444 A. P. COLEMAN 
and dichroism as the Methuen crystals, but not the hematite 
plates. 
As there is not much doubt that the Cobourg drift bowlders 
originated in the nepheline-syenite region to the northeast, they 
have been referred to here and may be considered in connection 
with the rocks previously described. 
In spite of the great variations in mineralogical composition 
to be seen in hand specimens all the rocks referred to have 
much in common; they are white to gray in color, generally 
schistose, often corundiferous, and present the same general 
habit, so that in field work they are naturally thrown together 
as nepheline-syenite and can be sharply distinguished from 
adjoining Laurentian gneisses, granites, and syenites. While 
not all of them contain corundum in large amounts they serve 
as a general guide to the discovery of the corundiferous rocks 
and are so used by prospectors for that mineral. Some of the 
ordinary syenites of the region, however, contain corundum also, 
and the largest crystals found occur in them. 
Just why the magma which has solidified into the groupof rocks 
described above should be so versatile in regard to mineralogical 
composition is not easily explained; but no other rock known 
to the writer shows so great a variety of types within short dis- 
tances as may be found in the nepheline-syenites. It may be 
that experiments such as those of Morozewicz' will give the clue 
to this variability, which seems to depend on the large proportion 
of alumina in the original magma. The corundiferous varieties 
of nepheline-syenite represent magmas supersaturated with 
alumina but not saturated with silica. 
A. P. COLEMAN. 
™See Review by T. A. JAGGAR: JOUR. GEOL., 1899, Vol. VII, No. 3, pp. 300, etc. 
