Blasdale.] 



Contribution to Mineralogy. 



339 



The results of the analyses of the two specimens are given in the 

 first two columns of Table III; for comparison five analyses of glau- 

 cophane, four from the type locality of Syra, and one from Zermat, 

 also the analysis of crossite, are included in the same table. In 

 both of the analyses here reported the close relationship of the 

 mineral to typical glaucophane is apparent. The only features 

 worthy of especial note are the relatively large amounts of combined 

 water and titanium. The latter element is easily accounted for in 

 the first analysis by its inclusions; probably these were also present 

 in the second specimen but escaped detection. Apparently the 

 water is due to hydrogen which displaces the alkali metals to a 

 certain extent as in the actinolite. Even larger percentages of 

 water (namely four) are reported by Barrois.* Compared with the 

 analysis of crossite, striking differences of the sesquioxide content 

 are worthy of note. From a chemical standpoint the most impor- 

 tant difference between the two minerals consists in the substitution 

 of nearly equivalent amounts of ferric oxide for alumina. Presum- 

 ably this difference is directly connected with the difference in the 

 optical orientation, and though the data is too incomplete to be con- 

 clusive, there are indications that there is a direct connection 

 between the position of the acute bisectrix and the relative amounts 

 of ferric oxide and alumina. In this connection, analysis No. 8, 

 representing a variety of glaucophane from Syra (Rhodus), is of 

 especial interest; the difference in the relative percentage of the 

 sesquioxides is here even more strongly accentuated, but, unfortu- 

 nately, no account of the optical properties of this variety appear to 

 have been published. 



In Table IV are given the calculations leading to the formula 

 of the mineral. It will be noted that the ratio of oxygen to 

 silicon is in both analyses slightly greater than corresponds to a 

 metasilicate. If, as in the actinolite, we represent the diad metals 

 by R", the elements H, Na, K, AlO, Fe'"0 by R', and by X a 

 mixture of Si 3 4 and Si0 4 in nearly equivalent proportions, but 

 with a slight excess of Si0 4 , the composition of the two samples 

 would correspond to the formulas R // i 2 -« X 12 . 5 R / 18 . 47 and R' / 13 . 35 



Barrois, Comptes Rendu, Tome 103, p. 221. 



