278 Wisconsin Academy of Sciences, Arts and Letters. 
Had there been no appreciable retention of water about the 
sand grains the concentration of nitrate in the successive wash¬ 
ings would have followed the simple law of dilution, changing,, 
as in the table, from 43.45 to .0000002 instead of from 35.75 
parts per million in the first 100 cubic centimeters to .11 parts, 
in the last washing, as was observed. 
Observed and Computed Concentration of Nitrate in Successive 
Washings of Sand in Distilled Water. 
No. 
Water retained 
in sand. 
Observed 
Concentration. 
Computed 
Concentration. 
Grams. 
Parts per million of Solution. 
1. 
12.7 
35.75 
43.4551000 
2. 
13.2 
3.30 
4.8969000 
3.. 
13.1 
.451 
.5710100 
4 ... 
13.4 
.174 
.0661390 
5. 
13.05 
.138 
.0078153 
6. 
13.3 
.128 
.0009023 
7. 
13.5 
.110 
.0001059 
8. 
13.5 
.110 
.0000126 
9. 
13.5 
.110 
.0000015 
10.. 
13.4 
.110 
.0000002 
Were the exact effective diameter of the sand grains experi¬ 
mented with known it would be possible to compute from the 
amount of water which did not drain away the minimum thick¬ 
ness of the water film which each controlled. It is certain that 
this diameter lay somewhere between .08 and .15 millimeters. 
The mean amount of solution retained after each of the 10 
washings by the 50 grams of sand was 13.265 grams, the small¬ 
est amount retained in any one case being 12.7 and the largest 
13.5 grams. Using this amount of water, which is 26.53 per 
cent of the dry weight of the sand, and a specific gravity of 
2.65 for the sand grains, the minimum thickness of the water 
film would lie somewhere between .008 millimeter and .015 
millimeters, where the diameter of the sand grain is taken at 
.08 and .15 millimeters. But since 13.265 grams is the amount 
of water which the sand was able to retain against gravity it 
