186 



Rev. Samuel Haughton, m.d., 



the following simultaneous equations, to determine the per-cent- 

 ages of the constituent minerals in the average granite : — 

 Let Q = the per-centage of Quartz. 



O „ „ Orthoclase. 



A „ „ Albite. 



W „ „ White Mica. 

 B „ „ Black Mica. 



(1) Silica, . 



(2) Alumina, 



(3) Iron peroxide, 



(4) Lime, 



(5) Magnesia, 



(6) Potash, . 



(7) Soda, . 



(8) Loss by ignition, 



7207 =100 Q+64-95 + 64-70 A+44-58 W+35-55 B. 



1481 = — 18-31 O + 21-80 A + 32-13 W+1708 B. 



225 = — — — 4-57 W+ 23-7 B. 



163 = — 0-25 O — 0-78 W+ 061 B. 



33 = _ o-58 O — 0-76 W+ 3-07 B. 



511 - _ 12-23 0+2-84 A + 1067 W+ 9-45 B. 



279= — 2-75 0+9-87 A+ 0-95 W+ 0-35 B. 



109 = — — — 534 \V+ 4-30 B. 



]f we select the four equations containing the largest percent- 

 ages, viz — The alumina, potash, soda, and iron peroxide equations, 

 we find after several reductions — 



A = 18-0 -f 0-156 W + 0-191 B. 



O = 37-65 — 0-909 W — 0-819 B. 



These equations show the manner in which the two feldspars 

 are related to the two micas. 

 We find finally — 



per cent. 



B = 5-81 



W= 19-16 



O = 15-44 



A = 22-10 



Inserting these values into the last seven equations, we ob- 

 tain — 



— 



Observed. 



Calculated. 



Diff. 



Alumina, . * 



1481 



1476-06 



+4-94 



Iron peroxide, . 





225 



224-99 



+0-01 



Lime, 





1G3 



22-28 



+ 140-72 



Magnesia, . 





33 



41-27 



-8-27 



Potash, . 





511 



511-45 



-0-45 



Soda, 





279 



280-64 



-1-64 



Loss by ignition, 





109 



126-89 



-17-89 



The agreement between calculation and observation is as close 

 as could be expected ; and the errors in the magnesia and loss by 

 ignition, are, doubtless, errors of observation, due to the small 

 magnitudes to be ascertained. The excess in the lime is real, and 

 must be accounted for by the existence of a small quantity of 

 paste, in the form of a silicate of lime. 



