2 M. Gay-Lussac on the [Jan , 



spirit embraced all the parts of chemistry, is the first person who 

 explained in a satisfactory manner the influence of heat in saline 

 solutions, " If," for example, says this illustrious chemist, " a 

 salt is very little soluble in \vater, and very fusible by heat, it is 

 clear that such a salt will be very httle soluble in cold water, 

 but very soluble in hot water. Such is the nitrate of potash, 

 and above all the hvperoxymuriate of potash. If another salt, 

 on the contrary, isVt the same time little soluble in water and 

 caloric, it will be little soluble both in cold and hot water ; and 

 the difference will not be considerable. This is the case with 

 sulphate of hme." 



" We see then that there is a necessary relation between 

 these three things ; the solubiUty of a salt in cold water, its 

 solubility in boiling water, the degree at which it melts by heat 

 alone, and by the assistance of water. That the solubility of a 

 salt in hot and in cold water is so much the greater the more 

 soluble it is in caloric, or, which comes to the same thing, the 

 lower the temperature at which it is disposed to melt."— (Traite 

 Elementaire de Chimie, ii. 39.) 



These principles, when we consider only some particular 

 examples, appear just and very clear. We conceive in fact that 

 if a salt melts at 2 12°^, it will mix at that temperature with water 

 in every proportion, however httle affinity it may have for that 

 liquid ; but they present a great number of exceptions, and to 

 determine their degree of accuracy and their generality, it is 

 indispensable to examine the solubility of a great number of 

 bodies. 



The determination of the quantity of salt which water can 

 dissolve is not a veiy difficult process. It consists in saturating 

 the water exactly with the salt whose solubility we wish to know 

 at a determinate temperature, to weigh out a certain quantity of 

 that solution, to evaporate it, and w^eigh the saline residue. 

 However, the saturation of water may present considerable 

 uncertainty, and before going further, it is proper to examine the 

 subject. 



We obtain a perfectly saturated saline solution in the two 

 following ways ; by heating the water with the salt, and allowing 

 it to cool to the temperature whose solubility is wanted ; or by 

 putting into cold water a great excess of salt, and gradually 

 elevating the temperature. In each case, it is requisite to keep 

 the final temperature constant for two hours at least, and to stiv 

 the saline solution frequently, to be quite sure of its perfect 

 saturation. By direct experiments made with much care, I have 

 ascertained that these two processes give the very same result ; 

 and that of consequence they may be employed indifferently. I 

 shall mention a few. 



I carried into the caverns below the observatoiy, where the 

 annual temperature does not vary more than -j-io-th of a degree, 

 two solutions of nitre and of sulphate of soda saturated at the 



