and Attached Water. 119 



Graham has shown how alumina, silica, oxide of tin, and 

 oxide of iron may be obtained in aqueous solution in the 

 colloidal state ; and Daubree has proved that at high tem- 

 perature and under pressure water disintegrates and decom- 

 poses certain rocks, and rearranges their constituents, and 

 has argued that, assisted by capillarity and high temperature, 

 many phenomena of volcanism and metamorphism are to be 

 attributed to that agent. He has thus vastly extended the 

 scope of the long-known action of water on silicates, and 

 showed that under conditions of temperature water alone 

 plays as potent a part in such disintegration as the fixed 

 alkalies were known to do under like conditions. My experi- 

 ments may perhaps be considered as leading us a step further. 

 For they show that water at a high temperature may not 

 only play the part of a solvent in the ordinary restricted 

 sense, but that there is in many cases no limit to its solvent 

 faculty ; in other words, that it may be miscible with certain 

 rocks in all proportions : that solution and mixture are con- 

 tinuous with one another. And this continuity, as my experi- 

 ments prove, is established in some cases, — and these indeed 

 with bodies having no chemical affinity with water, — at tem- 

 peratures not above the temperatures of fusion of those bodies 

 per se. This induces me to think that the replenishment of 

 water in rocks by capillarity, an action upon which Daubree 

 lays great stress and with regard to which he adduces many 

 striking experiments, is not an essential condition. I must, 

 however, leave the discussion of this question to penologists. 



[Note added June 16, 1884. — Since the above was in print 

 I have been favoured with a copy of a most interesting 

 memoir by Prof. Tilden and Mr. W. A. Shenstone on the 

 " Solubility of Salts in Water at High Temperatures," read 

 before the Royal Society of London, June 21, 1883 (Transac- 

 tions of the Royal Society, Part I. 1884). Accordingly, 

 wherever in the two memoirs there may be found similarity 

 in results or ideas, the priority is theirs. 



These gentlemen have apparently been to a considerable 

 extent guided by the conception that there is a relationship 

 between the solubility of a salt in water and its temperature 

 of fusion. And perhaps their main argument concerns this 

 relationship. They have made a special study of those inter- 

 esting cases in which the solid salt contains water of crystal- 

 lization. This branch of the inquiry I have rather deliberately 

 avoided, as I wished to establish the analogy between metallic 

 and dry-salt alloyage, on the one hand, and water-salt alloyage 

 on the other. Nevertheless they examined the solubility of 



