FLUME TRACTION. 



209 



TABLE 71. Comparison of values of I t for flume traction 

 over a bed of planed wood, with corresponding values jor 

 stream traction. 



Various qualifications and doubts being 

 omitted, the preceding paragraphs may be 

 generalized by saying that the sensitiveness of 

 capacity to slope is somewhat less in flume 

 traction than in stream traction. It varies in 

 both directions from a mean value expressed 

 by the exponent 1 .5, being greater as the slope 

 is less, as the discharge is less, as the fineness 



is less, and as the channel bed is rougher. 

 Efficiency for flume traction increases with 

 slope. 



CAPACITY AND DISCHARGE. 



Special series of experiments were made to 

 determine the variation of capacity with dis- 

 charge. In each series the conditions of slope, 

 width, and grade of debris were kept constant 

 and the discharge was varied. The observa- 

 tions are given in Table 72, and the same table 

 contains the adjusted values of capacity, to- 

 gether with the parameters of the adjusting 

 equations. Inspection of logarithmic plots 

 showed the propriety of adjusting by means of 

 the formula used with the data for stream 

 traction, 



<7=& 3 2-K) (64) 



and the computations were graphic. 



TABLE 72. Observations and adjusted data illustrating the relation of capacity for flume traction to discharge , for a rectangular 



flume of planed wood 1 foot wide. 



[L, observed load; C, adjusted value of capacity; Q, discharge.] 



Several values of capacity in Table 72 agree 

 as to conditions with values in Table 68, and 

 these values would be identical if the experi- 

 ments were homogeneous. A comparison 

 shows that the values given by the experiments 

 comparing capacity and discharge are in gen- 

 eral the greater, the average difference being 

 6 per cent. This is evidently of the nature of 

 systematic error and is probably connected 

 with some change in apparatus or in detail of 

 20921 No. 8& 14 14 



experimental method which occurred between 

 the making of the two groups of experimental 

 series. 



The sensitiveness of capacity to changes of 

 discharge varies with conditions. It is greater 

 as the discharge is less, as the slope is less, and 

 as the channel bed is rougher. It is relatively 

 great for the coarsest and finest of the de'bris 

 used and less for intermediate grades. Un- 

 der similar conditions it is less for flume trac- 



