216 Mr. F. Guthrie on Salt Solutions 



curate and scrupulous, we find that in many cases a less simple 

 numerical ratio between the salt and its water would often cor- 

 respond with the derived result far better than the ratio which 

 has been thence deduced. 



§ 68. With regard to the solidifying-point of the cryohy- 

 drate of iodide of sodium, I may here at once mention the ex- 

 ceedingly interesting fact which will be discussed in its proper 

 place, that while the cold of a cryogen formed by mixing 

 with ice any one of the other of these eight salts is so closely 

 near as to be considered identical with the solidifying point of 

 the corresponding cryohydrate, the cold of a freezing-mixture 

 consisting of ice and iodide of sodium far exceeds —15°, and in 

 fact reaches —28°, nearly the lowest temperature which I have 

 yet got. On this ground it would be really entitled to be placed 

 at the head of the list of the nine, where its water-worth has 

 placed it in Table (§ 66). 



§ 69. Speculation concerning Nal. — It is clear that when a 

 freezing-mixture is in action, the liquid portion is a solution of 

 the salt of such a strength that it resists solidification at the 

 temperature of the mixture. Accordingly there must be a cryo- 

 hydrate of sodinm which remains liquid down to — 28°. Is its 

 crystalline form so peculiar as not to respond by solidification to 

 any solid particles in the air, and so to be an exceptionally per- 

 sistent instance of supersaturation ? Perhaps the cryohydrate 

 can only exist in the liquid form. When it loses heat, it does 

 not solidify as a whole, but in two parts, each of which is less 

 soluble than the two together. For this reason the mixed solid 

 separated would in all its stages have the same composition as 

 the simple cryohydrate, while its separation, when once begun, 

 might keep itself in activity. The heat-tension then exhibited 

 would be the mean of the temperatures due to the solidification 

 of each constituent. 



I can scarcely assert that this is a satisfactory or even to my 

 own mind a clear explanation of the phenomenon. The diffi- 

 culty would be to some extent removed if we could get evidence 

 of an intermediate cryohydrate resembling that which exists 

 with NaCl. Of such a cryohydrate I have no substantial evi- 

 dence to offer at present. 



Cryohydrates of Alkaline Sulphates. 



To see whether the noticeable relation between sodium, am- 

 monium, and potassium is valid in other compounds of these 

 metals, their sulphates were examined. 



§ 70. Sulphate of Ammonium. — This body, when dissolved to 

 saturation in water, yields abundantly the ordinary hydrated 

 sulphate of ammonium when cooled to 0° C. The separation 



