109 [ 237 ] 



sors in this region) have preferred to be influenced by the latitudes given 

 by Major Pike, (which are very inexact, as I shall show hereafter,) rather 

 than to trust to their own experience. For instance: the distance from 

 St. Peter's to Sandy lake is generally adopted as from 500 to 600 miles : 

 it is by my surveys only 334 miles, viz : 180 miles from St. Peter's to 

 Kagiwigwa7i, or Crow-wing river, and 145 miles from this last to Sandy 

 lake. 



Finally, in 1839 I made a reconnoissance of the Mississippi from Prairie 

 du Chien to the Lower or Des Moines Rapids. The remainder of this 

 river has been laid down from the public land surveys. 



I might mention here, by the way, a fact very capable of modifying the 

 opinion of those who refuse still to believe in the prodigious progress 

 which civilization is making towards these distant and so little known re- 

 gions. When, in 1836, I ascended the Mississippi to explore its source, its 

 banks between the Lower Rapids and Prairie du Chien were almost des- 

 ert — some only of the principal towns of the present day were just coming, 

 as it were, out of the ground ; and when, at the close of 1837, I returned 

 to the eastern cities, 1 had only too much to answer those who asked my 

 opinion upon the many projects of speculation offered to the public, but 

 existing, as they said, only upon paper. Since that time I have had occa- 

 sion to devote two summers to the exploration of the regions contained be- 

 tween the Missouri and Mississippi ; and when, at the close of 1839, 1 de- 

 scended this last by canoe, to reconnoitre the portion between Prairie du 

 Chien and the Lower Rapids, I had to note in my field-book 37 towns, in 

 full development, upon the line where, two years before, there were not to 

 be seen the third part of this number. 



I have entered into the preceding detail in order to do justice to the la- 

 bors of geographers who have preceded me. But I must say that, in the 

 map which 1 present, 1 have everywhere preferred surveys made by my- 

 self, or by my direction ; because they rest upon numerous astronomical 

 observations — an advantage which my predecessors had not. Those who 

 will take the trouble to compare this map with other existing ones, will, 

 doubtless, not fail to remark the important alterations which these observa- 

 tions have introduced and justified in the astronomical geography of the 

 vast region it exhibits. 



Starting from Engineer Cantonment, the course of the Missouri is in- 

 clined more to the northwest ; the Great Bend, instead of being turned to 

 the east, faces the north ; and of all the embouchures of its tributaries, 

 some are lower, and others a great deal higher in latitude than was formerly- 

 supposed. 



The geographical position of Mini-ioakaiu or DcviVs lake, is also much 

 changed. This lake is invested with hills, and is perfectly isolated; it is 

 not, therefore, the source of several rivers falling into the Red river of the 

 IVorth, as drawn upon some maps; though, in times oi overflow, its waters 

 pass over a depression in the tjround, and run into the Sliayenoju. river. 

 It is the same way with the Mississippi: starting from the Falls of St. An- 

 thony upwards, all the river mouths bear too much to the west, and are too 

 low in latitude; and even its sources were placed more than three-quarters 

 of a degree too far to the wesf. Upon this line there is one important 

 mouth, the erroneous position of which has. till now, created jrreat disorder 

 in the geography of this region — I mean the niouth of the Kagiwigwan, 

 or Crow-wing river, the latitude of which, as given by Major Pike, is 27 

 minutes short. 



