1887.] Mr J. J. Coleman on a New Diffusiometer. 377 
The following experiments were made by diffusing caustic potash 
of three different densities into reddened dilute hydrochloric acid, 
which latter in every case required T45 milligramme of potash 
(KHO) per c.c. to turn the colour from red to yellow. The reservoirs 
of each were about 60 millimetres diameter, and contained about 
200 c.c. of alkali solution, so that its strength was approximately 
constant during the diffusions. 
The diffusions were made in a chest of wood, of 2 cubic metre 
capacity, with hollow walls filled with dry sawdust, this being again 
placed in a room of nearly constant temperature. Similar precautions 
were also adopted with the experiments verifying Ficks’ law, already 
detailed in this paper. 
It will be seen from this table that the result of the experiments 
demonstrate to the eye a fundamental law of diffusion, common not 
only to material particles, but to the imponderable agents, heat 
and electricity. Sir W. Thomson has also pointed out to me that, 
according to theory, and supposing the coefficient of diffusion is 
not variable, the heights of the columns in Table I. should have 
been identical, provided the acid had been regulated of varying 
densities to correspond with the alkali. 
Further experiments detailed in Table II. were then made which 
confirm this anticipation,* or rather, which show that if there is 
any variation in the coefficient it must be small. 
Experiments were also made with caustic soda (hfaHO), which 
are recorded in Table III., and I have found that the apparatus can 
also be made available for measuring the diffusibility of acids by 
diffusing them into slightly ammoniacal water coloured yellow with 
methyl orange, f We are thus supplied with a new method of 
ascertaining the diffusibility of a large number of chemical com- 
pounds, and also a method of checking the accuracy of the burette 
method of determining diffusibility, which I described in the Phil. 
Mag. in January last. The instrument may be also constructed on 
a larger scale for lecture demonstrations. 
* The difference between the heights was reduced to 6 millimetres instead 
of 31. See also confirmatory experiments detailed in Table IV., added 
September 1887. 
t See Table IY. for details. 
