1862.] on Mr, Graham* s Researches on Dialysis. 423 



whether water or alcohol, for instance. The diffusion of any particular 

 salt is scarcely affected by the presence of a different salt, though 

 materially affected by the presence of the same salt in the external 

 liquid. The general law seems to be, that the velocity with which a 

 dissolved salt diffuses from a stronger into a weaker solution is propor- 

 tional to the difference of concentration, quoad that salt, between the 

 contiguous solutions. The phenomena of liquid diffusion are mani- 

 fested most simply and uniformly when dilute solutions only are 

 employed. 



Jar Diffusion, — When the solution of the salt to be diffused, 

 instead of being placed in a vial, is conveyed by means of a pipette to 

 the bottomof a jar of water, the dissolved salt gradually rises through 

 the superincumbent water to a height or extent proportional to its 

 diffusibility. The results of jar diffusion bear out generally those of 

 vial diffusion. They show, moreover, the absolute rate or velocity of 

 the diffusive movement. Thus, during a fourteen days' aqueous diffu- 

 sion from ten per cent, solutions of gum-arabic, Epsom salt, and 

 common salt respectively, the gum-arabic rose only through T^ths of 

 the superincumbent water, or to a height of 55*5 millimetres ; while 

 the Epsom salt rose through the whole irths of the superincumbent 

 water, or to a height of 11 1 millimetres ; while the common salt not 

 only rose to the top, but could have risen much higher, seeing that the 

 uppermost or 14th stratum of water into which it had diffused, con- 

 tained about fifteen times as much salt as was contained in the upper- 

 most or 14th stratum of water into which the Epsom salt had 

 diffused. 



Diffusive Sepakations. — If a solution containing equal weights 

 of common salt and gum arable be poured into a jar of water, the ratio 

 of salt to gum in the dilute liquid in the jar will be the same as in 

 the original solution. But if the original solution be poured into a 

 diffusion vial from which the dissolved compounds can diffuse into 

 a jar of water, for every 100 milligrammes of salt, about 22^ 

 milligrammes of gum will pass out into the external water, or the 

 ratio of salt to gum in the dilute liquid in the jar will be as 100 

 to 22^y instead of as 100 to 100, or a separation of the gum from 

 the salt to the extent of 77^ per cent, will have been effected. 

 Again, when a solution containing 5 per cent, of common salt and 

 5 per cent. Glauber's salt (the diffusive rates of which salts are 

 to one another as 100 to 70) is submitted for seven days to the process 

 of jar diffusion, the upper half, or -^ths of superincumbent water, will 

 be found to contain 380 milligrammes of common salt and only 53 

 milligrammes of Glauber's salt, or the ratio of common salt to Glau- 

 ber's salt on the upper half of the liquid will be as 100 to 14, instead 

 of as 100 to 100, or a separation of the Glauber's salt from the 

 common salt to the extent of 86 per cent, will have been effected. Not 

 only the partial separation of mixed bodies, but the partial decomposi- 

 tion of definite chemical compounds maybe effected by diffusion. Thus 

 when alum, which is a double sulphate of the two metals, potassium 



