

 nature of the water film, and the converging drift of steam. Whether 



