932 
Journal of Agricultural Research 
Vol. XXVIII, No. 9 
Table IV .—Heats wetting of extracted and of unextracted colloidal material in soils 
Soil type 
Deter¬ 
mined heat 
of wetting 
per gram 
of colloid 
extracted 
from soil 
Heat of 
wetting 
per gram 
of residue 
after 
extracting 
colloid 
Colloid in 
residue 
by micro¬ 
scopical 
determi¬ 
nation 
Calculated 
heat of 
wetting 
per gram 
of colloid 
left in 
residue 
# 
Houston clay soil. _ 
Calories 
11.6 
10.5 
5.3 
4.6 
Calories 
2.1 
1.2 
a 1.9 
4.8 
Per cent 
40 
32 
11 
97 
Calories 
5.2 
3.8 
17.3 
5.0 
Houston clay subsoil__ 
Durham loam subsoil______ 
Vega Baja clay soil_ 
» This fine fraction contains considerable amounts of mica which may be responsible for a significant 
amount of this heat value. 
The results given in Table IV indicate in three of the four cases studied a wide 
difference in the heat values of the extracted and unextracted colloids. These 
results are in accord with the data obtained in adsorption studies. It thus 
appears that in some soils, the part of the colloidal material which is difficult to 
isolate may have a very different heat of wetting from the part which is readily 
extracted. This fact tends to bring about more or less inaccuracy in estimating 
the colloidal content of a soil from the ratio of heats of wetting of soil and sample 
of colloid. 
Evidence has been obtained that in the process of isolating and concentrating 
the colloidal material the adsorptive capacity of the colloid may be more or less 
altered. Apparently the heat of wetting property may also be subject to 
alteration. 
Since practically all of the heat of wetting of a soil is due to the colloidal 
material, the ratio, 
Heat of wetting of soil 
Heat of wetting of colloid * 
should show the true quantity of colloid in the soil when the ratio is based on a 
fair sample of all the colloidal material present, unless the colloidal material 
after extraction has a different heat of wetting capacity from what it has in the 
untreated soil. By very thorough work, I. A. Denison, of this bureau, has 
isolated 76 to 88 per cent of the total colloidal material present in three soils. 
Owing to the completeness of the extraction, the colloidal material must be very 
nearly representative of all the colloid in the soils. Nevertheless, Table V shows 
that when heats of wetting of these samples are used in the ratio, quantities of 
colloidal material are indicated which are considerably in excess of true values. 
The true quantity of colloid in the soil is taken as the quantity actually isolated 
plus the quantity of unextracted colloid left in the residues estimated micro¬ 
scopically. 
Table V gives the quantity of colloid in the soil indicated by the ratio and the 
heat of wetting values from which the ratio is calculated. The quantities of col¬ 
loid extracted and true values for colloid content of the soil are also given in this 
Table. 
The differences between the quantities of colloid indicated by heat of wetting 
ratios and by the combined extraction and microscopical determinations can not 
reasonably be attributed to failure to obtain a representative sample of the 
colloidal material for the heat of wetting ratios. Practically all the colloidal 
material was extracted for a sample. Moreover, it has already been shown that 
the so-called “ unextraetable ” colloid usually has a lower heat of wetting capacity 
than the extracted material. Hence, if all the colloid had been extracted, to 
produce perfect samples, the heat of wetting ratios would presumably have been 
