Vol. 57.] MOUNT GIKNAK, JTIIfAGAKH. 49 



the crystallization of the other constituents. The fact that a 

 magma with practically identical chemical composition appears in 

 so many instances to represent the portion of the rock that remained 

 longest unsolidified, is a very significant fact. It can only be inter- 

 preted as indicating that analcime is an eutectic compound; that 

 is, it represents a combination of silica, alumina, soda, and water, 

 with a little potash, which solidifies at a lower temperature than 

 any mixture of those substances which differs to any considerable 

 extent in the proportions of its constituents, or in having an admixture 

 of other materials : see [31 J. If, however, when such a magma was 

 all that remained to be solidified, the rock were cooled with sufficient 

 rapidity, we should have a glass with the same composition as the 

 analcime-magma. Hocks similar in all respects to that which I have 

 described, but with a matrix that is glassy instead of crystalline, 

 must therefore be expected to occur ; and we should accept nothing 

 short of actual traces of crystalline structure, as conclusive evidence 

 that we are dealing with an isotropic mineral, and not a glass. 



The question of the true nature of the type-rock from Brazil 

 must therefore be still left in suspense, though the existence of 

 fluidal structure in one locality tends to show that in places the rock 

 is in fact a glass, while the clear subcircular spaces may in other 

 instances raise some presumption in the contrary direction. But the 

 specimens which, through the kindness of Mr. 0. A. Derby, I 

 have had an opportunity of examining, show no spherical areas 

 approaching those of the Girnar rock in regularity of shape, and the 

 isotropic material filling them is much less clear and transparent. 

 I may add that the occurrence of the Brazilian rock in very narrow 

 dykes ' 10 to 20 or 40 centimetres in diameter ' in gneiss, which 

 was probably at a comparatively low temperature at the time of 

 the intrusion, would seem to favour the possibility of the matrix 

 being glass. 



I would deprecate any specialization of the use of the term 

 monchiquite in respect of the physical character of the matrix,, 

 where its chemical composition is that of analcime, as the distinction 

 between the glassy and the crystalline groundmass may imply only a 

 slight difference in the circumstances of consolidation, and there is 

 no reason why we should not have both glass and analcime in the 

 isotropic material visible in the same microscope-section. 



As indicated by Bosenbusch [lOJ p. 455, the chemical composition 

 of this groundmass differs from that of a nepheline-syenite simply 

 in the amount of water, and to some extent in the proportion of the 

 alkalies. Eosenbusch lays stress on the association of monchiquites 

 with nepheline-syenites [10] pp. 447, 455, [20] p. 538, & [24] 

 p. 236, of which the present occurrence is yet another instance. 



YI. The History of the Book. 



The question of the order of crystallization of the different 



minerals and the circumstances under which they were formed is 

 not a simple one. 



Q. J. G. S. No. 225. e 



