112 Chester and Cairns — Crocidolite 



A comparison of these figures with those given for abriach- 

 anite forces us to the conclusion that the two analyses are made 

 on essentially the same substance, and that Heddle's mineral is 

 simply a magnesian variety of crocidolite. The name abriach- 

 anite, if retained at all, maybe kept as the name of this variety. 

 In this connection it is proper to notice the crocidolite-like 

 mineral from Mexico described b}^ Bauer,* and referred by him 

 to asbestos. The table below gives his analysis and the calcu- 

 lated ratios. 



1. 



Si0 2 55-48 -925 



Fe„(X 12-32 -077 



A1 2 3 2-01 -019 f '° 96 



CaO 10-35 -185] 



MgO 17-23 -431 I .,_„„ 



Na a O 1-54 -025 f ' ' l6 



H 2 1-47 -082J 



Here, although lime has taken the place of ferrous oxide, and 

 there is much less soda and water, yet the ratio of Si0 2 : R a O, '. 

 RO is practically the same as in the last-named variety. It 

 also agrees with crocidolite in its physical and pyrognos- 

 tic characters. It stands then as a connecting link between cro- 

 cidolite and "amphibole, and perhaps indicates the true rela- 

 tions of the former mineral. We certainly cannot agree with 

 the conclusions of Dolter,f afterward sustained by Kenngott,:}: 

 who considers crocidolite merely a fibrous variety of arfvedson- 

 ite, for to carry out this idea, Kenngott is obliged to throw the 

 water out of consideration as non-essential, and to make no ac- 

 count of about five per cent of the silica in Stromeyer's analy- 

 ses, and of nearly seven per cent in Dolter's, which, as he says 

 himself, "is difficult to explain." And the true nature of arf- 

 vedsonite can hardly be considered as. settled, for Lorenzen's 

 recent analysis,! with the formula deduced from it, is so totally 

 different from any before it as to render its further examination 

 quite necessary. 



In looking over the figures of the various analyses quoted 

 above, the widest variation is noticed in those given for water, 

 varying from 1*58 to 5*58. Supposing the analyses all to be 

 made on essentially the same mineral this may result from 

 either of two causes. In the first place, where it is high, part of 

 it may be non-essential, as is the case in Haddle's analysis, where 

 0*95 per cent went off at 100°, leaving 3-82 as that actually be- 

 longing to the mineral. Then, where the result is low it may 

 have been determined by the loss when heated. Such a de- 

 termination will always be too low on this mineral, for at the 



*Z. f. Kryst., iv, 40, 1879. f Neu. Jahrb. f. Min., hi, p. 163, 1885. 



JMin. Mag., v, 50, 1882. § Neu. Jahrb. Min.,i, p. 158, 1882. 



