46 The American Geologist. January, 1905 
tlie areas measured are not d^ but quite different in the rock 
under consideration. 
The results of this brief study as to the efficiency of the 
methods employed cannot be offered as conclusive. They 
are such, however, as to emphasize the various phases of 
these methods and to indicate in a general way their lim- 
itations. 
Calculation of the mineral composition from the chem- 
ical analysis affords a means of classification of rocks and, 
in the case of holocrystalline rocks, of expressing more or 
less accurately their actual quantitative mineral constitution. 
Heavy solutions are a convenient and readily applicable 
means of separating the minerals of a rock and may be 
A'aluable in determining quantitative relations acording to the 
state of aggregation of these minerals and the range of spe- 
cific grvaity represented. The accuracy of the results will 
be directly proportional to the coarseness of grain and the 
readiness with which the minerals can be separated. . The 
recessary fineness of comminution and consequent loss as 
dust depend on the latter condition. The time required to 
make a separation of a rock bearing the average number of 
important minerals is but a small fraction of that needed for 
:i chemical analysis. 
Any one of the direct measurements with the micro- 
scope does not appear to fulfill even approximately the 
Ticcessary conditions for a statement of the complete quanti- 
tative composition of an igneous rock such as granite. They 
are easily applied, however, and may serve an excellent pur- 
pose in giving an idea of the comparative abundance of the 
<:onsiituents of a rock whether granitoid or porphyritic. But 
•as a basis for estimating the weight, or even volume compo- 
siticn. iVie present study has shown none of them to be all 
that could be desired. 
THE ORIGIN OF BITUMEN* 
By W. C Morgan, Instructor in Chemistry, U. C, Berkeley, Cal. 
From the early decades of the last century when the 
first chemical investigations were undertaken in this direc- 
•Reprinted from the California Journal of Technologry, Berkeley, Cal. 
