482 Mr. J. Gill on the Temperature of the 



confirmed this opinion, which is no doubt correct for simple 

 liquids in general, and also for solutions as to molecular action 

 in the phenomena ; that is to say, each particle of steam is 

 formed in a boiling solution under a tension corresponding to 

 the tension of the vapour of pure water boiling under pressure 

 at the same temperature, the difference between this higher tem- 

 perature and the common boiling-point of pure water being due 

 to the attraction between the particles of the salt and those of 

 the water. But it would naturally be supposed that, as soon as 

 the newly-formed particles of vapour broke through this sphere 

 of attraction in coalescing to form bubbles, they would at once 

 expand into steam of a tension corresponding to the pressure of 

 the general atmosphere under which they existed, and conse- 

 quently would form (under atmospheric pressure) common satu- 

 rated steam at 100°, unless they were superheated by the higher 

 temperature of the boiling liquid. 



It seems probable that the smallest visible steam-bubble 

 formed in the mass of a boiling liquid should be composed of a 

 vast number of individual vapour particles. It may be imagined 

 that the spherical layer of vapour particles forming the shell of 

 the bubble in contact with the surrounding liquid is at the real 

 temperature of the liquid, and under a tension corresponding to 

 that temperature ; but the mass of particles forming the body 

 of the bubble should apparently be at distances from the liquid 

 sufficiently great to place them beyond the attractive influence 

 of the dissolved salt; and if so, this small mass of steam may be 

 supposed, at the instant of its formation, to assume a density 

 and temperature corresponding to the local pressure. If it re- 

 mained for some time imbedded in the mass of hotter liquid, it 

 might naturally be supposed to become superheated ; but the 

 manner in which the communication of heat from the liquid to 

 the vapour might be effected is not perhaps easy to be imagined, 

 nor could the temperature of an individual bubble be practically 

 ascertained. But by surrounding a mass of steam with walls 

 kept constantly wet with a film of the hot solution, it may be 

 placed in circumstances nearly equal to those of a steam-bubble 

 in the body of the liquid, and thus its thermic character maybe 

 more conveniently examined. With this view I have used the 

 simple little apparatus shown by the annexed figure ; and the 

 results of some experiments I have made with it are, I think, suf- 

 ficiently remarkable to claim serious attention. 



The cylindrical vessel a of tin plate, 4 inches in diameter and 7 

 inches high, has a moveable inner concentric shell b, open above 

 and below, held in place by four slight elastic stays c c, leaving 

 a free passage of about 1 inch between its lower edge and the 

 bottom of the vessel a. The position of the moveable cover d is 



