100 
AMERICAN GEOLOGY. 
globe where they occur. In this class of rocks there is no order 
of superposition. In some districts gneiss may occupy a posi¬ 
tion contiguous to granite; in other districts mica slate may 
occupy the same position relative to granite, and gneiss may 
appear in many instances overlying hornblende rock, talcose 
or mica slate. The Hoosick Mountain and the Appalachian 
ranges, furnish many examples of the varied and variable 
collocations of these rocks. They may, therefore, be said to 
belong to one epoch. 
It has been stated, already, that gneiss differs from granite 
in its structure. Its particles are structurally parallel. Fig. 
18 represents this structure. 
This parallelism has been 
regarded, by distinguished 
geologists, as due to water. 
Of this statement more evi¬ 
dence is required : indeed 
when we consider the extent 
of the change required to 
convert any sediment into gneiss or mica slate by heat, and 
over such wide areas, we can scarcely fail to feel that the origin 
assigned is at least extremely doubtful, especially when it is 
considered that the crust had already cooled so much as to 
admit of the condensation of water upon it, and the for¬ 
mation of oceans in all quarters of the globe. While the meta- 
morphic view is regarded as inapplicable to this class of rocks 
over wide-spread fields the world over, I do not call in ques¬ 
tion those instances of local metamorphisms which occur in the 
Alps and other districts, where it is evident the disturbances and. 
changes by heat are very remarkable. But who has observed 
in this country, sandstones, conglomerates, slates, and limestones, 
which have been converted into gneiss and mica slate, or horn¬ 
blende rock. It is true rocks are altered. Chalk has been 
changed into a hard crystalline marble, but the area over which 
this change can be traced, is quite limited. Clay, too, has been 
baked, and under that process has become hard, and firm enough 
Fig. 18. 
