14 BULLETIN 1193, U. S. DEPARTMENT OF AGRICULTURE. 
CAUSES OF DISCORDANT RESULTS. 
POSSIBLE ERRORS IN THE METHOD. 
Considering the results as a whole, it seems apparent that the ad- 
sorption ratio of soil and sample of colloidal material must give at 
least an approximate idea of the colloidal content of a soil. The 
many close agreements between the ratios afforded by the adsorp- 
tions of such widely different substances as malachite green, water, 
and ammonia can hardly be conceived of as due to chance. ■ On the 
other hand, the occasional wide disagreements between the ratios 
calculated from the adsorptions of the different substances indicate 
that some important factor or factors influencing the relative adsorp- 
tions of the soils and colloids have not been taken into consideration 
in the methods employed. 
The changes which are supposed to take place in the colloidal matter 
of the soil on drying, moistening, freezing, etc., might, in the case of 
certain soils, have influenced the ratios. Beaumont (4) found that 
the water-adsorptive capacity of a soil was considerably reduced by 
drying at 110° C, and we found that the adsorptive capacities of 
several colloids for dye and water were reduced by about 7 per cent 
through heating to 110° C. Such effects may be very pronounced 
in freshly prepared colloids. 
In the case of freshly prepared colloids the phenomenon generally 
known as "aging" markedly influences the character and adsorptive 
capacity of the material (Van Bemmelen, 5; Freundlich and Schucht. 
12; Freundlich and Hase, 11). We found that the adsorptive 
capacities of silicic acid, " aluminum silicate," iron hydroxide, and 
aluminum hydroxide gels for malachite green, water, and ammonia 
varied greatly with the age of the material. In the case of soil 
colloids, however, aging, if operative at all, probably would not be 
marked inasmuch as the material must be in approximate equilibrium 
with the general soil conditions to which it is exposed. While we have 
not observed alterations in the adsorptive capacities of soils com- 
parable to those taking place in freshly prepared gels, a few soils were 
found which changed appreciably in their adsorptive capacity for 
malachite green when kept for several months in an air-dried con- 
dition. Many other soils tested remained sensibly constant in their 
adsorptive capacity when kept air-dried for more than a year. 
It is possible that some of the ratios given in Table 2 may be 
inaccurate as determinations of the percentage of colloidal material in 
the soil owing to unequal changes taking place, previously to testing, 
in the adsorptive capacities of the soil and the extracted colloidal 
material. This, however, could not be an error of very general occur- 
rence. Many duplicate determinations of the water-adsorption ratios 
made at wide intervals checked well. Miscellaneous data obtained 
indicate that if such changes do take place they are most likely to 
affect the adsorption of dye. 
There are, however, two general elements of uncertainty in measur- 
ing the colloidal content of the soil from the relative adsorptive 
capacities of the soil and sample of colloidal material. It is of course 
assumed in this method of calculation that the adsorptive capacity 
of the colloidal matter isolated from the soil is not altered during the 
process of isolation and preparation of the material. In other words, 
it is assumed that the sample of colloidal material as prepared has 
