IX. Summary: The common occurrence of the microcline form of potash-soda-felspar in pegmatites may be due to the crystallization temperature of the latter falling within the perthite exsolution range. At these relatively low temperatures the oblateness of the indicatrix is a maximum, and this, together with the fact that exsolution and crystallization of the two felspars are taking place simultaneously, provides the optimum conditions for the formation of microcline. The parallel orientation of the albite and the microcline twinning in coarsely perthitic microcline supports this view.

The derivation of the common potash-rich pegmatites from granite magmas cannot be satisfactorily explained by the existing binary or ternary crystallization diagrams. To overcome the various difficulties, the writer suggests that the residual granite magma has separated into two fractions, one rich in potash, which produced the ordinary pegmatites, the other rich in soda, which produces the complex pegmatites, and possibly much of the myrmekite of granites. This suggestion involves a modification of the binary diagram for temperatures below 750–800° C., and accounts for the following facts: The potash-rich character of the ordinary pegmatites, and their preponderance over the complex albite-rich pegmatites. The final graphic cotectic of quartz and microcline, or alternatively, of quartz and oligoclase. The absence of the ternary graphic cotectic of quartz, microcline, and plagioclase from these low-temperature crystallizations. The quartz-oligoclase cotectic composition of the myrmekite of granites, on the assumption that some part of the albite-rich immiscible magma has been retained in the crystal mush. A last-stage albite-rich magma which gave rise to the complex pegmatites.

This view of two immiscible magmas derived from a common granite magma has been put forward by other workers to explain special field problems.

Microperthitic exsolution causes an increase in density. Very high pressure would tend to promote exsolution and so modify the exsolution curve, and indirectly, the binary cotectic crystallization diagram, by increasing the immiscibility gap. Potash-soda-felspars which have crystallized from granitic magmas under very high pressures should thus tend to have a lower soda-content than the normal, and show the orthoclase rather than the microcline form.

The occurrence of potash-soda porphyroblasts in xenoliths and similar metamorphosed rocks under conditions which preclude an origin by direct crystallization from a magma, raises the question as to how far it is justifiable to regard holocrystalline rocks, even of typical granitic appearance, as products of direct magmatic crystallization. The same process which gave rise to these porphyroblasts, if continued for longer periods or at elevated temperatures, might produce holocrystalline granitic rock structures without the mass having even approached the liquid state.

For such rocks the ordinary solidus-liquidus diagram would have no meaning, but even under these conditions, crystallization of the potash-soda-felspars would still conform to the exsolution curve, whose immiscibility gap would depend on the particular formation temperature. A consideration of the composition and degree of heterogeneity of the contained potash-soda-felspars might provide useful information about the temperature and condition of formation.