286 Wisconsin Academy of Sciences, Arts and Letters. 
and farther than the color ingredients held in solution or in 
suspension, so that often there is an outer colorless, or nearly 
colorless, zone. Goppelsroeder has written a volume on this 
and related subjects, pointing out how the principle may be 
used in separating the ingredients of many complex organic 
compounds for analysis. I have here a blotter upon which has 
been dropped at one point mono-chroic writing ink and at 
another ordinary carmine ink, which serve to illustrate the 
phenomena in question. It is clear that we have here a dif¬ 
ferential capillary movement in which the solvent travels more 
rapidly than the other ingredients. The simplest explanation 
of the phenomena is to regard the several colored components 
of the ink disseminated throughout the solvent, but with each 
color particle surrounded by a layer of the solvent which it 
controls in the manner supposed with the silt in turbid water 
and dust particles in the air. As the solvent spreads by cap¬ 
illarity and invests each paper fiber in an envelope of fluid to 
the extent of entirely filling the pore space of the paper, 
through this entangled fluid the several color ingredients, loaded 
with their film of solvent, and thereby being relatively large, 
are floated and dragged forward in the currents set up. They 
are never able, however, to outrun the solvent and the smallest 
color particles are likely to travel faster and farther than the 
larger ones, thus bringing about the colored zones which appear, 
as the result of the separation. 
Ordinary water solutions, such as nitrates, sulphates and 
phosphates, appear to behave similarly in passing through 
porous media to the extent that the first portion escaping from 
the filter may be very much less concentrated than the original. 
Hor is this all. The residual portion of the solution not passing 
through the filter may become more concentrated than the orig¬ 
inal solution, apparently on account of the retention of some 
of the salts entirely on the outside of the filter. Taking a spe¬ 
cific case in which 500 cubic centimeters of a composite solu¬ 
tion containing K 2 S0 4 , MgS0 4 , Ca (H0 3 ) 2 and CaHP0 4 was 
slowly passed through a Pasteur-Chamberlin filter under a con¬ 
stant but gentle pressure. The solution as it escaped was col- 
