366 
DR. E. DIALERS ON THE UNION OF 
proportion of nitrate is mixed with water, is to be attributed principally to the fact of 
the nitrate being in more dilute solution than in the other case described ; for it is found 
on diluting a concentrated aqueous solution that the cold is produced in the earlier 
stages of the dilution. It is perhaps also to be attributed, to a slight extent, to the 
presence of more ammonia than in the other cases ; since it is possible that ammonia 
in the liquefied condition may generate a little heat when mixed with water, and this 
must mask the cold produced by the nitrate to a corresponding extent. 
The accompaniment of contraction of volume by absorption of heat, here illustrated, 
has received several explanations. Gay-Lussac, in accounting for it, considered that the 
specific heat of the mixture must be greater than the mean of the specific heats of the 
saline solution and the water ; but Thomsen’s* researches have shown that the specific 
heat of a mixture of an aqueous solution and water is less than the mean of the specific 
heats of the solution and water. Thomsen has also pointed out that where (as in the case 
of the salts of the fixed alkalies) the molecular volume is greater than the sum of those 
of its constituents, the molecular heat of the compound is greater, and that, on the 
other hand, where (as in the case of ammonia salts) the molecular volume is less, the 
molecular heat is also less ; so that it appears that volume and specific heat lessen and 
increase together. 
Another explanation is one that has been offered by Bussy and BuiGUETf, whose 
attention was drawn to this subject by their observation of the simultaneous contraction 
in volume and absorption of heat that occurs when liquid hydrocyanic acid and water 
are mixed together, and who have established that no apparent relation exists in these 
cases between changes in volume and changes in temperature. They ascribe the 
production of cold to the diffusion of the one liquid through the other, by which, 
according to their notions, each occupies a greater volume than it did before. But 
although after admixture each liquid does extend through a greater space, it does not 
do so by undergoing an increase in volume, but by being displaced by the other liquid, 
the space occupied by the mixture being filled up by the two liquids conjointly — 
assuming, that is (as is necessary in applying the above explanation), that the two liquids 
preserve their individuality during admixture. Now, unless diffusion involves expan- 
sion, it is hardly evident how it can afford any solution of the difficulty. 
F. Mohr J has suggested that the heat which disappears is the excess of that used up 
in giving to the mixture the property of remaining liquid at a lower temperature than 
water itself can, over that evolved in giving it the property of remaining liquid at a 
higher temperature than the water. This is apparently to assume that the latent heat 
of water is variable and inversely related to the temperature of freezing, just as the 
latent heat of steam is variable and inversely related to the temperature at which it is 
formed ; and the assumption seems probable enough. 
The anomaly that presents itself, when the density of the water is considered to 
* Pogg. Ann. cxlii. p. 337; Chcm. Soe. Journ. [2] ix. p. 468. t Ann. Chim. Pliys. [4] iv. p. 5. 
$ Deut. Chem. Ges. Ber. iv. p. 314. 
