10 BULLETIN 1122, U. S. DEPARTMENT OF AGRICULTURE. 
Ammonia gas was not absorbed at all by eight minerals and appre- 
ciable amounts were absorbed by only two, muscovite and limonite. 
Malachite green was absorbed in considerable amounts by muscovite, 
biotite, and chlorite, and to a lesser degree by tourmaline and several 
others. Four of the commonest soil minerals, on the other hand, did 
not absorb a measurable amount of the dye. Water vapor was 
absorbed by all minerals, although significant amounts were ab- 
sorbed by only five. The minerals which are usually present in soils 
in the greatest amount — quartz arid the feldspars — showed practi- 
cally no absorption for the dye, water, or ammonia. 
There appeared to be no correspondence between the amounts 
of malachite green, water, or ammonia absorbed by the different 
minerals. Rutile, for instance, was highly absorptive of water but 
not of malachite green or ammonia. On the other hand, tourmaline 
had a considerable absorption for malachite green, only a very slight 
absorption for water, and no absorption for ammonia. Limonite, 
which shows especially high absorption for water and ammonia, is 
usually considered as being a colloidal mineral. These results agree 
with Greenland's (11) conclusion that crystalline minerals as well as 
the gel minerals may show some absorption. 
While there is of course some proportionality between the outer 
surface exposed by any one mineral and its absorptive capacity, that 
is the finer any particular mineral is the more it will absorb, it is 
apparent from the above table that the absorptive capacities of the 
different minerals were not proportional to the outer surfaces ex- 
posed. Each mineral seems to have a characteristic absorptive 
capacity for the different substances. 
ABSORPTION BY THE MINERAL PARTICLES IN THE SOIL. 
With determinations available of the absorptive values of practi- 
cally all the important unaltered minerals found in normal soils, it is 
possible to make an approximate evaluation of the part such minerals 
have in the total absorption of a soil, provided the mineral com- 
position is known. ^ Thus, it is possible to calculate what is probably 
the truly noncoUoidal absorption of the fractions described under 
"^'Estimation of noncoUoidal absorption from soil fractions." With 
ithis in view, a quantitative mineralogicai examination was made of 
the ''fine fraction'' in each of the soils. The results were as follows: 
Huntington loam, soil, fine fraction, 50 per cent quartz, 2 per cent 
other crystalline minerals, and 48 per cent indeterminate colloid-like 
material. 
Huntington loam, subsoil, fine fraction, 75 per cent quartz, trace of 
'Other crystalline minerals, 25 per cent indeterminate colloid-like ma- 
terial. 
Sassafras silt loam, subsoil, fine fraction, 60 per cent quartz, 1 per 
cent muscovite, 1 per cent other minerals, 38 per cent indeterminate 
<}olloid-like material. 
Sharkey clay loam, soil, fine fraction, 50 per cent quartz, 5 per cent 
m.uscovite, 3 per cent other crystalline minerals, 42 per cent mdeter- 
minate colloid-Hke material. 
8 While optical methods are not applicable for the identification of minerals below 0.005 millimeter in 
jdiameter, crystalline outline can usually be distinguished in material down to about 0.001 milhmeter. 
