644 FRANK F. GROUT 
gabbro minerals are present and the composition of minerals is sur- 
prisingly uniform throughout. When any mineral differs from the 
average the variation is not visibly related to the position or to the 
associated minerals. 
A second type of variation is that from the general gabbro type 
to the granophyr. This is a change in mineral composition as well 
as a change in the essential minerals present (see Fig. 12). In the. 
field this change comes with surprising abruptness, after the 
Per cent Orthoclase. 
Per cent Quartz 
Fic. 12.—Diagram showing the break between gabbro and red rock at Duluth 
when plotted on the basis of quartz and orthoclase content. 
monotony of slightly varying gabbro bands, with an extreme only 
occasionally. In a few feet after the reddish tinge of granophyr is 
seen in the interstices of the gabbro none of the gabbro minerals 
are visible in the rock. 
This double process is distinctly contrary to the idea of Bowen’s 
paper, and there are few suggestions of the sort in the literature. 
There are, indeed, suggestions of different types of differentiation. 
Bowen outlines two series from gabbro to alkaline types.t| Harker 
has contrasted regional and local action.2, Lane speaks of wet and 
dry differentiation.s However, there is evidence at Duluth of two 
*N. L. Bowen, “Later Stages in the Evolution of Igneous Rocks,” Jour. Geol., 
XXIII (1915), December Supplement, p. 77. 
2 A. Harker, ‘“‘Tertiary Igneous Rocks of Skye,” Mem. Geol. Survey of the United 
Kingdom (1904), p. 410. 
3 A. C. Lane, ‘Wet and Dry Differentiation,’ Tufts College Studies, II, 30. 
