The Hydrion Differentiation Theory of Geotropism 75 
is only applicable when the dispersity is low, for if r = 10 /x/x 
[o-oi fji] the time to fall o-i mm. [100 /x] is 7 hours.” In the above- 
mentioned report we also find under Stokes’ Law that, using this 
equation, “we have a ready means of determining accurately their 
size from diameters of 1 mm. to and beyond the microscopic limit,” 
and again (op. cit. p. 36) that this “ method cannot conveniently 
be used for submicrons of less than 20 /x/x [0*02 /x] diameter.” Also 
(loc. cit) certain ultra-microscopes “when arranged horizontally are 
well suited for the purpose of determining the radius of submicrons 
from their speed of settlement by the application of Stokes Law.” From 
these quotations it would appear that “settling” in colloidal solu¬ 
tions is not only well known, but is carefully measured by some of 
those who deal in a practical way with such material. 
In the second section of his article Professor Blackman raises 
several points which are best considered seriatim. (1) The time 
factor , with particles of radius 0-2 /x to o-8 /x, does not present serious 
difficulties (see below for calculations). (2) “Perrin.. .allowed three 
hours for the completion of the process.” This statement is brought 
forward to support the view that particles even of 0*2 /x radius 
would cream too slowly, but Professor Blackman on a later page 
emphasises the fact that stimulation continues for some time after 
its commencement. Now it is the beginning of stimulation which 
occurs soon after the organ is displaced, and therefore we should 
consider, not the time taken by the particles to reach a stable 
equilibrium (as they did in three hours with Perrin’s gamboge), but 
rather the time which elapses before the redistribution of the 
particles becomes apparent. This time is indicated by Perrin, who 
states 1 that when the uniform emulsion is placed in a cell 100 fx high 
“a few minutes suffice for the lower layers to become manifestly 
richer in granules than the upper layers.” The resulting stimula¬ 
tion, if any, would, therefore, begin in a few minutes and continue 
for about three hours, thus satisfying the requirements both of rapid 
and of continued excitation (see also below). (3) Perrin’s data with 
a medium of viscosity 125 times that of water are next quoted. 
From this one infers that the critic considers protoplasm to be of 
a similar viscosity, but on the following page he takes a viscosity 
twice that of water for the protoplasmic medium. This is a wide 
range, and more restricted estimates are fortunately available. 
1 Brownian Movement and Molecular Reality, by J. Perrin. Eng. trans. p. 41. 
London. 1910. 
