380 On Re gelation. [1858. 



surrounded by water tend to retain their fluid state in both 

 directions at temperatures which are abundantly sufficient to 

 make it equally retain the solid or the vaporous state when either 

 of them is conferred upon it. There is nothing against the 

 assumption that ice has the like kind of power, i. e. the power 

 of retaining its solid state at temperatures higher than the 

 temperature of ice against water. Nevertheless, the fact is 

 more difficult to show ; still some experiments may be quoted 

 in favour of the view. If hydrated crystals of sulphate 

 of soda, carbonate of soda, phosphate of soda, &c.*, be care- 

 fully prepared in clean basins, by spontaneous evaporation 

 of the water they will retain their form unbroken, and their 

 hydrated state undisturbed, through the high temperatures of 

 a whole summer ; though if broken or scratched even in winter, 

 they will commence to effloresce at the place where the cohesion, 

 and with it the balance of force was disturbed, and will from 

 thence change progressively throughout the whole mass f. 

 As regelation concerns the condition of water, there is per- 

 haps no occasion to go further. Such facts as the following, 

 however, concern the extension of the principle and illustrate 

 the power of cohesion, especially in cases where it is coming 

 into activity. Camphor in bottles, or iodide of cyanogen in 

 proper glass vessels produce crystals sometimes an inch or 

 two in length, which grow by the deposition of solid matter 

 on them from an atmosphere unable to deposit like solid 

 matter upon the surrounding glass, except at a lower tempe- 

 rature. Crystals in solution grow by the deposition of solid 

 matter on them which does not deposit elsewhere in the solu- 

 tion : many such like cases may be produced. 



Returning to the particular case of regelation, it is seen that 

 water can remain fluid at temperatures below that at which ice 

 forms, by virtue of the cohesion of its particles, and in so far 

 the change is rendered independent of a given temperature. 

 Next, I rest on the fact that ice has the same property as 

 camphor, sulphur, phosphorus, metals, &c., which cause the 



* Philosophical Transactions, 1834, p. 74 ; or Exp. Res. Electricity, vol. i. 

 p. 191, note. 



t Such a case shows combined solid water at a temperature ready to separate 

 and change into vapour, yet not changing, because, as far as we can see, the 

 undisturbed cohesion holds all together. 



