958 THE JOURNAL OF GEOLOGY. 
The grorudite-tinguaite series traverses the diagram in an 
almost straight line, which rises steeply from the most siliceous 
end to the least siliceous, the alkali-silica ratio ranging from 
.09 to .24. The location of the so-called Sussexite, from Beemers- 
ville, Sussex county, N. J., which Brégger considers equivalent 
to a possible extreme form of differentiation belonging to this 
series, occurs in the diagram in direct line with the trend of the 
grorudite-tinguaite analyses. The alkali-silica ratio being .32. 
The few analyses of the differentiation facies of the igneous 
cores at Square Butte and Yogo Peak are merely indications of 
the trend of these series. In fact at Square Butte, there are 
simply the two end results of highly advanced differentiation. 
The trend of the line connecting these extremes is almost at 
right angles to that of the grorudite-tinguaite series. The trend 
of the line connecting the three parts of the Yogo Peak mass is 
almost horizontal. The two analyses of shonkinite lie close together, 
but those of the complementary syenitic rocks are separated 
from one another by considerable space. Assuming that the 
sodalite-syenite and absarokites described by Professor Merrill 
are complementary rocks, as their association in the field and 
their analogy with the rocks of Square Butte indicate, we find 
a line connecting their possible series would be nearly vertical, 
as in the diagram. The analysis of the rocks of the absorakite- 
shoshonite-banakite series occupy a rather broad belt in the 
diagram, which lies between the extreme of the Yogo Peak and 
Square Butte series. The least siliceous absarokites lying very 
near shonkinite and the quartz-banakites lying between the 
syenites of these series. From the position of the analyses in 
the diagram it is evident that the absarokite-banakite series as 
here constituted is more comprehensive than either of the two 
above mentioned, which are in fact distinct series, since each is 
confined to a particular rock mass. But in the case of the dikes 
and lavas in the Yellowstone Park no such subdivision of these 
orthoclase-bearing, basic rocks could be established. Occurring 
as they do as a group of related lavas, having like geological 
relations, and analogous chemical and mineral compositions, they 
