142 



CHABAZITE. 



(Mineralogy of New- York, page 353.) 

 This mineral has been found associated with mezotype, in fissures in 

 the gneiss, 23d street, N. Y. At Haerlem it also occurs in yellow or 

 brownish crystals with stilbite and heulandite. 



EPIDOTE. 



(Mineralogy of New- York, page 354.) 

 To the particular localities heretofore given, I have to add the occur- 

 rence of fine, though small crystals, in the gneiss near the old peniten- 

 tiary, in 23d street, New- York. 



TOURMALINE. 

 (Mineralogy of New- York, page 356.) 

 Crystals with very short prisms, generally resembling fig. 360 of the 

 Mineralogy of New-York, but variously modified, have been found in 

 the white limestone, near Amity, in Orange county. 



Green and brown crystals of tourmaline are credited to Chester, in 

 Warren county. Rep. on the Geol. 2d Dist., p. 64. 



BUCHOLZITE. 

 (Mineralogy of New- York, page 364.) 



In my remarks upon this mineral I stated, that its identity with silli- 

 manite, wihich was urged by some mineralogists, was inconsistent with 

 the difference in chemical composition. This arose chiefly from the 

 fact that Muir, under the direction of Dr. T. Thomson, found sillimanite 

 to contain a large per centage of zirconia. But the early analysis of 

 Bowen and the more recent ones of Prof. Connell, Prof. Norton and 

 Erdmann, failed to detect the presence of this earth. The composition 

 of a specimen of sillimanite from Chester, Penn., as given by Erd- 

 man, is, silica, 40*08; alumina, 58*8S; protoxide of manganese, 

 0*74, which approaches to that of andalusite and gives the same for- 

 mula. 



According to a more recent notice of Dr. T. Thomson, sillimanite 

 and fibrolite are identical with bucholzite. The species should be dis- 

 tinguished by the name of silicate of alumina. (Phil. Mag. and Ann., 

 xxvi, p. 536.) Rammelsberg has also shown the identity of bucholzite 

 with xenolite of Nordenskiold. 



