42 THOMAS STEKRY HUNT ON A NATURAL SYSTEM IN 



these protosilicates may, like the feklspars, be formed alike by aqueous aud by igueous pro- 

 cesses, and that a broad distiuctiou as to origin must be drawu between those anhydrous 

 silicates, of both sviborders, which are due to aqueous deposition, and are often associated 

 with calcite and wàth quartz, and the same species which are found in plutonic rocks, aud 

 may be the result of iTystallizatiou from a cooling igneous mass. Propertosilicate solutions, 

 by like reactions with magnesian salts may, by double exchange, give rise to compounds 

 like iolite, the magnesian micas and the chlorites, while the source of glauconite is 

 probably to be sought in the reaction between dissolved protosilicates and ferrous solu- 

 tions, followed by a partial peroxydation of the resulting hydrous ferropotassic silicate. 

 Thus, as we have endeavoured to show, while natural processes, both igneous and aqueous, 

 unite in giving origin to the protosilicates aud the protopersilicates, it is the transforma- 

 tion of the latter by subaerial aqueous action, which, by removing the protoxyd bases, 

 generates the persilicates. 



§ 42. While the distinctions between the siiborders which have been deduced from their 

 genetic history are generally well defined, there are, however, two remarkable examples 

 which serve to connect the protosilicates with the protopersilicates. "With the typical 

 species of amphibole and pyroxene, which are protosilicates, there have hitherto been 

 included certain compoviuds which, while apparently identical with these in external 

 characters, contain notable j)ortions of alumina. Taking as a type of the aluminous 

 amphiboles, a lime-magnesia pargasite, we find for the atomic ratio of protoxyds, alumina 

 and silica, 2 : 1 : 3, which recjuires alumina 15'9, silica 422. This, if the alumina 

 replaces atomically, as has been suggested, a portion of the silica, would give the ordinary 

 amphibole ratio of 2 : 4. It is, however, a true protopersilicate, having the same atomic 

 ratios as melilite. The example offered by species like amphibole and pargasite, which, 

 from physical characters alone, would be referred to the same genus, shows in a striking- 

 manner the importance of chemical considerations in mineralogy. The most highly alu- 

 minous pyroxenes contain only about one half as much alumina as pargasite, and might 

 perhaps be regarded, together with the less aluminous amphiboles, as crystalline admix- 

 tures of protopersilicates of the pargasite type with a homœomorphous protosilicate. 



A wider view of the problem, however, leads us, while admitting the possible exist- 

 ence of such admixtures, to see that from melilite and pargasite, in which the atomic ratios 

 of protoxyd and ahimina are I'O : 0'5, we have a series in which the ratio becomes, as in 

 humboldtilite, 1"0 : 04, and so on to the less aluminous amphiboles and pyroxenes, iu 

 which it becomes I'O : 01 or even less, establishing a transition through these species to 

 the protosilicates. In like manner, towards the other limit of the protopersilicates, we 

 find this ratio changing from 10 : GO to 10 : 90 and 10 : 120, in indicolite, rubellite and 

 the muscovitic micas ; thus marking the transition to gem-like persilicates like andalu- 

 site and topaz, and to persilicate micas like kaolinite and pyrophyllite. 



^ 43. We may conceive the relation of the three suborders to each other to be repre- 

 sented by a design of two l)ands of ecjual breadth, but of unlike color, aud of diminishing 

 intensity of color, protracted in opposite directions along a common course, for a consider- 

 able part of which the tw^o bands overlie or rather blend with each other. The 

 unmingled portions of these two color-bands represent the protosilicates and the persili- 

 cates. As a result of such an arrangement, the protopersilicates, towards the protosilicate 

 end of the continuous series, include comparatively little alumina, as iu melilite, pargasite 



