452 Dr. C. Callaway — On River Curves. 



very few which come in on the concave. The same principle is 

 seen to obtain in the Ganges, the Indus, the Po, the Danube, and 

 every other great river system which I have examined. It is need- 

 less for me to multiply examples, as my statements can be easily 

 tested by the inspection of reliable maps. 



The coincidences I have pointed out cannot be accidental. They 

 clearly indicate the operation of some law. They strongly suggest 

 that the divergence of rivers in the direction of their affluents is 

 due to the affluents themselves. How the divergence is produced 

 I will now attempt to explain. 



Let us suppose that a river is running straight forward, and 

 a tributary comes in on one side at right angles. After heavy rains 

 the affluent will bring down sediment, and will discharge it, or 

 a part of it, into the main stream. Under ordinary conditions, 

 a large proportion of this sediment will be carried obliquely down 

 the river, and deposited on the opposite side, according to the 

 law of the parallelogram of forces. Let AB (Fig. 1) be the 

 river, and CD its affluent. Then let the line ah represent the 

 force of the river in magnitude and direction, and cd the force 

 of the affluent. Then a large part of the sediment brought in by 

 CD will be carried along the line c6, and deposited at h. Some 

 of the detritus may sink to the bottom before h is reached, and some 

 of it will probably be carried down below h ; but there must be 

 considerable deposit at h, for at that point the lateral force cd will 

 be balanced by the reaction of the bank, and the carrying power 

 of the current will be reduced. Of course, the finer sediment will 

 be transported further than the coarser particles, and h will be 

 a short line rather than a point. The growth of this new land 

 at 1) will cause a deflection of the current, which will impinge upon 

 the opposite bank lower down, and will begin to excavate. A series 

 of curves will thus be initiated according to principles which need 

 no explanation in these pages. 



C 



B Jr a . /\ 



Fig. 1. — Affluent (C D) entering river (AB), and depositing sediment at h. 



Not only will the deposit at h grow by additions from the 

 affluent CD ; it will form a barrier projecting into the stream, and 

 in the slack water above it sediment from the main river will be 

 thrown down. 



According to the explanation I have just given, the normal 

 position of the affluent should be a little ahove the convex curve, 

 and this is very frequently found to be the case. But the new 

 sediment thrown down above the projection formed at h will cause 

 the projection to grow up-stream, so that in time it will reach as 

 high as the tributary, or even higher; and the latter will enter, not 



