Vapours of Boiling Saline Solutions. 487 



rent of steam sweeping over the surface, and continually remo- 

 ving the superheated particles supposed to exist, and replacing 

 them with particles of fresh saturated steam at 100°, it is evident 

 that the cooling brine must go on absorbing and condensing 

 more and more steam until in time it would become so diluted 

 as not to differ much from pure water; and, as an inevitable 

 consequence, the temperature must lower in proportion. Now 

 experiment proves that under the conditions mentioned the tem- 

 perature does not lower ; consequently we may conclude that no 

 communication of heat takes place between the brine at 109°, 

 and the steam at 100°. There seems to be reason to believe 

 that similar results, varying in degree, would be obtained with 

 aqueous solutions of other salts, and with liquids in general when 

 their boiling-point is raised by their holding some substance in 

 solution, — probably also with any amalgam of mercury boiling 

 at a higher temperature than pure mercury. It might also be 

 fairly deduced from analogy, that hot aqueous solutions, though 

 refusing to communicate heat to colder saturated vapour of water, 

 would freely equalize their temperature with a saturated mixture 

 of air and steam at 100°, the air alone receiving heat in the pro- 

 cess directly from the liquid, and communicating heat to the 

 steam with which it is mixed. 



It is not easy in these experiments to maintain the requisite 

 circumstances sufficiently constant to obtain very precise results. 

 In the apparatus described above, the cooling of the vessel a from 

 exterior influences causes the film of brine with which its inner 

 surfaces are coated at first to absorb and condense steam. The 

 outer surface of the shell b would radiate heat to the colder walls 

 of a ; and the films of brine with which it is coated on both sides 

 would consequently absorb and condense steam, and so get gra- 

 dually washed, while the heat conducted upwards through the 

 metal from the boiling brine below would then tend to dry the 

 surfaces and thus superheat the steam. In a variation of the 

 experiment the bottom of a was enlarged, and a concentric case 

 was soldered round it, which was kept full of brine maintained 

 at the boiling temperature by the lamp below ; but it was per- 

 ceived that the slight hydrostatic pressure of the column of liquid 

 was sufficient to make the boiling-point of this mass of brine 

 higher than that of the brine boiling at h ; and hence superheat- 

 ing of the steam would result directly from the drying of the 

 inner surfaces of a. If the experiments are made as described 

 in the preceding pages, the first trial generally shows a rapid 

 fall of thermometer No. 2 to near 100°. If a little time is allowed 

 to elapse and the experiment is again tried by simply reheating 

 the thermometer, it does not generally show the same fall of 

 temperature the second time. The brine should be removed each 



