216 
MR. C. BARRINGTON BROWN AND PROFESSOR J. W. JUDD 
If this view of the origin of the gem-bearing limestones be accepted, we may 
possibly find means of accounting for the formation of the gems, corundum, and 
spinel themselves. 
It may he remarked at the outset of our enquiry that the association of corundum 
and spinel, with such varied minerals in different cases, points to the conclusion that 
these gems have not always owed their origin to the same set of causes. 
In the great belt of corundiferous rocks in the eastern part of the United States, 
corundum has been shown to be constantly associated with ultra-basic rocks (peri- 
dotites), more or less completely converted into serpentine, and included in a series of 
highly metamorphic masses. 
But corundum is no less frequently found in association with more acid rocks. 
Near Mozzo, in Piedmont, corundum occurs in a felspar-rock, in the Riesengebirge in 
granite, and in the Zanskir mountains of Cashmere and many other localities in 
gneissic rocks. 
At Pipra, in S. Rewah, and at other points in Central India, masses of corundum- 
rock of enormous thickness and extent have been found interfoliated with the 
crystalline gneisses and schists.* 
Corundums are found, probably produced by contact metamorphism, in basaltic 
lavas (Unkel-on-Rhine, and especially near Le Puy, &c.), and in blocks ejected from 
volcanoes (Laacher See, Niedermendig, Konigswinter, &c.). 
In limestones, corundum occurs usually in association with tourmaline and chon- 
drodite, as at Orange Co., N.Y., and Sussex Co., Jersey, St. Gothard, Naxos, and 
other localities ; and at other times, as in Burma, in limestones without the minerals 
containing boric acid and fluorine. 
There are two other modes of occurrence of corundum which are of especial 
interest. Henri Ste.-Claire Deville showed!' that when the natural hydrated 
aluminium oxide of Baux (bauxite) is fused with caustic soda, digested with water, 
and then treated with nitric acid, a few grains of considerable hardness are left 
behind. These hard grains resist the action of acids, but dissolve in fused potassium 
bisulphate. On analysis they were found to be aluminium oxide containing traces 
only of iron, titanium, and vanadium. The hard grains were by these tests proved 
to be corundum. 
M. Moissan has shown that the remarkable iron-masses of Ovifak, Disco Island, 
Greenland, contain disseminated grains of corundum sapphire.j 
Spinel probably occurs in all, or nearly all, the different associations in which we 
find corundum, and, indeed, the association of the two minerals with one another is of 
frequent occurrence, 
* See ‘Mining Mag.,’ vol. 11 (1895), p. 57 ; and ‘ Records of GeoL Survey of India,’ vol. 5, p. 20, 
and vol. 6, p. 43. 
t ‘ Ann. de China. et de Phys.,’ vol. 61, p. 309; ‘ Chemical News,’ vol. 4 (1861), p. 341. 
J ‘ Compfc. Rend.,’ 116, p. 269. 
