16 BULLETIN 1122, U. S. DEPARTMENT OF AGRICULTURE. 
Table 5. — Absorption ofgels- 
Description of sample. 
Absorption per gram 
of gel. 
Mala- 
chite 
green. 
H2O. 
NH3. 
Silica gel: 
Freshly prepared 
Commercial product 
Aluminum hydrate gel: 
Freshly prepared 
Chemical Warfare Service product. . 
Ferric hydrate gel: 
Freshly prepared 
Chemical Warfare Service product. . 
Aluminum sihcate gel, freshly prepared 
Peat 
Peat, particles below 1 micron 
Average values 
Gram. Gram. 
0. 0348 0. 8658 
.0080 I .2015 
.0889 i .6155 
.0735 .3500 
.0920 ! .3090 
.1160 ; .3106 
.5334 I 1.0888 
.2736 I .2907 
.0982 ! .2990 
Gram. 
0.1426 
.0525 
.0478 
.0504 
.1274 
.0563 
. 1443 . 4812 
.0736 
The different gels, just as the mineral powders, varied widely in 
their absorptive capacities for malachite green, water, and ammonia. 
It is significant that on the whole the absorptions of the gels were 
much greater than those of minerals ground below 1 micron, and 
were in a class with those of the ultra clays. The average absorption 
of all the gels w^as somew^hat below^ that of the ultra-clay average for 
malachite green, but higher for water and ammonia. 
A close agreement between the average absorptions of the gels and 
ultra clays is, of course, not to be expected, since the ultra clays would 
not be composed of equal amounts of each of these gels. Moreover, 
the absorptive capacities of sj^nthetic gels vary greatly, according to 
what appear slight differences in methods of preparation, and they are 
also profoundly affected by changes generally described as aging. 
Differences between the absorptive capacities of our gels and those of 
the chemical company and the Chemical Warfare Service are doubt- 
less due to differences in both method of preparation and age of 
product. 
The absorptive data on colloidal mineral powders and synthetic 
gels are, on the whole, in harmony with the view that the larger part 
of the colloidal material in soils is not made up of mineral particles 
but of substances in a gel-like condition. In itself the absorptive 
data is not conclusive evidence, but it is confirmatory of similar con- 
clusions arrived at in other investigations. 
GENERAL DISCUSSION. 
The data accumulated in this work indicate that, as a rule, con- 
siderably less than 5 per cent of the absorptive capacity of a soil is 
located in the unaltered mineral particles and that 95 per cent or 
more of the absorption is due to the colloidal material. It is realized 
that this distribution of the absorptive capacity of a soil, which was 
determined for malachite green, water, and ammonia gas, might not 
hold for all substances, although these three substances differ widely 
in their chemical and physical properties. 
If 95 per cent or more of the absorption of the whole soil is due 
to the colloidal constituents, the accuracy of absorptive methods for 
