[ 530 ] 



L. A Study of an Apparatus for the Determination of the 

 Hate of Diffusion of Solids dissolved in Liquids. By 

 Albert Griffiths, M.Sc.(Vic), A.R.C.S.(Lond.)* . 



Section I. 



FOR some time the author has been engaged in some 

 experiments with the object of determining the coeffi- 

 cient of diffusion of bodies dissolved in water. It may be 

 some years before experimental results of any decided value 

 are obtained ; and perhaps he will be allowed to give an 

 account of the calculations involved, and of the methods he 

 has adopted in obtaining an idea of the probable order of 

 magnitude of the errors to which the apparatus is liable. 



The apparatus consists of a 



Fig. 1. 



vessel V, divided into two com- 



partments, U and L, by a dia- 

 phragm through which pass a 

 number of equal vertical tubes 

 of which two only are shown in 

 the figure. Two tubes, A, B, 

 provided with stop-cocks, pass 

 into U. One, A, just enters U ; 

 the other passes down to the 

 bottom. Similarly two tubes C 

 and D pass into L ; C just enters, 

 and D reaches to the top. The 

 vessel is first completely filled 

 with water, and in the case of 

 a substance which produces an 

 aqueous solution with a specific 

 gravity greater than unity, the 

 tubes A and B are closed, and 

 the solution is passed in through 

 C, the water of L being allowed 

 to escape through D. Diffusion 

 along the vertical tubes now com- 

 mences, and the compartments U 

 and L are alternately and periodi- 

 cally refilled with pure water and 



solution respectively. A quantitative analysis is made of the 



liquid taken from the upper compartment. 



Section II. — Calculation of the Coefficient of Di fusion. 

 Lord Kelvin has solved the problem of the flow of elec- 

 tricity along a cable possessing appreciable capacity. This 

 * Communicated by the Physical Society : read Feb. 24, 1899. 







B- 





-A 



















U 

















L 















D" 



-c 







