FLOOD-PLAINS. 14 



I v 



same law. It is true the curvature is faint, so faint that in the field it 

 may he imperceptible to the eye, and in diagrams most authors ignore 

 it, and represent the cross-section of a flood-plain by a straight line; but 

 the curve, though slight, is real. If compared with the debris fan of a 

 mountain torrent it would be found to correspond to the outer margin, 

 where it blends into the plain and the convexity is slight. Strictly 

 speaking, a flood-plain extending many miles along a river cannot be 

 likened to a single alluvial cone; it is a composite structure made up of 

 a multitude of overlapping cones, each having its apex upstream at that 

 point where the silt composing it left the channel. The overlaps obscure 

 the convexity of individual cones, and the net result is a curve of large 

 radius and low convexity. 



Cross-section of constructive River. — The typical cross-section of a river 

 having a flood plain — a constructive river— so far from showing a straight 

 line from the channel to the bluffs, presents a convex curve of deposition, 

 besides a number of other distinct elements. If no terraces are present 

 the section will be as in figure 10, each half of the valley presenting a 



Figure 10. — Cross-section of a constructive Rivt r. 



a b = Weather curve at crest of bluffs ; ic^= Water curve of erosion ; e d = Swamp ; de= Water 

 curve of deposition. 



weather curve, two water curves of opposite character, and a swamp 

 along the line of intersection of the water curves. The presence of ter- 

 races adds much complexity. The swamp at c d is caused by a de- 

 pressed surface and an impervious subsoil. Only the finest argillaceous 

 sediments spread so far from the channel or from the bluffs, and these 



cut their channels deeper; and these instances, conspiring with the fact that the surfaces of flood 

 plains are alluvial, and with the fact that many terraces in glacial regions are carved from uncon- 

 solidated drift, have led some American geologists into the error of supposing that river terraces 

 in general are records of sedimentation, when in fact they record the stages of progressive cor- 

 rasion." 



The question about the true significance of terraces is aside from the present discussion, but the 

 ordinary meaning of flood-plain is too well settled to disturb it by so radical a departure as that 

 here proposed. Since Gilbert himself admits it to be descriptive of'a real phenomenon, we might 

 conclude that there are two kinds, flood-plains of eorrasion and flood-plains of sedimentation, and 

 that all we need is to distinguish these and use the term in each case with its proper adjunct; but 

 a closer analysis reveals two difficulties. In the first place, since tin' products of eorrasion on one 

 side of a stream are deposited as sediments on the other side, it turns out that the plain of cor. 

 rasion, or planation, is also a plain of sedimentation. This difficulty tends to merge the two kinds 

 into one ; but there is an opposite and more radical difficulty which rends them apart. The plain 

 of planation is not properly a flood-plain at all. Lateral eorrasion is nut exclusively a flood phe- 

 nomenon, though it is active in floods. The plain of planation differs so much from true flood de- 

 posits that the term flood plain ought to he restricted to the latter, as is usually done, 



