1873.] i>Vi [Genth. 



always imbedded in marble or granular limestone when found in situ ; 

 the marble lies upon mica slate, gneiss or granite. The minerals associ- 

 ated with the emery and corundum are : diaspore, gibbsite, zinc-spinel, 

 pholerite, ephesite, margarite, mica (muscovite '?) chloritoid, black tour- 

 maline, chlorite, magnetite, hematite, limonite, rutile, ilmenite and titan- 

 iferous iron. 



Tbis represents very much the list of the species which occur with the 

 corundum and emery of our chromiferous serpentine region and it is not 

 improbable that the emery beds in Asia Minor and Greece are of analo- 

 gous age. 



I have now to speak of the occurrence of corundum or of minerals re- 

 sulting from its alteration in a series of slates considered by E. Emmons* 

 as belonging to his Taconic System. 



The first corundum in them was discovered in 1852, by Dr. C. L. Hun- 

 'ter,f in Gaston County ; it occurs at Crowder's Mountain and Clubb's 

 Mountain ; and it has lately been discovered under similar circumstances 

 in Rutherford County. It is associated with rutile, cyanite, damourite, 

 lazulite, pyrophyllite, gold and quartz. 



There are reasons to believe that the pyrophyllite beds in Orange, 

 Chatham, Moore and Montgomery Counties are analogous to the corun- 

 diferous strata of Gaston County ; and the same appears to be true for 

 those at Graves Mountain, Lincoln County, Georgia. 



There are two occurrences of conundrum in cyanite in the gneissoic 

 rocks at Swannanoa Gap, Buncombe County, and in Wilkes County, 

 North Carolina ; neither of which seem to have any connection with the 

 serpentine belt. 



I have yet to mention the occurrence of corundum in Laurens District, 

 South Carolina, which is probably in mica slates. All the specimens 

 which I have seen from this locality are rough bipyramidal crystals from 

 one to three inches in length and from one-fourth to one inch in thick- 

 ness. They are all more or less altered. Most of them are coated with 

 a micaceous mineral into which many of the crystals are completely con- 

 verted. 



I shall not dwell iipon the crystals of corundum, ruby and sap- 

 phire which are frequently met with in the heavy sands of gold washings 

 as they occur in too minute quantities to have any bearing upon the 

 principal questions which forms the subject of this investigation. 



After thus giving a brief outline of the geological features of the 

 occurrence of corundum, I shall now enter upon the consideration of the 

 minerals which are fonnd in association with the same, but will first make 

 a few remarks with reference to the chemical analyses which are em- 

 braced in this paper. 



The greatest pains were taken to select the very best and purest ma- 

 terial ; and all the small fragments, before being reduced to powder, were 



* Ebenezer Emmons, Geol. Report of the Midland Counties of North Carolina , ia56. 

 tC. L. Hunter— Sill. Journ . [2: X V , 373. 



