290 ASBESTOS AND ASBESTIFORM MINERALS— MERRILL. voL.xvin. 



If the foregoing is correct, it may seein, on first tliougUt, that we should 

 find asbestiform augites, enstatites,' and other members of the pyroxene 

 group. This does not necessarily follow, since these minerals, as is 

 well known, are peculiarly subject to alteration under conditions of 

 strain, giving rise to actinolitic, tremolitic, and talcose products. These 

 may or may not be asbestiform, according to local conditions. It is my 

 present belief that the asbestos form is never a result of original crys- 

 tallization, but is always secondary, the original mineral doubtless being 

 an orthorhombic or monoclinic pyroxene, or j)erhaps an amphibole. 

 The references made to the works of Blum, Heddle, Sandberger and 

 others, in the earlier parts of this paper, seem to point to this conclu- 

 sion. It is possible in such cases that the mineral derived from the 

 rhombic magnesian pyroxenes may take the form of anthophyllite, and 

 those from monoclinic lime-magnesian pyroxenes that of tremolite. 

 Such a rule can scarcely be considered as universal, since in many cases 

 the mineral undergoes more or less chemical as well as molecular alter- 

 ation under these conditions. The absence of appreciable quantities 

 of alumina in the asbestos proper is perhaps the strongest argument 

 against its derivation from augite or other aluminous pyroxenes, though 

 it is doubtless to such an origin that we can trace the uralites from 

 Nahant and Maiden. 



There is ample field here for further observation, and should this 

 paper be effective in causing collectors to note more carefully than 

 heretofore, not merely where the mineral occurs, but how it occurs and 

 with what associations, it will serve at least one good purpose. 



' Dana, on p. 389 of his " System of Mineralogy," latest edition, mentions the possi- 

 bility that " some asbestus may properly belong to the pyroxene grouii." It is 

 evident that, with the possible exception of the uralites from Maiden and Nahant, 

 none of the samples examined by the writer can be referred to the monoclinic 

 pyroxenes, though on strictly chemical grounds many of those called anthophyllite 

 might equally well be called enstatite. 



