PLATE LXIV. 



BELTRAMI COUNTY, 1S99. J. E. TODD. 



That portion of Beltrami county which was submerged by lake Agassiz (except 

 Beltrami island) is separated from that which was not by a line running east and 

 west approximately through town 150 north, but toward the east boundary of the 

 county bearing northw r ard so as to enter towns 151 and 152. The area of lake 

 Agassiz has the well known physical features of the basin of the Red River of the 

 North, excepting a small tract, not well known, which lies in the north central part 

 of the county, which rose above the surface of that lake, increasing in size in the 

 later portion of its history until it included probably fifteen to twenty townships, 

 and was at last blended into the marshy tracts that still surround it, lying not much 

 above the Lake of the Woods. That part of the county lying south of the above 

 mentioned line is one of moraines and lakes, with all the diversity which marks the 

 "park region" of Otter Tail and Becker counties. 



The water divide between the waters flowing south to the Mississippi and those 

 tributary to the Red River of the North is quite invisible. There is an intricate 

 inosculation by means of many lakes and winding streams over an undulating, broad 

 tract, so that it is with.considerable difficulty that the actual divide can be followed. 

 Some lakes, near the watershed, having no visible outlets, probably are drained both 

 ways by entering the gravel beds of the drift. Lake Julia is described by Beltrami 

 as one of that kind. He considered it the source of the "Bloody" river as well as 

 of the Mississippi, but Mr. Todd says it lies, with other lakes, in the course of a broad 

 depression or channel that crosses the main divide and which once, probably during 

 the later part of the ice-period, was a water-way of drainage from the upper morainic 

 regions of the Mississippi to the valleys tributary to Red lake. 



According to the direction of the moraines in the southern part of the county, 

 Mr. Todd has inferred that some of the southern ones were formed by an ice-lobe 

 extending westward from the region of lake Superior, about the same time that the 

 greater lobe of the Red River valley was on the northern part of the county, the two 

 lobes having a reentrant interlobate area about where Itasca lake lies. Into this 

 interlobate area were accumulated not only larger amounts of morainic debris, but 

 also a more copious discharge of water, the latter having its escape to the westward 

 through a channel that crosses the moraine about a mile northwest from lake Itasca, 

 and at a later date through a valley that extends southwestward from upper Kice 

 lake and into Norman county, and at a still later date by the Clearwater valley. It 

 is also apparent, according to Mr. Todd's descriptions and interpretations, that a 



