l6o ' GEOLOGY: N. L. BOWEN Proc. N. A. S. 
manner and perhaps many other associations particularly where the 
differentiation is markedly discontinuous both chemically and spatially. 
Monomineralic types as members of composite intrusives. — Differentia- 
tion as a result of settling of crystals and as a result of deformation are 
by no means mutually exclusive; indeed, it seems probable that in one 
case, in particular, deformation may step in to bring to completion a task 
begun by gravitative accumulation, namely, in the case of the production 
of monomineralic types of excessive purity. For anorthosites, by way of 
example, it is probable that, besides the sorting of plagioclase crystals 
by gravitative action, a squeezing out of the interstitial liquid may have 
contributed to their extreme purity in many cases. If the squeezing con- 
tinued after a high degree of compaction of crystals had been obtained a 
clastic structure might be developed and the common protoclastic structure 
of anorthosite may perhaps be interpreted as evidence that a squeezing 
out of residual liquid has played an essential part in its production. 
Monomineralic types as simple ''intrusives." — Peridotite as a sill -like 
"intrusive" would appear to be capable of production by a somewhat 
similar combination of crystal settling and deformation. If the intrusive 
material was originally simply basaltic, if in early stages of crystalliza- 
tion the settling of olivine crystals to the floor was pronounced and if at 
this time deformation occurred such that in certain places the roof was 
down- warped the result might be that the whole width of the sill in these 
places would be occupied by peridotitic material. The down-warping 
of the roof at this early stage of crystallization might not only remove 
the supernatant liquid from above the accumulating olivine crystals but 
might be of sufficient vigor that the interstitial liquid of the peridotite 
should be largely expressed and a mass of typical dunite result. The 
mass of peridotite or dunite so formed would have all the ear-marks of 
an ordinary intrusive sheet and there would be no internal, textural evi- 
dence that it had not crystallized from a molten mass of sensibly its own 
composition though no peridotitic or dunitic liquid ever existed in that 
region. The formation of this peridotitic differentiate connotes the 
possibility, indeed the probability, of a complementary granitic differ- 
entiate which may be represented as one member of a composite sill in 
those parts of the sheet that have been widened during the deforming ac- 
tion. The association of peridotitic intrusives with composite intrusives 
of this type is rather common and merits examination on the part of field- 
workers with the above suggestion as to origin in mind. In the asbestos 
region of Quebec the peridotite (now serpentine) masses are described by 
those best acquainted with them in the field as synclinal sheets.^ This 
designation together with the association of diabases, diabase breccia, 
and granitic intrusives, suggests that the region may furnish a concrete 
example of the origin of a sheet-like peridotite "intrusive" after the manner 
described above. 
