T. F. Jam.ieson — Changes of Level in the Glacial Period. 345 



quantity of water it discharged, and the length of time the lake 

 existed, is further shown by the size of the channel which the river 

 flowing out of it has cut through the thick bed of drift to the south- 

 ward. This channel forms a long trough, from 100 to 225 feet in 

 depth, commencing at Lake Traverse and extending for about 250 

 miles along the course of the Minnesota River to its junction with 

 the Mississippi. The fact of the excavation of this remarkable 

 trough being due to the outflow of the lake which formerly filled the 

 basin of the Eed Eiver was pointed out by Gen. G. K. Warren in 

 1868. 



The beaches left along the margin of the vanished lake are still 

 quite distinct, and have lately been made the subject of careful 

 examination by Mr. Warren Upham, of the U. S. Geological Survey.^ 

 He has executed a series of levellings along the uppermost beach for 

 a distance of 142 miles, extending from its outlet at Lake Traverse 

 northward to Maple Lake, which is about 20 miles east of Crookston, 

 and one interesting result of his work is to show that the beach does 

 not form a horizontal line, but slopes gradually upwards from south 

 to north along the whole distance he carried his levels. Its height 

 at the south end is about 85 feet above the present level of Lake 

 Traverse, or 1055 feet above the sea ; while at Maple Lake, where the 

 levelling stopped, it is 1180 feet above the sea, showing a gradual 

 rise of 125 feet in the course of 142 miles. " Through this distance," 

 says Mr. Upham, " the upper beach clearly marks one continuous 

 shore-line, and the accuracy of our levelling is attested by close 

 agreement with railroad surveys at five widely-separated points." 



The ascent of this beach from Lake Traverse northward is at the 

 rate of about -A- of a foot per mile for the first 60 miles. Further 

 north, through the next 80 miles, its rate of ascent is somewhat 

 greater, varying from three-fourths of a foot to 1\ foot per mile. 

 The portion of the upper beach thus examined extends through 

 a prairie region very favourable for exploration and levelling. 

 Further on it turns to the E. and N. through a trackless forest 

 with impassable swamps, where it would be difficult to trace its 

 course. 



There are two other beaches, at lower levels, left by the lake as 

 it gradually subsided, owing to the wearing down of its outlet. 

 These also have been examined by Mr. Upham with somewhat 

 similar results. Both are found to have a gradual upward slope to 

 the north, as compared with the surface which a sheet of water would 

 now have if confined in this valley. The second beach, in the space 

 of 150 miles, rises about 70 feet with a nearly uniform slope of a 

 little less than half a foot per mile. While the third or lowermost 

 beach was found in the course of 135 miles to rise northward to the 

 extent of 50 feet, and at a later period of the lake's existence to half 

 that amount. 



Assuming these levellings to be correct, and assuming also that 

 each of the supposed beaches does actually represent one of the old 



^ "W. Upham, Lake Agassiz, Bulletin of the Minnesota Acad, of Nat. Science, 

 vol. ii. p. ■I'dQ, 1882. 



