REPORT OF THE CHIEF ASTRONOMER 137 



SESSIONAL PAPER No. 25a 



used in mapping the rocks overlying the Purcell lava in the McGillivray range. 

 Erosion has there, within the Boundary belt, removed the equivalents of the 

 Phillips and Roosville formations. 



These Gateway beds are so similar in composition and habit to those across 

 the Kootenay and already described that a special account of the former is not 

 necessary. They are marked by an unusual wealth of ripple-marks and anne- 

 lide trails and borings. Several beds of ferruginous and metargillitic quartzite, 

 occurring some 300 feet above the Purcell Lava, carry remarkably large and 

 perfect cubes of more or less limonitized pyrite. These range from 1 cm. to 

 4 cm. or more in diameter and form most conspicuous elements of the rock. 

 (Plate 15.) They often form simple interpenetration twins. The crystals 

 seem to have grown in the original mud either before or during the period of 

 its consolidation. On any other supposition it would be difficult to under- 

 stand how space was made for their growth; the lamination of the rock imme- 

 diately surrounding each crystal is usually quite undisturbed and not crinkled 

 or bowed around the crystal. 



The specific gravity of eight hand-specimens, representing types for the 

 whole Gateway formation in the McGillivray range, varies from 2.-646 to 2-747, 

 averaging 2- 637. 



STRUCTURE OF THE PURCELL MOUNTAIN SYSTEM. 



As already remarked there are special physical difficulties in the way of 

 discovering the structure of the Purcell system at the Forty-ninth Parallel; 

 hence the details of structure are not as well understood as are the structures 

 in the eastern ranges. Enough facts are in hand, however, to show that the 

 Purcell system is, like the Galton-MacDonald mountain group, chiefly com- 

 posed of great monoclinal fault-blocks. Of these twelve have been determined 

 without much residual doubt. Most of them are found in the Yahk and Moyie 

 ranges. The McGillivray range shows a tendency towards the structure of 

 terranes characterized by open folds. 



Between Gateway and the summit the Kitchener (Siyeh) and Purcell 

 Lava beds are warped into a broad, unsymmetrical anticline. The dips average 

 '35° N.E. on the eastern limb, a steepness of dip which would rapidly carry the 

 top of the entire Purcell series of sediments far below the level of the Devonian 

 limestone at Tobacco Plains. The distance between the limestone and the 

 most easterly of the outcrops (Purcell Lava) across the drift-covered Purcell 

 Trench is eight miles. We can only conjecture the structures beneath the drift 

 cover. Those actually visible indicate that the Rocky Mountain Trench is, at 

 the Boundary line, located on a zone of combined faulting and down-flexure. 

 In all probability the faulting has had the dominant control in locating the 

 trench. 



The western limb of the broad anticline shows northwesterly dips of 15° 

 to 20°. The convergence of strike lines on the two limbs shows that the fold 

 pitches gently to the north. 



