188 PROFESSOR THOMAS GRAHAm's SCIENTIFIC WORK. 



temperature, and the reverse substitution of water for the basic ether 

 of sulphethylic acid at a higher temperature, had been represented as 

 depending on the augmented elasticity of the ether vapor at the higher 

 temperature. Mr. Graham showed, however, that ether couhl be very 

 readily formed by heating the mixture of sulphuric acid and alcohol in 

 sealed tubes — that is, under conditions in which the augmentation of 

 volatility due to heat -was 2)(iri passu counterbalanced by the diminution 

 of volatility due to pressure. Altogether, Mr. Graham supported the 

 contact theo^'y of ether formation, as opposed to the then received re- 

 a(;tion theory; but several of his experiments afforded clear, though in- 

 deed supererogatory, su[)port to the reaction theory soon afterward in- 

 troduced by Williamson. 



In addition to the memoirs cited above, the question of hydration 

 formed an express or incidental subject of many other of Mr. Giaham's 

 investigations. It is noteworthy that, for him, osmosis became a me- 

 chanical effect of the hydration of the septum ; that the interest attach- 

 ing to liquid-transpiration was the alteration in rate of passage conse- 

 quent on an altered hydration of the liquid ; that the dialytic difference 

 between crystalloids and colloids depended on the dehydration of the 

 dialytic membrane by the former class of bodies only ; and similarly in 



man\ other instances. 



III. 



Movements of liquUls under pressure. Transpiration. — That the ve- 

 locities with whi<;h different liquids, under the same pressure, issue 

 from a hole in the side or bottom of a vessel should be inversely as 

 the square roots of their respective specific gravities is a projiositiou 

 deducible from well-known mechanical principles. As demonstrated, 

 however, by Dr. Poiseuille, this law is not applicable to the case of 

 liquids issuing und(>r pressure tlirough capillary tubes. In addition 

 to determining experimentally the laws of the passage of the same 

 liquid — that the velocity is directly as the pressure, inversely as the 

 length of the capillary, and directly as the fourth power of the 

 diameter, and that it is accelerated by elevation of temperature — 

 Dr. Poiseuille further showed that the rate of passage of different liquids 

 through capillary tubes is for the most part a special property of the 

 particular liquids ; and that while the rate of passage of water, for 

 instance, is scarcely affected by the presence of certain salts in solution, 

 it is materially accelerated by the presence of chlorides and nitrates of 

 potassium and anunonium, and materially retarded by the presence of 

 alkalies. He also showed that while the rate of passage of absolute 

 alcohol is much below that of water, the rate of passage of alcohol 

 diluted with water in such proportion as to form the hydrate, nGC2 0. 

 3 Aq, is not only much below that of alcohol, but also below that of 

 any other mixture of alcohol and water. 



Some time after Dr. Poiseuille's death Mr. Graham, starting from this 

 last observation, took up the inquiry. Giving to the phenomenon itself 



