56S MR. W. A. RICHARDSOX : A MJCROMETRIC [vol. lxxix, 



The rock is non-porphyritic, with a medium and uniform grain 

 throughout. No pleochroic haloes were observed, and the highly- 

 refracting mineral grains are rare in the mica. 



Dr. W. Pollard has made a complete analysis of china-stone 

 from Groonvean Quarry, 1 lying a little to the north-east of 

 Trethosa (No. 46). A doubtful trace of zirconia and high values 

 for phosphoric pentoxide and fluorine are reported, agreeing with 

 the measured values for the minerals concerned. It is interesting 

 to compare mineral composition calculated from this analysis 

 with the measured data. The procedure for calculating 'normative ' 

 minerals has been to some extent departed from, so as to make 

 calculation fit in more closely with minerals known to be present. 

 The lime was allotted, as usual, to apatite and fiuorite ; excess 

 fluorine was then calculated as topaz, and finally excess alumina 

 was allotted to water as kaolinite. In other respects the 

 ' normative ' procedure was followed, and the results are tabulated 

 below : — 



Quartz 31-3 



Orthoclase 31'0 



Albite 25-2 



Anorthite O'O 



Apatite 1*3 



Fiuorite 2-6 



Topaz 1-2 



Kaolinite 5'9 



Corundum 0\5 



Other normative minerals I'l 



Total 100-1 



The high values for apatite, fiuorite. and topaz agree well when 

 compared with the values for neighbouring places in Table IV 

 (facing p. 552). It is noticed that, after these allotments have 

 been made, no lime is available for the formation of anorthite, 

 and this is in agreement with microscopic determinations. With 

 regard to the principal minerals, quartz is slightly too high, 

 orthoclase too low, and albite about correct for the position of the 

 sample on the map. It must be remembered, however, that the 

 orthoclase is evidently somewhat altered, and some material must 

 have been removed ; a little albite must be transferred to orthoclase 

 to form perthite, and some orthoclase (or the kaolin) to form 

 gilbertite. If these adjustments could be made, it appears likely 

 that the sample would fit into its place within the quantitative 

 boundaries. 



If the three types of granite are compared quantitatively by 

 means of their averages (Nos. II— IV, Table V, p. o-'A). it is seen 

 that the biotite-granite is richest in orthoclase, the lithionite- 

 granite in quartz, and the gilbertite-granite in plagioclase. The 

 total micas are nearly constant. In order to compare these 

 granites mineralogically with others, Nos. V--VII in the same table 



1 J. A. Howe, ' Handbook to the Collection of Kaolin. &c. ' Mem. Geol. 

 Surv. 1914, p. 192. 



