INVESTIGATIONS IN MINNESOTA AND NORTH DAKOTA. 345 



north, wbicli became a oasiii that wouid remain filled with water, and 

 it naturally follows that after this water-course had ceased to receive 

 supplies from the north, the small amount of water that would escape 

 from this lake, with its diminishing velocity, would not erode a granite 

 bed sufficiently to rapidly reduce the depth of this lake. 



(2) The ancient shore line of Lake Traverse can be traced a mile 

 below the present limit of the lake and would indicate a former union 

 with Big Stone Lake since the days of active running w^ater in this 

 channel. The character of both the fauna and flora would bear out 

 this theory, since the difference in vegetation in the valley and on the 

 hillside is much more marked and pronounced between the lakes than 

 it is below the foot of Big Stone Lake or above the head of Traverse 

 Lake. Likewise the windings of the old bed of the Little Minnesota 

 River would indicate little or no opposition from running waters. 



The similarity of species of fishes found in the two lake basins would 

 indicate that these two lakes had at one time been connected and that 

 there was free water communication between the parts, while the 

 number of comparatively unimi)ortant and minor differences (which, 

 however, are quite noticeable and constant, and in some cases reach 

 almost varietal importance) show clearly the landlocked condition of 

 the fishes of Lake Traverse and amount almost to proof that these 

 lakes have not been united in recent years. 



LAKE TRAVERSE. 



Lake Traverse forms the greater part of the northwestern boundary 

 of Traverse County, one of the western tier of (counties in the State of 

 Minnesota. It is a long, narrow body of water, varying from 1 to IJ 

 miles in width. The valley is of a regular trough shape, sloping from 

 the top of the hills to the bottom of the lake. The water of the lake 

 lies low in this trough, with but a narrow margin of level land between 

 it and the steep hillsides. 



The lake is shallow, with a maximum depth of 30 feet and an average 

 of possibly 15 feet, the depth fluctuating to the extent of 2i to 4 feet in 

 periods of about four years. This variation is closely connected with 

 the rainfall, but is said sometimes to hapijen, without apparent cause. 

 The lake at the time of our visit was about 3 feet deeper than it had 

 been for four years. The water frequently falls so low that the large 

 drift bowlders project from the surface of the water in the middle 

 of the lake. The shore line is quite regular, marking a more than 

 usual stony area in the drift. The water is warm, 77° at the time of 

 our visit in July, and, owing to its shallowness and the high winds that 

 stir the lake to its bed, the top and bottom temperatures are practically 

 the same. The water is comparatively pure, never containing clay in 

 sufficient quantities to cause it to look turbid; it is soft and contains 

 but little mineral salts. The hills on eitlier side of the lake are 150 to 

 200 feet high and composed entirely of drift, a greater part of which 



