The KetreetKi ti'd II. — Winchell. 81 
the Lake Superior sandstone and the sandstones of the upper 
Keweenawan. It is true that the St. Croix beds there lie on 
the trap, the base of the former being a coarse conglomerate 
made up largely of trap material from the trap range. There 
are, however, some important considerations, relating to this 
occurrence, which are usually overlooked, viz. : 
The overlying non-conformable strata at St. Croix falls are 
not of the age of the Lake Superior sandstone, but younger, 
although probably a part of the same great formation. The 
fossils that are found in the strata ai-e those that characterize 
the horizon of the St. Lawrence limestone, which is the horizon 
of the original Dicellocephalus. The strata are, in part, not a 
sandstone, but a dolomyte, and below them, further south, are 
about 1,000 feet of siliceous sandstone. These lower layers 
were penetrated in sinking the deep well at Stillwater. They 
are well known both toward the east, in Wisconsin, and to- 
ward the west, in Minnesota, where they have been named 
Dresbach and Hinckley sandstones. It is these lower sand- 
stones that appear on the Lake Superior shore involved with 
the traps. The facts at St. Croix falls demonstrate that the 
subsidence which was going on during the time of the typical 
Keweenawan and later was yet in progress during the deposi- 
tion of the St. Croix formation, and that whatever fractures 
there may have been, in the Keweenawan rocks, incident to 
the movements of the crust in that region, were covered by 
the later sediments non-conformabl_y. The St. Croix strata 
are very far above the base of the sandstone formation, and it 
is necessary to find the base and show it is a non-cf)nformable 
conglomerate to warrant the assumption of a great erosion 
interval after the Keweenawan. This important element is 
lacking in several other similar non-conformities, if not in all 
of them so far as described. 
Prof. Chamberlin refers specifically to the phenomena at the 
falls of Black river in Douglas c()unt3% Wis., described by Mr. 
Sweet.* But there is here nothing to prove tlic liDrizou of the 
non-conformable sandstones. They are assumed to be later 
than the sandstones overlying and tilted with the traps, liut 
the descriptions and the figures of Mr. Sweet do imt iiinke 
that a necessary relation. Indeed one of Mr. Sweet's figures 
*Geol. of Wiseonsiu, vol. iii, pp. 340-317. 
