PETROLOGICAL ABSTRACTS AND REVIEWS 273 
recalculation to 1oo and representation to the nearest whole number 
seems simpler to the reviewer, especially if the calculation is performed 
on a slide-rule. 
Pp. 68 to 161 are devoted to nearly 1,300 analyses recomputed to 
these four values; the chemical values used are taken from Osann’s 
former tables and are referred to by number. In Table I the rocks are 
arranged according to the decreasing S, increasing Al, and decreasing F 
values; in Table II according to the AICAIk values. Table III is 
arranged under the ordinary rock-names, the SAIF, etc. values being 
given as well as references to the literature. 
Of all the analyses computed, only two coincided in all four relation- 
ships, namely, the quartz-monzonite from Elkhorn, Montana, and an 
average of four analyses of the Butte ‘‘granite.”’ Since the two rocks are 
from the same batholith, the similarity is not surprising. 
To follow the discussions of the various relationships, it is necessary 
to look at the accompanying diagrams. No attempt, therefore, will be 
made to summarize them here, and the reader is referred to the original 
paper 
While the present work deals only with the igneous rocks, a number 
of examples of sedimentaries and crystalline schists are given for com- 
parison. A second part, dealing with sediments and schists is to follow. 
This work represents an enormous amount of patient labor. It 
should be of extremely great value for the visualization of chemical 
differences in rocks. 
Osann, A. “Uber topische Gesteinsparameter,” Sifzb. Heidel- 
berger Akad. Wiss. Math.-naturw. Kl., 1914, A 26, pp. 15, 
pls: 3, fig. i. 
Rocks of a petrographic province, or from a rock-mass showing 
zonal or other differentiation, or from any igneous body and its satellites, 
when plotted in a triangular diagram after Osann’s well-known system, 
show clearly their mutual relationships. If it is desired to show the 
general relationships of any igneous to the average igneous rock, it is 
only necessary to indicate the latter by a point properly placed. In this 
paper the mean of Clarke’s and Washington’s average rocks is taken 
and is recalculated in the Osann system. This gives a rock not far 
from the pyroxene-amphibole-biotite-diorite from Electric Peak. The 
latter, however, has more dioritic characters, the former more mon- 
zonitic. To show more clearly the variations of other rocks from this 
mean, it is shifted to the center of the triangle and the co-ordinates 
