King—The Suspension of Solids in Fluids. 287 
lected in 50 cubic centimeter separates, except the first two 
which were 25 cubic centimeters each and there was left be¬ 
hind, not passing the filter 80 cubic centimeters. Each of'these 
.separates was analyzed and it was found that while the original 
solution contained in 500 cubic centimeters 176.5 mgs. of K, 
the first 25 cubic centimeters coming through contained only 
77.4 mgs. per. 500 c. c., 100 mgs. less, and the 80 cubic centime¬ 
ters not passing the filter had increased in strength to 179.5 
mgs., or a gain of 1.7 per cent. In the case of calcium the re¬ 
sults stood: original, 137 mgs.; first 25 cubic centimeters, 55 
mgs.; not passing filter, 146 mgs.; or a gain of 6.5 per cent. For 
the magnesium the results stood: original, 164.6 mgs.; first 25 
cubic centimeters 79.6 mgs.; not passing filter, 168.2 mgs., or a 
gain of 2.1 per cent. In the case of nitric acid the results 
stood: original, 530 mgs.; first 25 cubic centimeters, 259.6 
mgs.; not passing filter 540 mgs., or a gain of 1.9 per cent. 
For sulphuric acid the results found were: original, 806 mgs.; 
first 25 cubic centimeters, 340 mgs., not passing the filter, 
862.5 mgs, or a gain of 7 per cent. 
Results similar to these have been repeatedly secured with 
potassium nitrate alone, with potassium sulphate and with 
sodium bicarbonate and with sodium chloride alone, although 
in the last case the concentration of sodium chloride outside 
of the filter was invariably very small. 
The simplest explanation of such differential movements of 
salts through porous media as these, and the retention of them 
outside the filter, appears to be to suppose that either the whole 
of the water, or some portion of it, is apportioned to the salts 
held in solution in such a manner as to place a molecule or 
group of molecules of each salt at the center of a sphere of 
water capable of moving as a unit and not fundamentally dif¬ 
ferent from what we have supposed in regard to the suspension 
of silt, unless difference in magnitude be regarded as such. With 
such a compounding of water and salt there would be less free¬ 
dom to movement of the salt through the filter and a separation 
of the salt and water, more or less complete, would necessarily 
result. Such a compounding, too, would be qualitatively, at 
least, in harmony with the elevation of the boiling point and 
