Professor W. M. Davis — River Meanders. 



147 



soon systematically exaggerated into meanders again. Abundant 

 verification can be given of this scheme of development, by observing 

 the behaviour of actual rivers, or more simply by studying the new- 

 edition (1900-1901) of the preliminary maps published by the 

 Mississippi Eiver Commission at St. Louis, Missouri, on* which 

 the river channels as determined by surveys in 1893-5 are over-printed 

 in red upon the results of surveys in 1881-3, printed in black. 



Fig. -1. 



Fig. 3. 



Any river that we now see meandering in an open alluvial plain 

 must have been meandering a long time. Its individual meander 

 curves must have already advanced down the valley over considerable 

 distance.s ; and it is, I believe, chiefly for this reason that tributaries 

 are taken in where the main stream bends toward them. To make 

 this clear, let a number of tributaries be added at random to the 

 latest river course, drawn in Fig. 2. Four of them are shown in 

 Fig. 3 : one enters the river at a convex bend, the other three at or 

 near concave bends. Now let the normal changes of the meanders 

 continue still farther, as in the dotted lines of Fig. 4. In the first 

 of these changes, tributary D is taken in by the convex curve next 

 above the concave curve that it entered before. In the second change, 

 the mouth of the tributary B is similarly transferred to a convex 

 curve. In the third change, the same fate overtakes tributary A. 

 During all these changes, tributary C has been only a little shortened ; 

 it still enters on a convex curve. When this curve comes to move 

 down the valley the tributary will prolong its course, but will 

 continue to enter the main stream on the up-valley side of a curve, 

 until it is captured by the approach of the next following curve. 

 It therefore appears natural enough that tributaries should usually 



