of Northwestern Canada. 125 



connected manner completely aciossthe present position of the 

 Cordillera belt.* 



In the Queen Charlotte Islands, massive conglomerates im- 

 mediately overlie that part of the section which has been 

 referred to as the Queen Charlotte Islands formation. These, 

 it has been suggested by Mr. Whiteaves, represent the horizon 

 of the Dakota, and this reference is there strengthened by the 

 fact that the conglomerates (2000 feet in thickness) are in turn 

 overlain by shales holding Inoceramus jproblematicus. This 

 occurrence of conglomerates appears, however, to have more 

 than a local significance, for similar conglomerates are now 

 known to occur in the same (overlying) position relatively to 

 the earlier Cretaceous fauna in the northern part of Van- 

 couver's Island, on the Lewes Biver, in the upper part or at 

 the summit of the Tatlayoco, Jackass Mountain and Skagit 

 series previously referred to, and are again found to overlie 

 the Kootanie formation in the Rocky Mountains, forming 

 there a portion of the thickness of beds between the Kootanie 

 and Benton and consequently in all probability referable to 

 the Dakota. 



The constant or very frequent appearance of such massive 

 conglomerates at or about the Dakota horizon, may fairly be 

 taken to represent the initiation of an important and general 

 subsidence, which seems to correspond as closely as possible 

 with that referred to by Mr. Hill as the second great Cretace- 

 ous depression. It must be added, however, that in the north- 

 western portion of the continent, this second subsidence was 

 not so profound as that described in the Arkansas-Texas re- 

 gion, and was interrupted, in the area of the plains, by at least 

 one well-marked brackish-water and land epoch, represented by 

 the Belly River and Dunvegan series of rocks. 



The earlier Cretaceous rocks, here more particularly referred 

 to, and named in widely separated portions of their extent the 

 Kootanie and Queen Charlotte Islands formations, are again 

 clearly analogous to Mr. Hill's Comanche formation, with which 

 they have the same upward limit, and like it extend downward 

 far beneath the base of • the Cretaceous of the Interior Conti- 

 nental Plateau. In comparing the earlier Cretaceous rocks of 

 these two portions of the continent, however, we find that 

 though a distinct unconformity exists between the summit of 

 the Comanche and base of the Dakota of the southwestern 

 region of the United States, no such physical break is yet 

 known as between the Kootanie or Queen Charlotte Islands 

 formations and the Dakota ; while the very great thickness of 

 these formations, so far as it goes, may be regarded as tending 



* Cf. on Triassic, Trans. Royal Soc. Can., vol. i, sec. 4, p. 144. Annual Report 

 G-eol. Surv. Can. 1885, p. 161 B. 



Am. Jour. Sci.— Third Series, Vol. XXXYIII, No. 224.— august, 1889. 

 8 



