E. W. Hilgard— Objects and Interpretation of Soil Analyses. 189 
ment, it is not in itself very instructive, as it leaves the relative 
amounts of the two substances altogether indefinite. eter- 
mination of the organic matter b combustion, or by extraction 
deau, the aid of which at least a uniform minimum determin- 
ation becomes possible. 
I have not devised any method for the direct determination 
of the water of hydration, although there are cases in which it 
would be very desirable to have this item, for the determina- 
tion of the condition of the alumina and ferric oxide. 
I have in a few cases determined the amount of szlica — 
in boiling solution of sodic carbonate in the crude soi 
this determination is often beset with almost insuperable me- 
chanical difficulties, from the diffusion of the clay in the alka- 
line liquid. It does not appear to promise results of sufficient 
importance to justify such labor; the more, as by the method 
of Grandeau, the actual available amount of silica can probably 
be better determined. But I have found the determination of the 
silica soluble in the alkaline carbonates, in the “insoluble residue” 
of the acid extraction, of very great interest. Evidently, in so 
far as it is derived from the decomposition of clay, “ kaolinite,” 
it should stand in a definite ratio to the alumina dissolved by 
acid, and this is often very aha the case. But were 
fact, Peace accounts for a great deal of the otherwise in- 
comprehensible variations in the properties of soils and certain 
clays, which I shall hereafter discuss. should also mention 
in this connection that I have strong evidence of the presence 
of still another hydrous ey — to saponite, in some o 
the tertiary “ prairie soils’ f the Southern States; the peculi- 
arities of which, when ies cultivation, have seemed unin- 
telligible. 
T have not yet been able to extend the method of Grandeau 
for humus extractions over a sufficient number of widely dif- 
ferent soils of well known characteristics, to consider the claim 
of its furnishing a definite measure of the available plant-food 
in the soil, as definitely established. But thus far I have found 
nothing to contradict this probable assumption, and much tend- 
Ing to its confirmation; and I hope to be able to continue the 
