of Fotential Enm^gy of TAquid Sitr faces. 43 



meats show precisely this result. I will cite a few examples 

 in support of the last conclusion. 



1. If the solid wetted by the liquid has its maximum density 

 and consequently its minimum surface, the specific • heat is 

 itself a minimum at this point, and takes a greater value on 

 both sides of the temperature of maximum density. This 

 was proved by M. S[)ring, by a very elegant mode of experi- 

 mentation, for the alloys of Rose and Darcet immersed in oil *. 

 But in the immediate vicinity of the maximum anomalies may 

 present themselves, on account of the eventual change of sign 



. dT 

 of -^- ; this, again, was observed by the same physicist for the 



above alloys ; I shall return to it a little further on. 



2. We know how water resists congelation when it is en- 

 closed in capillary spaces, and to what degree organized bodies 

 can support severe cold with impunity because their fluids are 

 enclosed in vessels of microscopic minuteness. In these cases 

 the contact-surface between the solid and the liquid, relatively 

 to the volume of the latter, is enormous. 



3. But the present theoretic deduction is confirmed in an 

 unexpected manner by the remarkable experiments of the 

 eminent physicist M. Melsens f, relative to the truly astonish- 

 ing quantity of heat necessary to detach a volatile liquid fi'om 

 the surface of a very porous substance such as wood-charcoal. 

 Permit me to cite one of those experiments. 23 cubic cen- 

 tims. of sulphide of carbon are introduced into a Faraday tube 

 containing 30 grams of charcoal ; the long branch, containing 

 the charcoal, is introduced into a tin tube filled with water, 

 which is slowly brought to ebullition ; the short branch of the 

 first tube dips into a freezing-mixture which is kept between 

 15° and 17° C. below zero. Under these conditions the v/ater 

 must be heated for moi'e than an hour, in order to collect 2 or 

 3 cubic centims. of sulphide of carbon in the short branch. 



4. Lastly, I will mention a very curious property of gun- 

 cotton covered with a very inflammable liquid (for example, 

 benzine, ether, sulphide of carbon, &c.) and exposed to the 

 action of the electric spark : the liquid alone ignites ; the 

 gun-cotton does not explode. 



II. Let us noAV suppose t invariable, and see what quantity 

 of heat must be supplied to cause the surface S to vary be- 

 tween the limits Si and S2. If Q denote this quantity of heat, 



* "Sur la dilatation, la chaleur specitique des alliages fusibles," &c., 

 £uU. de CAcacl. Roy. de Belgique, 1873; t. xxxix. p. 5-18. 



t " Notes cMiiiiq^Lies et cliimico-pliysiques, 5^ note," § " De la tension 

 dea liquides volatils au contact du charbon," Mem. Couronnes et autres 

 Mem. pahlies par VAcad. Hoy. de Belyique, t. xxiii. 



