and Shepard's Treatise on Mineralogy. 34 1 



I 



has shown, in the case of some of the titaniferous silicates, to the for- 

 mation of blue oxide by the loss of oxygen in titanic acid. Prof. 

 Webster has the subject in hand for a chemical analysis, and he ha 

 taken measures to procure a larger supply from the locality." 



Several recent analyses were made for the treatise in the course 

 of its progress, which were also sent by the analysts to Mr. Dana, 

 and were quoted in reviewing his work. 



We could have wished that Mr. Alger had used his judgment 

 more unsparingly in condemning species that are bad. We 

 run over his book, alluding to some errors which seem of suffi- 

 cient importance to demand at least a passing remark— especially 

 those that may convey abroad a wrong impression with regard 

 to American species. 



Cyprine, (p. 22.) This mineral is united by Mr. Alger with 

 garnet instead of idocrase, in consequence of an analysis by 

 Mr. Richardson,— evidence of no weight, inasmuch as garnet 

 and idocrase are admitted by chemists to have the same compo- 

 sition. The structure of Cyprine is identical with that of ido- 

 crase, from which it differs only in its light blue color. 



ithite, (p. 32.) Xanthite is retained as a species. We 

 have seen crystals the same in form with those of idocrase, and 

 re appears to be no doubt of their identity. 

 Fibrolile, (p. 108.) This mineral is united with kyanite. The 

 horolite of Count Bournon from the Carnatic has long been con- 

 sidered Bucholzite, with which it is probably identical. The 

 na me fibrolite was early applied in this country to a variety of 

 kyanite in short fibrous prisms, supposed at the time to be identi- 

 cal with Bournon's mineral. 



Davidsonite, (p. 121.) Rammelsberg states that this mineral, 

 according to the investigations of Breithaupt, Plattner and Lam- 

 Paaius, is nothing but beryl. Lampadius obtained for its c< 

 slt ion, silica 66-10, alumina 14-58, glucina 13-02, magnesia 1-16, 

 Peroxyd of iron 0-52, water 80.* 



Hudsonite, (p. 127.) The Hudsonite of Beck, here admitted 

 to the rank of a species, is identical with augite in structure, and 

 ,s near Hedenbergite in chemical characters. 



Thulite of Thomson, (p. 131.) This species, here retained, 

 n °ugh hesitatingly, is now considered a rose-colored epidote. 



Xa 



the 



mpo 



* Handworterbuch, i, 189. 





V °h xlvii, No. 2.— July-Sept. 1844. 44 



