374 Proceedings of Royal Society of Edinburgh, [july 15 , 
their crystalline product was found by them not to yield glucose when 
it was heated with dilute sulphuric acid. Hence they concluded 
that strophanthin is not a glucoside (Comptes Rendus de VAcademie 
des Sciences, Ixxxiv., 1877, p. 261 ; and Journal de Pharmacie 
et de Chemie, xxv., 1877, p. 177). The glucosidal character 
of strophanthin, however, has now been amply demonstrated by a 
large number of experiments which I have made, and by the ex- 
periments of subsequent observers, and especially by those of A. 
W. Gerrard, described in an interesting paper published this year 
( The Pharmaceutical Journal and Transactions, 14th May 1887, 
p. 923). Further, the solution obtained when strophanthin is 
decomposed by sulphuric acid has been fermented with yeast, and 
carbonic acid, representing 23*64 per cent, of glucose, has been ob- 
tained. 
4. On a New Diffusiometer and other Apparatus for Liquid 
Diffusion. By J. J. Coleman, F.I.C., F.C.S. 
Supposing a tall glass tube open at both ends be cemented into a 
reservoir packed full of common salt, and the tube then carefully 
filled up with water, and the whole apparatus immersed overhead 
in a jar of water, in a few days or weeks (depending upon the 
length of the tube) particles of salt will arrive at the top of the tube 
and diffuse into the water atmosphere. 
When this condition arrives, which Fick calls “ dynamic equi- 
librium, 5 ’ diffusion takes place at a uniform rate, the mathematical 
expression of the process being stated as follows : — 
Let K denote the quantity of salt which in a normal state of 
diffusion passes in a unit of time through a unit of horizontal 
section of a cylindrical tube whose height is equal to the unit of 
length, this being called the diffusion coefficient ; also let Q be the 
quantity of salt which in the time t flows from the mouth of the 
tube ; S its horizontal section ; d the density of the liquid at the 
bottom ; and h the height of the tube ; then 
Q = Kc& 
h 
Experiments subsequent to those of Fick have caused some doubt 
as to whether the “ coefficient of diffusion” is the same for all 
