OCCLUSION OF IGNEOUS ROCK 419 



wehrlite, Iherzolite and pyroxenite — rocks all free from feld- 

 spar — correspond closely in chemical composition with our 

 serpentinoids, though differing in the purely magnesian base of 

 the original pyroxenes. 



No feldspar or other mineral of alkaline composition has 

 been distinguished in thin sections of our serpentinoids, nor the 

 "pseudo-mesh structure," indicative of serpentinized feldspar,^ 

 nor the presence of alkalies in any analysis.. The original con- 

 stitution of a gabbro, even of a basic type, is therefore elimi- 

 nated. The poverty in alumina and excess of lime show clearly 

 that the original rock did not consist of an aluminous silicate, 

 such as the aluminous augite or hornblende of an ordinary 

 pyroxenite, limburgite, theralite or similar ultra-basic rock. 

 They are however consistent with the composition of a diallage 

 aggregate. 



From a correlation of the results of the chemical analysis 

 with those of the microscopical examination of this serpentinoid, 

 the following conclusions become impressed. 



1. The remarkable excess of lime. This is not consistent 

 with the view that " on the assumption that all of the water is 

 united in the serpentine, the proportion of this mineral is to the 

 total mass as about 4 to 5." - In the composition of serpentine 

 magnesia is never replaced by lime. It becomes necessary 

 therefore to assign this content of lime entirely to remaining 

 molecules of diallage, mainly left in form of bastite, tremolite 

 and hydrated amphibole. 



2. The low percentage of silica. This indicates that a portion 

 of the base must be free from that combination, /. c, magnesia 

 as hydrate, in the form of brucite. A little calculation will 

 show that no possible mode of combination of any silicates of 

 probable occurrence could reduce the silica below 37 per cent., 

 as found in the analyses. It is interesting to note the still lower 

 percentage of silica and higher of magnesia in Nuttall's old 

 analysis : silica 30, ferric oxide 2, magnesia 52, and loss by 

 ignition 16. These figures, if approximately correct, would 



' Finckh, loc. cit., 109-110. 

 ^ Newland, loc. cit., 316 



