48 
PENNSYLVANIA ANORTHOSITES 
expect to find, if not a felsic differentiate above the anorthosite, 
at least traces of a residuum either as interstitial material in the 
anorthosite, or in pegmatite dikes, intruding the chilled border. Mil¬ 
ler makes no reference to pegmatites; and while it is admitted that 
typical anorthosite is by no means a pure plagioclase rock, and that 
sodic and potassic feldspars and even quartz are sometimes found 
in small amounts, the chemical analysis of typical Adirondack anor¬ 
thosite indicates the presence of only .92 per cent K^O, and there 
is not even a trace of the volatile constituents common to nearly 
all igneous rocks. 
In the application of these hypotheses to the Honeybrook an¬ 
orthosite, the criteria which characterize Bowen’s and Miller’s 
hypotheses, respectively, may be given in outline. 
Bowen’s theory carries with it the following criteria: 
1. Anorthosite should be a coarse-grained labradorite rock of 
very nearly monomineralic composition by reason of its being a 
collection of plagioclase crystals which have accumulated through 
gravitative settling. 
2. For the same reasons anorthosite could never have been 
molten and should not, therefore, occur as dikes. The presence 
of protoclastic texture or granulation should be common to all 
moved anorthosites. 
3. Anorthosite should be associated with gabbro, pyroxenite 
and peridotite, and with syenite or other silicic rocks. • 
4. The silicic differentiate should overlie the anorthosite; it 
may or may not show intrusive relations with the anorthosite; 
there may be a transition facies between the two rocks. 
5. The gabbro, which represents the undifferentiated parent 
masrma should form a chilled-border facies around and above the 
differentiated series. 
Miller’s theory is based on the following criteria in the Adiron¬ 
dack anorthosite. 
1. Anorthosite is a coarse-grained labradorite rock of very 
nearly mono-mineralic composition by reason of its being the 
upper differentiate of a gabbroid magma from which the metasili¬ 
cates and orthosilicates have separated by gravitative settling. 
2. A gabbroic chilled border surrounding and overlying the 
anothosite marks the limits of the undifferentiated intrusion. 
3. Anorthosite may occur as dikes, since it was once to some 
