88 CORUNDUM^ ITS, OCCURRENCE AND DTSTRTBUTTON. 
found. The chroniite at all these localities shows the presence of a 
certain percentage of alumina and magnesia. Thus the small excess 
of magnesia unites with a definite amount of the alumina, but instead 
of separating out as spinel it separates out in the chromatic molecule, 
as previously mentioned. The greater part of the alumina separates 
out as corundum. 
5. In a number of these peridotite formations feldspar has been 
found, which is undoubtedly one of the original minerals of the rock 
and not a secondary product. At the Cullakeenee mine. Buck Creek, 
Clay County, N. C., there is at one of the border corundum veins a 
large amount of feldspar and hornblende which have separated out 
with the corundum. Again, at the Bad Creek mine. Sapphire, Jackson 
County, the corundum is found associated with feldspar and biotite 
mica. In both places they are lime-soda feldspar, and this occurrence 
of feldspar and corundum in a peridotite rock indicates that the molten 
magma contained some of the alkali and alkali-earth oxides, as NaJ), 
K^O, and CaO ; that a portion of the alumina united in the formation 
of the feldspar molecule, and that the rest of it separated out as 
corundum — these minerals separating out from the still molten 
magnesian magma. In a magma of this type there is present a large 
amount of magnesia, which forms the magnesian silicates, but 
apparently has no tendency to unite with the alumina to form the 
spinel molecule; but, on the other hand, the small amount of the 
alkalies and alkali-earth oxides do unite with a definite amount of the 
alumina to form the feldspar molecule. This, again, would seem to 
indicate that in the presence of a certain amount of the alkali and 
alkali-earth oxides and silica, provided there is enough silica present 
to unite with both these oxides and the magnesia, there is no tendency 
for the magnesia to unite with the alumina to form any type of mag- 
nesia-alumina minerals. That the feldspar in these peridotites is a 
contemporaneous mineral with the corundum and not due to its 
alteration is clearly demonstrated by the examination of thin sections 
of the rock under the microscope, which shows the sharp, angular 
masses and fragments of corundum embedded in tlie feldspar, and 
also the intimate mixture of the feldspar and the corundum with large 
porphyritic feldspar crystals in a groundmass of the finer-grained 
olivine. Dr. G. H. Williams " has investigated a feldspathic perido- 
tite of eastern Maryland, and his investigations are in accord with the 
conclusions drawn above — that the feldspar is older than the olivine 
and separated out from the still molten magnesian magma. He 
says: ^ " The structure of these rocks is therefore quite an exceptional 
one for the family of peridotites; first, on account of the porphyritic 
« Bull. U. S. Geol. Survey No. 28. " Ibid., p. 53. 
