J. W. Mallet, Esq., on the Analysis of Euclase. 105 



differ widely.* If, on the other hand, alumina and glucina be 

 isomorphous, the composition of euclase coincides with that 

 of andalusite. 



4 Al 2 3 3 Si 8 . 

 part of the alumina being replaced by glucina. An impor- 

 tant objection to the idea of any real connection between 

 these minerals, however, arises from the fact, that they occur 

 in different crystalline systems, andalusite belonging to the 

 right prismatic, while euclase is in the oblique prismatic sys- 

 tem. 



There was one minor point in connection with Berzelius's 

 analysis which it was interesting to examine with special 

 care, namely, the occurrence or not of a small quantity of tin 

 in euclase, and I, therefore, took particular pains in testing 

 all the reagents for this metal before using them, and made 

 a separate blowpipe experiment on the mineral itself, with 

 the object of reducing the tin directly. Even by the latter 

 method there was no difficulty in distinctly ascertaining its 

 presence, and there can, therefore, be no doubt of its really 

 existing in the pure mineral. 



The occurrence of traces of this metal in other silicates, 

 as beryl, epidote, and a magnesian garnet, meteoric stones, 

 and in several ores of titanium and tantalum, has been re- 

 marked by different analysts, especially by Berzelius, and is 

 certainly a very curious fact, when we consider the extremely 

 small number of minerals in which tin forms a leading con- 

 stituent, and the improbability of such minute quantities 

 being essential to the composition of the species in which 

 they occur. — (Journal of the Geological Society of Dublin, 

 vol. v. part iii., 1853, p. 206.) 



* The angles of crystals of the two species also differ considerably. 



