35^ The American Geologist. December, i9(k> 
presses; itself in the mineralogical composition by the presence 
of hercynite. 
Origin. Several modes of formation suggest them- 
selves for a rock of this character. 
The hypothesis of its origin by means of a dififerentiation 
in the magma which gave birth to the gabbros and plagio- 
clasytes must apparently be rejected, since the silicoferrolytes 
and the plagioclasytes can not be regarded as the opposite 
poles of a similar series. It will be noted in this connection 
that there does not exist in the silicoferrolyte the abundance 
of calcium, of magnesia, and of titanium which ought to be 
found if this hypothesis were correct, and which exists in 
nearly all the analogousrocks whose analyses are cited above, 
and whose origin has been explained by the aid of this 
hypothesis by the authors who have studied them. Further- 
more, the existence of quartz would be very difBcult to ex- 
plain if one accepted this hypothesis. 
The same reasons render improbable the hypothesis ac- 
cording to which the silicoferrolyte would be the result of an 
intimate mixture of a gabbro and of a cjuartzose sedimentary 
rock, a cjuartzyte, for example. Indeed, an examination of the 
chemical composition of the normal gabbro and of the silico- 
ferrolyte (after deduction of about lo per cent of quartz) shows 
that no rock mixed with the gabbro in equal parts could give 
birth to the silicoferrolyte : and even if one admit that one part 
of the gabbro could digest three parts of a foreign rock, the 
latter would have to have had the following composition: 
iron oxides, 90%, alumina 6.5%, and water 3.5%, and even 
then the resultant rock would contain no free quartz, which 
microscopic study shows to be quite abundant in the silicofer- 
rolyte. 
Furthermore, the existence of numerous grains of quartz 
enclosed by the fayalite. and by the magnetite seems an in- 
surmountable argument against the hypothesis of the forma- 
tion of this rock b\' pure igneous fusion. Indeed, it is difificult to 
understand how free c[uartz could have remained intact in a 
molten magma whose consolidation would have given birth 
to an orthosilicate of iron and to a very large amount of mag- 
netite. 
There remains, therefore, only one hypothesis which con- 
