existing Physical Causes during stated Periods of Time. 275 



deep on one side and shallow on the other. The principal 

 deposit therefore takes place on the shallow or quiet side, and 

 the principal removal occurs from the deep side where the cur- 

 rent runs more quickly. 



This may explain why the traveller on the Mississippi sees for 

 hundreds of miles a caving bank on one side, and an advancing 

 sand-bar on the other (Lyell). When the action of the river 

 is also unequal on its two banks in different places along its 

 course, a channel consisting of curves instead of straight lines 

 must be produced. When each curve, however, had assumed 

 the complete horse-shoe form, the water, by travelling round the 

 outer circumference of the bend, will have its effective speed 

 reduced to that on the inner or shallow side. The current would 

 thus become more nearly equal in all parts of the channel, and 

 necessarily the deposit likewise ; and in winter it would have a 

 nearly equal tendency to excavate the banks on both sides, which 

 condition of equilibrium might last for some time. 



Hutton, in 1795, has remarked, that there is evidence of de- 

 nudation in every country where at any time of the year the 

 streams carry off any particles of the superficial soil*. The 

 Mississippi must derive its vast supplies of mud from thousands 

 of such tributaries; for it could obtain them from no other 

 source, unless we suppose it abstracts them from its own plains. 

 Certainly in many places soil is being removed fron^ one part or 

 other of its plains ; but an equal quantity must be added to some 

 other part, for the river could not make a permanent inroad into 

 its plains without enlarging its channel. This it does not do, 

 or it would be able to carry off the winter-freshets without over- 

 flowing, and the present artificial bank would be unnecessary. 



I have thus briefly referred to observations made by British 

 engineers which may throw some light on the causes of periodical 

 floods, and changes of channel in rivers, and also upon the for- 

 mation of alluvial plains along their course. These questions 

 need not further be entered into, because the limited growth of 

 alluvial plains and deltas may be best illustrated by tracing the 

 alteration in the mean level of a large part of North America 

 that would be consequent upon a denudation sufficiently exten- 

 sive to furnish the alluvium said to exist in the valley of the 

 Mississippi. On the borders of the Gulf of Mexico at the pre- 

 sent time marine strata are forming within a short distance of 

 the fluviatile, and frequently alternate with them, because spaces 

 of the sea- shore are enclosed by banks of river-mud and con- 



* Our clearest streams run muddy in a flood. The great causes, there- 

 fore, for the degradation of mountains never stop as long as there is water 

 to run ; although, as the heights of mountains diminish, the progress of 

 their diminution may be more and more retarded. Op. cit. vol. ii. p. 205. 



