Mr Iliifjli Miller on Rioer-l'erracinrj. 291 



by an Ainerican geologist * in language so scientifically 

 graphic that we make no apology for (|uoting it. 



" When the stream is progressively building up its bed 

 outside of the gate, it is obvious tliat it cannot long occu])y 

 one position, for if it persisted in running for a very long 

 time in one place, it would build an embankment. Its 

 position soon becomes unstable, and the slightest cause will 

 divert it to a new bed, which it builds up in turn, and which 

 in turn becomes unstable, and is also abandoned. The 

 frequent repetition of these shiftings causes the stream to 

 vibrate radially round the gate as a centre, and in the lapse 

 of ages it builds up a half-cone, the apex of which is at the 

 gate. The vibration is not regular but vacillating, like a 

 needle in a magnetic storm ; but in the long run, after very 

 many shiftings, the stream will have swept over a whole 

 semicircle, with approximately equal and uniform results. 

 The formation thus built up is an ' alluvial cone.' . . . 

 There is one feature which the eye seldom recognises or even 

 suspects. The profiles are not truly conical, but are slightly 

 curved, instead of having a truly rectilinear slope. They 

 are concave outwards, the slope being a little greater near the 

 apex, and slightly or sometimes insensibly diminishing to- 

 wards the periphery. . . . -|- It is a surprisingly har- 

 monious result of a process which in its elements is apparently 

 irregular,, and becomes regular only by averaging the results 

 of its constituents. . . . The cone appears to be built up 

 of long radial or sectoral slabs, superposed like a series of 

 shingles or thatches." 



This is the history of all deltas and alluvial cones. In the 

 case of small fans, such as those that border alluvial vallevs, 



* Capt. Button, Geology of the High Plateaus of Utah, 1880, p. 2-20. 



t Capt. Duttou finds the slope near the apex to lie usually between 2° and 

 85°, while near the periphery it reduces to 1° or 2"^. In the fans laid out at 

 the mouths of streamlets on the sides of valleys in the North of England, I 

 find the slope to range to 6'' or more. But there is a perfect gradation in 

 radial slope between the river-delta of fine mud in which the slope is some 

 very low fraction of a degree, up to the talus of rude blocks in which it may 

 exceed 35°. The flattening-out to which Capt. Button refers is of course due 

 to the fact that the coarser discharges at the gate-mouth appn.iximate more to 

 the talus-condition. 



