Dr. J. E. Reynolds on certain Mineral Silicates. 287 



portant felspars of the volcanic rocks, just as albite and ortho- 

 clase are of the granitic rocks. Of these four (and other inter- 

 mediate compounds), however, leucite and orthoclase are the 

 most directly comparable, being both potassium and aluminium 

 felspars, evidently formed on precisely the same type, yet differ- 

 ing by the compound molecule SiO 6 . It now becomes a matter 

 of much interest to attempt an explanation of this remarkable 

 and yet methodic variation in the aciditic portion of these felspars. 

 With the assistance derived from some known properties of the 

 silicic hydrates, and the Rev. Professor Haughton's valuable 

 translation of Durocher's memoir on Comparative Petrology*, I 

 think it is possible to understand the conditions under which 

 one or other of these felspars can be formed, and the mode of 

 production of either. We are already well aware of the fact that 

 leucite is convertible into orthoclase by removal of potash and 

 alumina by aqueous action ; but this information gives little aid 

 in the direction we require. 



It has been already stated that the silicic hydrates are very 



unstable bodies, desiccation of a hydrate, H 2 ^0 , a t 100° C. 



\Si/ 

 being sufficient to convert it into the polysilicic hydrate H 2 ^0 , 



Again, the mineral Randanite of Salvetat, already several times 

 referred to, when dried completely at ordinary temperatures has 



W 



its composition represented by the formula H 2 ^o . on drying 



at 100° C. for a long time, however, its composition then is 

 \Si'V 

 H/<0 . Possibly, on still further heating, another compound 



would be obtained. It is easy to understand, therefore, how a 

 silicic acid may yield a polysilicic acid by the simple loss of 

 water. But the metallic base of a silicate cannot be got rid of 

 so easily. Heating of the compound alone cannot alter it, not- 

 withstanding the marked tendency of a silicate of a low type to 

 pass into a polysilicate ; for as the temperature increases, this ten- 

 dency likewise grows. Assuming that a typical silicate is ca- 

 pable of existing in a semifused magma, and that its formula is 

 x Si/ 



(5K 6 H-|Al) 2 /\° 6 ^ it obviously cannot lose its basic elements, 

 /O 6 \ 



* Haughton's 'Manual of Geology,' 1866, p. 16. Durocher's important 

 essay is here given in full, and contributes probably the best aid to be found 

 to the study of the paragenesis of minerals. 



