138 ANNALS NEW YORK ACADEMY OF SCIENCES 



(manganite) or dioxide (pyrolusite) ; b) as hydrate (pyrochroite) and car- 

 bonate (in wad, rhodochrosite, etc.) ; c) with siHca abundant, as manganese 

 hydrosilicate (bementite). 



Lime: a) as a residual remnant of one of the calcareous silicates (augite, 

 diopside, tremolite, anorthite, etc.); b) as carbonate (calcite) or calcium- 

 magnesium carbonate (dolomite, ankerite) ; c) with silica abundant, as one 

 of the newly formed calcium hydrosilicates (gyrolite, okenite, xonotlite), 

 or as aluminum-calcium hydrosilicate, (sloanite ?). 



Magnesia: a) with silica scanty, as oxide (periclase), hydrate (brucite), 

 ferro-magnesium hydrate (p\'Toaurite), aluminum-magnesium hydrate 

 (hydrotalcite), magnesium carbonate (magnesite, breunerite, mesitite) or 

 hydrocarbonate (hydromagnesite, hydrogiobertite, etc.); b) with silica 

 abundant, as one of two of the magnesium hydrosilicates (deweylite, 

 sepiolite) or as aluminum-magnesium hydrosilicate (pyrosclerite). 



Alkalies: with alumina and ferrous oxide (as a chlorite); with lime, 

 as a hydrosilicate (certain zeolites). 



It should here be noted that in a study of micro-aggregates of a different 

 origin, e. g., from development within a lower zone of metamorphism, a 

 quite different series of constituent minerals would need to be considered. 

 The preparation of any such series, in the present incomplete knowledge of 

 the conditions of origin of mineral species, would require careful investiga- 

 tion of associations, relationships and all other evidence at hand. One 

 conclusion from such study will be remarked in the series above given: 

 that many minerals, the occurrence of which in distinct and crystallized 

 specimens has been set down by the mineralogist as uncommon or even 

 very rare {e. g., brucite, periclase, deweylite, gyrolite, anthosiderite, etc.), 

 may yet be shown to occur abundantly, in dissemination through rock 

 formations and mineral aggregates in obscure or entirely invisible forms. 



In calculation of mineral constitution from the analyses of such micro- 

 aggregates, the chemical formulas of the constituent minerals, so far as they 

 have been determined with certainty, may be accepted and used as absolute, 

 and as far preferable in most cases to any actual analysis of a mineral, on 

 account of the universal intermixture of impurities in the latter, even in the 

 best crystallized and apparently purest specimen. 



The facts show, in my opinion,that all mineral substances have a definite 

 composition and character, that none are intermediate or transitional, that 

 even from decay or other mode of dissociation of a complex mineral com- 

 pound only independent minerals of simpler but exact composition are 

 derived. If this be true, we shall have little need of resort here to h}'po- 

 thetical chemical compounds but may perhaps rely entirely on determined 

 formulas for all calculations of mineral constitution. 



