pratt.] MODES OF OCCUREENCE. 33 
where dotted by the small grains of garnet, which rarely show good crystal form. 
The garnet occurs associated also with the plagioclase. 
The plagioclase occurs twinned according to the albite law only. In sections 
perpendicular to 010 the lamelke show extinction as great as 30 , and the plagio- 
clase is therefore rich in lime and as basic as labradorite, which it probably is. 
It shows strong evidence of shearing movement in the rock; it is often broken, 
exhibits rolling extinction, and the albite lamella? are curved and bent. It runs 
along the planes of schistosity between the feldspars and forms a mosaic of 
angular broken grains. 
Staurolite was found in rather broad, irregular grains, and rutile in small, 
irregular grains and well-crystallized prisms. 
Professor Pirsson has indicated that the character and structure of 
this rock, composed chiefly of amphibole, labradorite, and garnet, 
suggest most strongly thai it is a metamorphosed igneous rock of the 
gabbroicl family. During metamorphism the augite of the gabbro 
would be converted into the l)ro\vn hornblende; anyiron ore thai was 
present would be taken up by 1 lie hornblende and garnet. The 
rutile would have resulted from the titanic acid thai is a regular com- 
ponent of the iron ores in these gabbro or diabase rocks. Staurolite 
is a mineral thai would be rather naturally expected, as il is usually 
a mineral of metamorphism, and its natural home is in the schistose 
rocks. The feldspar lias suffered llie least (excepl the corundum) 
chemically, and shows only the shearing of dynamic processes. 
The corundum does nol occur in crystals, bul in small fragments 
and in elongated nodules, which are cracked and seamed and appear 
to have been drawn out by the shearing processes. The general 
character and shape of the fragments of corundum would indicate 
that they were original constituents of the igneous rock and were not 
formed during its metamorphism. 
The exact classification. of this rock is not easy, bin ii is probably 
an amphibole-schist. 
CORUNDUM IN GNEISS. 
Corundum has been found in North Carolina in the ordinary gneiss 
of the same belt of crystalline rocks in which the peridotites occur. 
A number of occurrences, part of which may be corundum in gneiss, 
are described under the head of mica-schist; some of the rock that 
was at first thought to be a gneiss is now known to be a quartz-schist. 
In the eastern pari of Clay County, N. C, on the southern slope of 
Cross Ridge of the Chunk} 7 Gal Mountains, just above Thumping 
Creek, corundum has been found in the gneiss at a number of points. 
The rocks are so covered with soil and decomposed rock that the 
exact relation of the corundum-bearing gneiss to the normal gneiss 
can not be determined, but from what can be seen the former appears 
to be in narrow bands cutting through the latter. In structure the 
gneiss is distinctly laminated and very fine grained, except the por- 
tions immediately surrounding the corundum, where its constituents 
Bull. 180—01 8 
