RELATION OF CAPACITY TO DISCHARGE. 



145 



always greater than the corresponding value 

 of K, usually much greater. It was found by 

 trial that, within the range of conditions real- 

 ized in the laboratory, the difference between 

 values of efficiency computed directly by means 

 of (71a) and values computed indirectly by 

 means of ( 64) is not large, its order of magnitude 

 being that of the probable errors. 



The control of duty and efficiency by dis- 

 charge is further considered in the following 

 section. 



COMPARISON OF THE CONTROLS OP DIS- 

 CHARGE AND SLOPE. 



CONTROLS OF CAPACITY. 



We are now in position to compare the in- 

 fluences exerted by slope and discharge, sever- 

 ally, on capacity. The general fact brought out 

 by the comparison is that capacity is more sen- 

 sitive to changes of slope than to changes of 

 discharge, but the difference in sensitiveness is 

 not the same for all conditions. 



TA BLE 36. -Comparison of the index of relative variation.i, . for capacity and slope, with the index, i 3 ./or capacity and discharge. 

 [Foi 1 grades (A) to (D) the data are for S 1.0; for grades (G) and (II) for S 1.2.] 



In Table 36 values of the index of relative 

 variation are brought together from Tables 15 

 and 32. The selection includes all such as cor- 

 respond in respect to debris, trough width, 

 slope, and discharge, with the exception of 

 those of trough width 0.23 foot, which appear 

 to be anomalous. There are 64 pairs of values. 



Of the 64 comparisons, 62 show capacity as 

 more sensitive to slope, 2 as more sensitive to 

 discharge. The two exceptional cases are from 

 experiments with debris of grade (D) and with 

 channel width 0.66 foot; and the data from 

 those experiments were reexamined in search 

 for an explanation of what seems an anomaly. 

 No explanation was found, and, as the observa- 

 tions are supported by the estimates of pre- 

 20921 No. 86 14 10 



cision, it remains probable that there are real 

 exceptions to the general rule. 



The means of the 64 values of \ and i 3 are, 

 severally, 1.93 and 1.42; and the ratio of the 

 first to the second is 1.36. On the average, 

 the sensitiveness of capacity to slope is one- 

 third greater than the sensitiveness of capacity 

 to discharge. 







To ascertain the variation of the ratio -* 



H 



with discharge, the values in Table 36 were 

 specially grouped for the taking of partial 

 averages. The first group gave comparative 

 ratios for discharges of 0.093 and 0.182 ft.'/sec., 

 by means of four sets of index values, each set 

 agreeing as to all conditions other than dis- 



