Lavis — On the Structure of Rocks. 141 



manner, so that such minerals as leucite and pyroxene at Vesuvius, 

 or the latter mineral at Stromboli, are able to gradually increase in 

 size and perfection, which will proportionally diminish the crystal- 

 lizability of the remaining vitreous matter. Professor Samuel 

 Haughton 1 has shown that the remaining paste consists of a very 

 fusible basic glass with an approximate composition of 2RO, SiO 2 

 containing much iron protoxide. On the other hand, a large supply 

 of lava brought up from below with considerable rapidity has little 

 time for the growth of individual crystals, but the whole mass 

 undergoes a microcrystalline change until no, or very little, vitreous 

 matter remains to feed the further increase of individual crystals. 

 We have a parallel in such a case as the following : — If we make a 

 solution of some salt very soluble in boiling water, but very slightly 

 so in cold, and we cause such a solution to cool moderately quickly, 

 the salt will separate itself in a granular crystalline state ; but if 

 such cooling be made to take place gradually during many days, 

 very fine, perfect, and large individuals will replace the granular 

 types. Now, when a microcrystallization takes place, it will so 

 separate the remaining vitreous material that even under the 

 microscope little will be discernible, so that it is very difficult to 

 detect it or appreciate its amount. But where suitable conditions 

 favour the growth of large crystals in a similar magma, the vitreous 

 matter that remains will be more concentrated, and therefore more 

 apparent both to the naked eye and under high magnifying powers. 

 The histological character of any cooled magma, with regard 

 to its mineral components, is a question of profound interest, 

 which, up to the epoch of the attempts of artificial reproduction 

 of different types, aided by microscopical research, remained a 

 very obscure subject. When we have to deal with the fused com- 

 ponents of any single mineral in a pure state, the researches of 

 Messrs. Fouque and Levy demonstrated that, so far as laboratory 

 experiments go, the critical point of crystallization is near that of 

 the fusing-point of a mineral. We should, therefore, expect that 

 in a leucitic lava, the leucite would be first to separate as crystals, 

 to be followed by felspars, and last by pyroxene. It is a well- 

 known fact that some of these crystallized simultaneously. This 



1 ' ' Eeport on the Chemical, Mineralogical, and Microscopical characters of the 

 Lava^ of Vesuvius from 1631 to 1868."— Trans. Roy. Ir. Acad., vol. xxvi., p. 141. 



