386 Royal Society : — 



conclusions are arrived at, namely : — (1) that the only nucleus 

 capable of suddenly crystallizing any one of such solutions is a salt 

 of the same kind as that dissolved ; and (2) that all bodies, solid, 

 liquid, or aeriform, which apparently act as nuclei, are really con- 

 taminated with a hydrate of the salt that forms the supersaturated 

 solution. 



I cannot refrain from expressing my admiration at the unwearied 

 skill and patience with which these two memoirs were prepared. 

 The experiments were repeated by hundreds, and under a large 

 variety of circumstances, so that it seems scarcely possible to 

 entertain any doubt as to the validity of the conclusions arrived at. 

 I had not seen these memoirs until long after the publication of my 

 second paper on this subject* ; or I should have hesitated in offering 

 it to the Royal Society without special reference to them. What 

 I did see was a very brief abstract of M. Grernez's memoir in the 

 ' Comptes Bendus;' and to this I refer in my first paper f, quoting 

 the experiments and the decisive objections of M. Jeannel in oppo- 

 sition to M. Grernez's conclusions. 



The experiments of Mr. Liversidge $ are identical in principle with 

 those of M. Viollette, and some of them mere variations, such as the 

 proof that bodies greedy of water and capable of being hydrated do 

 not produce crystallization ; only M. "Viollette made use of calcined 

 sulphate of copper instead of calcic chloride &c.§ Moreover no 

 fresh proofs are wanted as to the non-nuclear action of the modified 

 salt or of the anhydrous salt on supersaturated solutions of Griauber's 

 salt. 



Eecently MM. Gi-ernez || and Viollette % have each, on the occasion 

 of the publication of my third paper (namely, that written in con- 

 junction with M. Van der Mensbrugghe**), called in'question the 

 integrity of my experiments, M. Grernez insisting that the oils and 

 other liquids employed by me contain salts of the same kind as 

 those of the solutions they apparently acted on. 



Such an assertion as this seems to me to be very difficult of 

 proof ; and it seems equally bold to assume that the air of an 

 open garden in the country contains salts of various kinds, watching 

 their opportunity to get into my flasks and vitiate the results of 

 my experiments. 



I desire to invite special attention to a statement made in my 

 second paper — the more so because, without desiring for a moment 

 to call in question the negative results obtained by MM. Grernez 



* Phil. Trans. 1871, p. 51, 



t Phil. Trans. 1868, p. 660. The reference to M. Jeannel is in the abstract 

 of this paper contained in the ' Proceedings of the Royal Society,' May 28, 1868, 

 p. 405. 



% Proceedings of the Royal Society, vol. xx. p. 497. 



§ On the surface-tension theory, it is not inaccurate to say that absolute alcohol 

 robs the solution of water, and so determines crystallization, since the mixture 

 of alcohol and water degrades the tension of the solution. 



U Comptes Rendus, vol. lxxv. p. 1705. •[ Ibid. vol. lxxvi. p. 171. 



** Proceedings of the Royal Society, vol. xx. p. 342. 



