Gabbroid Rocks of Mi?inesota. — Winchell. 201 
those used for the chemical analysis, mounted in balsam pre- 
sented one grain nearly oriented, from which the measure of 
2d gave: 2E=^62° 24'. Thus a whole series of variations 
exist in this one rock. 
2E = i3«i6' 
2E = 22055' 
2E ^-- 48°33' 
2E = 4903' 
2E = 62'' 24 ' 
To this series might be added two measures from the py- 
roxene of thegabbro rocks near Duluth, which resulted: 
2E* = 56° 33 ' 
2Et = 67" about 
And it may be added that even higher values very possibly 
exist, which it has been impossible to measure. At the same 
time not a single instance has been found in which it was not 
possible to measure the angle when the section was exactly 
perpendicular to the bisectrix. It therefore seems probable 
that a continuous series of values exist from the smallest ob- 
served to the normal value (2E=iio° about.) 
The dispersion is weak with p>u about «g. So far as ob- 
served the maximum extinction angle in the vertical zone does 
not exceed 30°. Distinct twinning occurs rather uncommon- 
ly parallel to /^'(loo). The mineral often presents a peculiar 
texture, or rather two textures, which deserve mention. In 
one case the large areas divide themselves between crossed 
nicols into a number of irregular areas of slightly, but dis- 
tinctly different extinction angles. Often, but not always, a 
very slight difiference in color can be detected in the different 
areas, one being very pale green, and the other colorless, or 
pale pink. In this texture the cleavages show no perceptible 
change in direction in passing from one area to another, in- 
dicating that crystallographically the whole anhedron belongs 
to a single crystal, while optically it may be divided into half 
a dozen areas, undoubtedly of slightly variable composition. 
*See chapter VI, Orthoclase gabbro. 
tMeasured on a section of a diabase from near Duluth, kindly 
loaned the writer by Prof. A. Lacroix. of the Museum d'Histoire 
Naturelle, Paris. 
