.i8 
1.66 
2 1. go 
15.20 
6.60 
6.71 
4-54 
13.81 
•71 
3-03 
2,95 
8.22 
6.33 
3.83 
3-09 
1.61 
I.U5 
I.Q2 
2.29 
•33 
.25 
100.80 
99.70 
2.84 
Gabbroid Rocks of Minnesota.— IVinchcll. 293 
I U III 
Si02 52-48 49-15 45-65 
TiOg 126 
AI2O3 15-47 
Fe.Os ■ 5-14 
FeO 9-25 
MnO 51 
MgO 2.55 
CaO 7-27 
Na20 3-26 
K2O 1-75 
HgO 1-24 
PgOs 29 
100.47 
Sp. Gr 2.81 
-2.84 -2.86 
I. Orthoclase gabbro fairly rich in orthoclase, (i797) from Duluth. 
Minnesota, No appreciable BaO nor SrO ; F not determined. 
II. "Hornblende gabbro," from Dulnth. Minn.; by A. Streng: 
Neues Jahrb. f. Miner, etc. 1877. p. 177. MnO and Fl not deterijiined. 
TiOc and P2O., determined separately, and not included in the total. 
III. Orthoclase gabbro (1797), rich in hornblende, from Duluth. 
Minn. No appreciable BaO nor SrO; F not determined. 
The orthoclase gabbro constitutes therefore an interme- 
diate phase between the family of gabbros and that of the mon- 
zonytes, and is not a monzonyte properly speaking; the name 
proposed by Irving, and here used, is therfore to be retained. 
Origin. In conclusion, the reasons for which the ortho- 
clase occurs as a real agent of corrosion in relation to the 
plagioclases and other elements of this rock call for discussion. 
Several possible explanations present themselves: 
1 . Introduction of the orthoclase by means of infiltrating 
waters after the consoHdation of the rock. 
This theory would explain very well the abundance of hem- 
atite and other alteration products in the rock, but it cannot 
explain the constancy of orthoclase everywhere in such a large 
mass. It is therefore to be rejected. 
2. Production of the orthoclase by cxomorphic metamor- 
phism exercised by numerous dikes of orthoclastic rocks 
(augitic syenytes, etc.) which cut the rock here under discus* 
sion. 
