affecting Communication of Heat. 421 



temperature of 212°. In the former case the contact of ice with 

 overcooled water determines an instantaneous congelation of part 

 of the water with a rise of temperature to 32° ; and in the latter 

 case the contact of steam at 212° causes a sudden ebullition of 

 the overheated water, with a fall of temperature to 212°, and so 

 of analogous experiments with other substances. This, by the 

 old school, might have been taken as a kind of negative proof, 

 — a repugnance on the part of Nature to permit an approach 

 towards the inversion or suspension of a physical law. The 

 immediate cause of the phenomena has been referred by modern 

 investigators to molecular conditions and polar influences in the 

 ultimate particles of matter; and as long as the phenomena 

 were confined to circumstances like those mentioned above, we 

 might perhaps rest satisfied with such explanation. But in the 

 case of aqueous solutions there appears to be something like a 

 more positive proof of the above imagined law. 



A saturated solution of common salt in water boils, under 

 atmospheric pressure,, at 226°, and yet the steam which it emits 

 is only 212°; and so also of other solutions which boil at tem- 

 peratures even beyond 270°. It is well known that though the 

 presence in a liquid of a substance dissolved in it may, to a con- 

 siderable amount, raise the temperature at which the liquid boils 

 under a given pressure, yet, unless the dissolved substance enters 

 into the composition of the vapour, the relation between the 

 temperature and pressure of saturation of the latter remains un- 

 changed. It is not difficult to conceive that the affinity or 

 attractive influence between the particles of the water and those 

 of the dissolved body should hold down the water-particles in 

 close association with the solid particles against the force of heat 

 tending to separate them, and that thus the boiling-point of the 

 solution should be raised ; but it is not so easy to conceive why 

 the steam-bubbles rising from the bottom of a mass of a solu- 

 tion of potash boiling at 272° should escape from the surface 

 as saturated steam of 212° without any indication of having 

 been superheated by contact with the liquid 60° hotter, and 

 by radiation during their transit through this liquid from the 

 bottom to the surface. It seems very probable that, as sup- 

 posed by Regnault, molecularly each particle of steam is formed 

 as saturated steam at a pressure actually corresponding to the 

 temperature of the boiling liquid, the chemical attraction of the 

 dissolved salt coercing it in the same manner as it holds down 

 the particles of water against the act of boiling through a 

 thermometric range of 60° in the case under consideration 

 (from 212° to 272°) ; but when a number of steam-particles 

 coalesce to form a bubble, it is reasonable to suppose that the 

 smallest visible steam-bubble must be of enormous dimensions 



