364 G. W. Tyrrell — Bekinkinite of Barshaw, Renfrewshire. 



It was also found necessary to fix the amount of olivine in the 

 bekinkinite. It was assumed that olivine was present in the same 

 amount (18-5 per cent) as it occurred in the norm, calculated 

 according to the methods of the American Quantitative Classification. 

 This assumption was much strengthened by the fact that the amount 

 of olivine (serpentinized) in the rock, measured by the Rosiwal 

 method, came to 18-6 per cent. In the calculation of lugarite it was 

 found necessary to fix the amount of analcite present. This was 

 taken as 12 per cent, since rough measurements (micrometric) of the 

 amount of analcite present in two varieties of the Barshaw lugarite 

 gave 1 1 • 1 and 13*8 per cent respectively. 



The various compositions, as given by the two methods, are set out 

 in Tables II and III. 



The duplicate measurement of variety 2 of lugarite (Table II, 

 col. Ill) was made a year earlier than that of col. II, and on fewer 

 slides. Notwithstanding, the agreement between the - two is- 

 sufficiently good. 



TABLE III. 

 Modes obtained by Calculation fkom the Chemical Analyses. 



I. Bekinkinite. Barshaw. 



II. Bekinkinite. Barshaw. Second calculation. 

 III. Lugarite. Barshaw. 



Analyses of titanaugite from Craw ford John, and of barkevikite 

 from Lugar, were used in Nos. I and III ; in No. II an analysis of 

 augite from monchiquite of Serra de Tingua, Brazil (analysis q of 

 table xii, Cross, Iddings, Pirsson, and Washington, Quantitative 

 Classification of Igneous Hocks, 1902) was used in place of the 

 Crawfordjohn titanaugite. 



