PUBLICATIONS. 113 
through the earlier accumulations, and have been exposed to view by 
subsequent erosion. 
The gabbros in question are among the more recent rocks, though 
not the latest, and exhibit no trace of crushing, recryStallization or 
other signs of metamorphism. They appear to have remained in the 
condition in which they originally crystallized. Moreover, no great 
terrestrial disturbance has affected the region since the time of their 
eruption. ‘They are said to form sheets or sills varying from a few feet 
to many yards in thickness, each band consisting of many parallel layers 
of lighter and darker material, which correspond to the trend of the 
sheet itself. The component layers vary in thickness from mere paste- 
board-like lamine to beds a yard or more in thickness. ‘They are 
sometimes as parallel and regular as sedimentary deposits. But, traced 
along the strike, they are apt to vary in thickness and even to die out. 
Their appearance is quite like the banding of gneisses. 
The microscopical study of these rocks shows their mineral con- 
stituents to be like those of normal olivine-gabbros, and the differences 
between the light and dark bands to be due to differences in the rela- 
tive proportions of the minerals. 
The crystallization of the mass is continuous across the bands, the 
individual crystals interlocking in such a manner as to make it evident 
that all crystallized from a molten magma at approximately one time. 
The white bands in some cases consist almost entirely of labradorite, 
while the extreme dark bands are made up of augite and titaniferous 
magnetite. 
The cause of these differences in composition of alternate layers of 
gabbro is considered to be a differentiation of the magma previous to 
its intrusion among the rocks in which it consolidated. The banding 
is the result of the intrusion of a heterogeneous magma. The original 
shapes of the more or less differentiated masses that composed this 
heterogeneous magma are not known, their intrusion through fissures 
would produce a laminated arrangement. 
A comparison of these rocks with certain ancient gneisses is drawn, 
especially with the anorthosite rocks of Canada, and with certain ultra- 
basic portions of the Lewisian gneiss of Scotland; the similarity in 
their structures is pointed out, and a correspondence in their origin is 
suggested. 
The analogy between the banding in both these kinds of rocks 
strengthens the view now generally held by geologists, that the older 
