220 ALBERT JOHANNSEN AND E. A. STEPHENSON 



Washington (p. 225), orthoclase was taken as K2O • AI2O3 • 6Si02 

 in the place of the actual composition. In all probability the 

 orthoclase-plagioclase percentages of the Rosiwal and specific- 

 gravity determinations are more nearly those of the actual minerals 

 present than those computed from the chemical analysis. It will be 

 seen that the sums of the feldspars, 58.09, 61.4, and 59.81, are 

 not greatly different. ■ 



The sums of the leucocratic constituents in the three columns, 

 80.95, 83 .00, and 82 .02, are also close together, as are the biotite 

 and magnetite values. 



The only noteworthy variations in any of the determinatior^s 

 are in the values of the remaining constituents — the pyribole and 

 the pyrite. Here the values for the former are 5 .42, 7 . 5, and 6 . 54, 

 and for the latter, i .86, 0.3, and o. For the sum of the two con- 

 stituents in each column, however, we have 7.28, 7.8, and 6.54. 

 In the first place the sum of all the constituents in the analysis 

 calculated by Cross, Iddings, Pirsson, and Washington in column i 

 is only 99 . 15, as against 99 9 and 99 .81 in the other two columns; 

 consequently the first-column values should be slightly increased. 

 The calculated amount of pyrite (i .86) is greatly in excess of that 

 seen in any slide or in the hand specimen". The fraction of i per 

 cent given by the thin section is much more nearly correct. When 

 the greater part of the pyrite is added to the pyribole in column i, 

 the values are more nearly the same. In the specific-gravity deter- 

 mination the magnetite and biotite were removed before the pul- 

 . verized material was placed in the Sonstadt solution, which was then 

 diluted until the pyribole and the small amounts of pyrite, apatite, 

 and zircon present came down. Therefore these latter minerals are 

 included with the pyribole in that column. The rock analysis and 

 the thin sections were clearly from material slightly more melano- 

 cratic than the pulverized rock used in the specific-gravity 

 determination, as is shown by the difference between the leuco- 

 cratic-melanocratic ratio of columns i and 2 as against column 3. 



