C. HE. Tilley—Para-Gneisses vn South Australia. 309 
The parameters correspond closely with those for the “ group 
mean” of the analyses cited by Grubenmann. 
The abundant minerals garnet and sillimanite serve as indices 
of the origin of the dominant potash felspar. The high-grade 
metamorphism of argillaceous material results usually in the forma- 
tion of potassic felspar from the more aluminous constituents as the 
micas, either sericite or biotite. It is obvious that the formation of 
new felspar from these constituents involves the setting free of an 
excess of alumina to be disposed of. In these rocks this has appeared 
either as garnet or sillimanite or more rarely spinel. 
Dealing first with garnet. This is essentially an almandine type 
in which the FeO constituent is dominant, as the analysis shows. 
Its formation can be represented by an equation of the type 
(1) 2H,K Al,(Si0,),) + 3810, => 2KAIS8i,0, + 2Fe,MgAl,(S10,)s 
Mg,Fe, (810,), -++2H,0 
biotite + quartz = orthoclase + garnet 
The excess alumina is here combined with the iron magnesia 
constituents in garnet. 
The validity of an equation of this nature is evident by the 
relationships of certain garnets as seen in thin slices, where its 
association with biotite and quartz is one of genetic significance. 
The reversibility of this equation is postulated at least for those 
cases where the degradation of garnet to biotite is to be observed, 
the potash being obviously derived from neighbouring orthoclase. 
In some of the rocks garnet occurs in association with green spinel, 
and this relationship is genetic (cf. Fig. 2 (b)). The equation for this 
reaction is of the type? 
(2) 2H, KAI,(Si0,), + | Si0, =2KAISi,0, + Fe,MgAl,(Si0,), ++ 
MgFe,(Si0,), FeAl,O, + 2H,O 
Those types in which quartz and felspar are the chief constituents 
and garnet and sillimanite very subordinate, can obviously only 
have derived that amount of felspar from biotite or sericite in 
accordance with the amount of alumina set free to be absorbed in 
garnet and sillimanite. The felspar in excess of this amount must 
be derived by the recrystallization of original felspar, and where 
this was large the original sediment was an arkose. 
_* In these equations the quantitative proportions of orthoclase to garnet, 
silimanite, and spinel will obviously depend on the ratio H: K in the mica 
involved. In this, and following equations, the mica must be regarded as 
mechanical mixtures of sericite, and biotite of the composition shown in 
equation (1). For simplicity the gross composition of these mixtures is shown. 
Whether Clarke’s view, or the hypothesis of Tschermak for the constitution 
of the micas is adopted, it is quite clear that the main types muscovite and 
biotite form mix crystals only to a very limited degree. In a recent paper in 
a study of a large number of analyses of the mica group, H. E. Boeke (‘* Die 
Grenzen der Mischkristallbildung in Muscovit und Biotit’’: Newes Jahrbuch, 
1916, pp. 83-117) shows that the compositions of muscovite and biotite are to 
be pieced in two distinct and separate fields, each of which is quite limited 
in extent. 
