210 ANNUAL OF SCIENTIFIC DISCOVERY. 



beds of bog-iron ore in the upper peninsula of Michigan, on the 

 tributaries of the Monistique River. It occurs in a halt-desiccated 

 bog-covering several townships. It is of remarkable purity, and 

 great but unknown depth. It lies directly in the track of the 

 projected railroad intended to connect the Northern Pacific Rail- 

 road with the railroad system of Michigan. The ore can be 

 floated down the Monistique and its tributaries to Lake Michigan, 

 in the immediate vicinity of an excellent harbor. This immense 

 deposit is undoubtedly derived from the disintegration of the 

 hematites and magnetites of the contiguous region on the west. 

 The ore will possess great value for mixing with the other Lake 

 Superior ores. 



The third note made was on the discovery of an ancient outlet 

 of Lake Superior. Following the White Fish River from the 

 head of Little Bay de Noc, we find it occupying a broad and deep 

 valley, walled in on both sides by limestone cliffs attaining an 

 elevation of 120 feet. The head-waters of this river literally 

 interlace with those of the Au Train River, which runs north into 

 Lake Superior. Here is a vast valley of erosion but little elevated 

 in an}' part above the present level of Lake Superior. Through 

 this the waters of that lake must have flowed in a powerful stream 

 in that earlier epoch when all the lakes stood from 50 to 800 feet 

 higher than at present. There are many evidences of glacier 

 action along this valley. The stria3 at Marquette, near the head 

 of the valley, point north and south. In short, the evidences lead 

 to the conviction that a vast glacier stream once traversed this 

 valley, and was probably the agency by which it was excavated. 

 Little Bay de Noc is but the prolongation of this valley at a lower 

 level; and, indeed, the whole basin of Green Bay seems to be 

 but a phenomenon of erosion belonging to the epoch of the same- 

 glacier system. 



THE MAUVAISES TERKES IN COLORADO. 



Prof. Marsh, in a letter to Prof. Dana, dated Aug. 12, 1870, 

 says : 



" The scientific expedition from Yale College, while recently 

 examining the geology of northern Colorado, discovered an 

 extensive outcrop of the true Mauvaises Terres, or White River 

 formation, at a point nearly 200 miles south of the region where 

 it had been previously identified. The locality first detected, 

 which contained all the characteristic fossils of this deposit, was 

 on one of the branches of the Little Crow Creek, about 5 miles 

 south of the Wj'oming State line. The strata there observed 

 consisted of at least 150 feet of light-colored clays, overlaid by 

 conglomerates and sandstones about 200 feet in thickness. The 

 lower portions of the clays are the true Titanothcrium beds, con- 

 taining many remains of the Titanotherium Proutii. Above these 

 are similar clay deposits, corresponding closel} 7 in age with those 

 on the White River, and marked by abundant remains of Oreodon 

 Culbertsoni, Testudo Nebrascensis, Helix Leidyana, and many 



