chap. vi. Trachyte and Basalt. 137 



usual order of eruption be inverted. From the high 

 degree of fluidity of most basaltic lavas, these perhaps, 

 alone, would in many cases reach the surface. 



As we have seen that crystals of feldspar, in the 

 instance described by Von Buch, sink in obsidian, in 

 accordance with their known greater specific gravity, 

 we might expect to find in every trachytic district, 

 where obsidian has flowed as lava, that it had proceeded 

 from the upper or highest orifices. This, according to 

 Von Buch, holds good in a remarkable manner both at 

 the Lipari Islands and on the Peak of Teneriffe ; at 

 this latter place obsidian has never flowed from a less 

 height than 9,200 feet. Obsidian, also, appears to have 

 been erupted from the loftiest peaks of the Peruvian 

 Cordillera. I will only further observe, that the specific 

 gravity of quartz varies from 2 # 6 to 2*8 ; and therefore, 

 that when present in a volcanic focus, it would not tend 

 to sink with the basaltic bases ; and this, perhaps, 

 explains the frequent presence, and the abundance of 

 this mineral, in the lavas of the trachytic series, as 

 observed in previous parts of this volume. 



An objection to the foregoing theory, will, perhaps, 

 be drawn, from the plutonic rocks not being separated 

 into two evidently distinct series, of different specific 

 gravities ; although, like the volcanic, they have been 

 liquefied. In answer, it may first be remarked, that 

 we have no evidence of the atoms of any one of the 

 constituent minerals in the plutonic series having been 

 aggregated, whilst the others remained fluid, which we 

 have endeavoured to show is an almost necessary con- 

 dition of their separation ; on the contrary, the crystals 

 have generally impressed each other with their forms. 1 



1 The crystalline paste of phonolite is frequently penetrated by 

 long needles of hornblende ; from which it appears, that the horn- 

 blende, though the more fusible mineral, has crystallised before, or 



