472 HENRY S. WASHINGTON 
Assuming that the figures given in the table above are 
roughly true (which is quite hypothetical), it will be of interest 
to calculate the composition of the magma as a whole. The 
results of this calculation, which are admittedly crude and of 
little reliability on account of the character of the data employed, 
are as follows: 
SiO; LiO,; Al, O7, He, On, He®@,) Mz @ i: Ca@t Na, Oy Ke © Ee @ rere: 
GijasiaeLs0 14.5 167 An Opp aneLiw7 2.00 (2 3574. Ol O14 —OOr 
The result corresponds in general with the idea of the magma 
derived from examination of the analyses, though it is perhaps 
somewhat higher in MgO and CaO, and hence more monzo- 
nitic, than we might have expected. Of the rocks analyzed it 
approaches most closely to that of the akerite (No. X), but it 
shows less SiO, and alkalies, and more MgO and CaO than this. 
A rock of this composition would probably be found among the 
more basic akerites or more acid diorites. 
Leaving this aside for the present it will be evident that the 
main course of differentiation (assuming that such has taken 
place), has been to form a large series of granites, quartz-syen- 
ites,and-diorites, which pass into one another more or less grad- 
ually through transition forms. Among these there is a quite 
gradual gradation of the oxides, as will be seen on reference to 
the table of analyses. A rather peculiar feature is the increase 
of Al,O, in the basic members, which is quite unusual. Parallel 
series from other regions analogous to this might be mentioned, 
but it seems scarcely worth while to do so. 
The series of granito-dioritic dikes, which correspond so 
closely both mineralogically and chemically to the granolites, 
must be classed as aschistic’, z. e., which are not separate differ- 
entiation forms of their magmas, but only dike forms of the 
partial magmas which solidified elsewhere as granolites. 
The foyaitic series presents a somewhat different problem. 
These rocks are evidently connected genetically with the main 
*BROGGER: of. cit., Vol. I, p. 125. I am in some uncertainty in thus rendering 
into English BROGGER’Ss words aschist and diaschist. The termination -ic would seem 
to be better than -ous or -ose, which latter is already in use in schistose, denoting 
structure. 
