274 W. R. BROWNE AND W. A. GREIG. 
hornblende and biotite. The present example would seem 
to show that olivine can crystallize out from magmas con- 
taining an appreciable proportion of free quartz, although 
of course the possibility must be considered that the 
magma may have been acidified by the gravitative settling 
of some olivine. 
The discrepancy between normative and modal quartz 
is rather puzzling. If the scheme of crystallization out- 
lined above is correct, then the norm, which represents 
crystallization under anhydrous conditions, should be poorer 
in quartz than the mode by the amount formed as a by- 
product in the crystallization of hornblende and biotite. 
But in fact the Rosiwal analysis shows something less than 
6 per cent. by weight of quartz, whereas the norm contains 
nearly 8 per cent. It may be that the quartz is not uni- 
formly distributed through the rock, and that the portions 
measured were poorer than the average in this mineral. 
The comparative lack of hornblende and biotite in the 
analysed monzonite-porphyry is probably to be regarded as 
due to its having cooled relatively quickly and without the 
concentration of water and other mineralizers which 
exercised such an important influence on the quartz- 
monzonite. 
The positions of apatite and iron ore in the crystallization 
order deserve some notice. These minerals are usually 
among the first to crystallize from igneous magmas, but 
in the present instance they appear almost entirely as 
inclusions in the later minerals, biotite, orthoclase and 
quartz, and seldom (in the case of apatite, never) in pyroxene 
or plagioclase. This is particularly noticeable in those 
rocks containing a groundmass, the apatite and most of the 
iron ore being concentrated in the spaces between the 
larger felspars, which are conspicuously free from inclusions. 
It may be that, as suggested by Hlsden for the apatite of 
