A TYPE OF IGNEOUS DIFFERENTIATION 627 
THE ROCKS OF DULUTH 
The Duluth gabbro and its differentiates are intrusive into 
Middle Keweenawan flows, probably at a considerable depth. At 
Duluth it is evident that a feldspathic gabbro was intruded and 
cooled before the main mass of more basic gabbro was intruded. 
The mass is so large that it must have required thousands of years 
to cool, plenty of time for its differentiation into the many types 
now found. Some differentiates assume an intrusive relation to 
those earlier to crystallize, the most conspicuous case being the 
intrusion of ‘“‘red rock,” or granophyr, into gabbro. When studied 
in detail the mass shows by the intimacy of the geologic connection 
that, with all its variety, it is essentially a single geologic unit. 
Not only is the red rock related to the gabbro by association and 
intermediate phases, but over most of the area the two portions 
of the gabbro cannot be distinguished. Even at Duluth the two 
masses are not everywhere distinguishable. The averages differ 
only about 10 per cent in the amount of feldspar. It is believed 
that no great error will be introduced if the whole mass of data 
on the Duluth gabbro is considered as a unit. The main gabbro 
at Duluth and in many other outcrops is conspicuously banded.* 
The form has been named a lopolith? (Fig. 1). 
The descriptive petrography of the formation need not be 
rewritten here in detail. However, this study adds a few points 
from the type locality, where the exposures are exceptionally clear. 
The new data also make it possible to present a consistent, though 
not at all final, summary of the petrography of the whole mass. 
THE GABBRO 
The diagrams of the modes from measurements (Fig. 2) and the 
norms from analyses (Figs. 3 and 4) show the variation in the two 
gabbro masses at Duluth, without regard to position in the mass. 
It is evident that no simple linear series could be arranged on the 
' Frank F. Grout, “‘Internal Structures of Igneous. Rocks,” Jour. Geol., XXVI 
(1918), 430. 
2 Frank F. Grout, ‘‘The Lopolith,” Am. Jour. Science, XLVI (1918), 516. 
3 The references and a correlation of the varying nomenclature are given by A. N. 
Winchell in U.S. Geol. Sur. Mon. 52, pp. 395-407. 
