120 REGINALD A. DALY 
else the immense crops of olivine, pyroxene, magnetite, etc., were 
dissolved in still deeper, very hot magma which may still be liquid. 
If it is this magma that has been erupted in post-Cambrian time as 
basalt, then the original magma was not typically basaltic but more 
salic. On the other hand, there is no field evidence of the assumed 
generation of ultra-femic differentiates below the pre-Cambrian 
granite terranes, even if some special mechanism existed whereby 
the sunken crystals of this early differentiation were arrested at 
levels above the source or sources of the primitive basalt. Since 
peridotites, most probably derived from the relatively small volumes 
of basaltic magma involved in post-Cambrian igneous activity, 
have been erupted from time to time, it is reasonable to suppose 
that much more numerous and larger bodies of peridotites should 
have been erupted during the pre-Cambrian, if the pre-Cambrian 
granites were differentiated from basalt. The fact is that peri-. 
dotites are by no means conspicuous in the pre-Cambrian complexes. 
As noted elsewhere, the writer is more disposed to regard the 
basalt of the world as itself a primeval differentiate, the sunken 
part of an intermediate magma, of which the other risen part is 
the material now constituting the granitic terranes of the pre- 
Cambrian. 
Rapid transitions between phases of differentiated imjections.— 
Though in each differentiated sill or laccolith the salic and femic 
phases are generally separated by a layer of rock chemically inter- 
mediate between the two, this layer is often very thin. Thus, at 
Square Butte, Montana, the transitional rock between the syenite 
and shonkinite of the well-known laccolith is only “a few inches or 
a foot or so” in thickness. Pirsson describes the transition as 
“extremely abrupt.’ Considering the large scale of this dif- 
ferentiation and remembering the distribution of the sunken crystals 
in Bowen’s experiments, such abruptness of transition is hardly to 
be expected if the differentiation were due merely to fractional 
From Bowen’s account of the evolution (p. 40) it is not easy to see how the 
potash of basalt could be concentrated in the proportion seen in granite, assumed to 
represent the residual liquid of an initially basaltic magma, unless the soda of the 
granite were much more abundant than the potash; yet in average granite potash 
dominates over soda. 
2L. V. Pirsson, Bull. 237, U.S. Geol. Surv., 1905, p. 51. 
