Genth.] ^64 [Sep. 19, 



to suggest, that the chromiferous and nickeliferous serpentines and talc 

 slates owe their existence to the decomposition of chrysolite rocks ;* an 

 opinion which has since been proved to be correct for those rocks wherever 

 they have beeu found ; and lately by very careful examinations of ser- 

 pentines and serpentine-like rocks made by Richard von Drasche.f 



C. U. Sbepard| endeavors to show that the real composition of these 

 rocks is not that of chrysolite but of villarsite. I have, therefore, re- 

 quested Mr. Th. M. Chatard to analyze a specimen from the Culsagee 

 Mine or Corundum Hill, near Franklin, Macon County, North Carolina, 

 and for comparison, add my old analysis of a variety from Webster, Jack - 

 son County, North Carolina. 



Chatard. Genth. 



Si0 2 =41.58 41.89 



A1 2 3 = 0.14 trace 



Feb = 7.49 7.39 



NiO (tr Co & Mn) = 0.34 0.33 



MgO = 49.28 49.13 



CaO = 0.11 0.0G 



Ignition = 1.73 0.82 



Chromic iron, etc = 0.58 



100.66 100.22 



Villarsite is only an intermediate product of decomposition between 

 chrysolite and serpentine. A rock, which approaches the latter mixed 

 with a triclinic feldspar, occurs under similar circumstances as the 

 Cullakenee or Buck Creek Mine in Clay County. 



An analysis of a specimen which, however, was not quite pure, gave 

 me : 



SiO a (by difference) = 35.19 



A1,0 3 = 0.64 



FeO = 9.70 



MgO =40.99 



Jo-nition = 13.48 



100.00 

 The outcrop of the Culsagee Mine extends over thirty acres and the 

 strata developed there are, according to Prof. Shepard (loc. cit.) : 

 " 1. Chrysolite rock, somewhat mixed with anthophyllite ; 



2. a layer of micaceous rock ; 



3. a seam of chalcedony ; 



4. a stratum of chloritic rock (ripidolite); 



5. the same through which the corundum is irregularly diffused, some- 

 times in narrow veins or widening out to several feet." 



*F. A. Genth— Sill. Journ. [2] XXXIII, 202. 



t R. von Drasche, in G. Tschermak's Mineralogische Mittheilungen 1871, 1. 



}C. U. Shepard-Sill. Journ. [3] IV, 112, 



