528 



BAGNOLD 



[CHAl'. 21 



Further, wliile there can be httle doubt that the threshold of bed disturbance 

 is definable as a threshold stress, it is by no means certain that under all 

 conditions the available power oj will be sufficient to maintain the transport 

 of the distiu'bed grains, either in the steady or in the oscillating cases. If the 

 value of to were to be insufficient, it is reasonable to sup])ose the result would 

 be a piling up of the excess disturbed grains to form raised features of some 

 kind. 



CO 



cc. 

 o 



20 30 40 50 



SEMI-AMPLITUOE, r (cm) 



Fig. 6. Threshold of grain movement on an vinrippled bed by oscillating water movement 

 in terms of orbital velocity and amplitude of water displacement over the bed. 



On this view it may well be that the threshold of bed movement will be fomid 

 to be better defined in terms of the power ojt in the dimensionless form 

 (jotlips —p)9D[{ps —p)gDlp]'^ than in terms of the bed stress, t<, in the dimension- 

 less form dt = rtl{ps -p)(jD. 



This problem, like so many others in the field of sediment movement, calls 

 for critical experiment designed specifically to throw light on the physics. It is 

 unlikely to be solved by speculations based on the results of engineers' experi- 

 ments designed for particular practical purposes. 



