176 



TBANSPORTATION OF DEBBIS BY RUNNING WATER. 



different system, intersecting or tending to 

 intersect the curves for the components. Evi- 

 dently the indexes of relative variation are 

 inharmonious, and evidently the inferences 

 drawn from data for the slope of 1.4 per cent 

 would not be duplicated by a discussion of 

 data from a slope of 1.0 per cent or 1.6 per cent. 



200 



o 



(0 

 Q. 

 <U 



O 



100 



Slope 



FIGURE 59. Curves of capacity in relation to slope for grade (C), grade 

 (E), and mixtures of those grades. The ratios of the components in 

 the mixtures are indicated. 



Graphic comparison has been extended to the 

 remaining data of Table 60, so far as dual mix- 

 tures are concerned, but no marked discordance 

 has been discovered outside of the (BF) group. 

 Although the cause of the exceptional discord- 

 ance has not been found, I believe that it 

 should be ascribed to some exceptional though 

 unknown circumstance and not be permitted 

 to nullify the otherwise harmonious testimony. 

 The tenor of that testimony is that the relation 

 of capacity to slope is substantially the same 

 for mixtures as for simpler grades of debris. 

 There is, however, a noteworthy qualification 

 to this statement, in that the values of a have 

 a smaller average for mixtures than for com- 

 ponent grades. 



The single comparison possible between 

 results obtained with different discharges indi- 

 cates that with mixtures, just as with their 

 components, capacity increases with discharge 

 in more than simple ratio. 



MIXTURES OF MORE THAN TWO GRADES. 



Experiments were made with a mixture of 

 three grades, and with a mixture of five. The 



former was observed with discharges of 0.182 

 and 0.363 ft. 3 /sec., the latter with discharges of 

 0.182, 0.363, and 0.545 ft. 3 /sec. (See Tables 4 

 (J) and 60.) 



Brief consideration only will be given to the 

 data from the mixture of three, (AjCA), 

 because the points on which they bear are more 

 fully covered by the data from the mixture of 

 five. Grades (A) and (C) differ from one 

 another in fineness much less than either 

 differs from grade (G) . The triple mixture may 

 therefore be thought of as half coarse and half 

 fine, with the distinction that the fine half is 

 made up of two grades. By comparing data 

 from it with data from the closely related mix- 

 tures (AA) and (CA), m which the fine portion 

 is of a single grade, we may throw light on the 

 question whether the advantage to traction 

 which is obtained by substituting two grades 

 for one may be augmented by further diversi- 

 fication. With discharge 0.363 ft. 3 /sec. and 

 slope 1.4 per cent the capacities for the three 

 mixtures are 



(AA) (AAG 2 ) (CA) 

 89 98 106 



As the capacity for the triple mixture has a 

 value midway between those for the two dual 

 mixtures, no advantage is indicated for the 

 greater diversification. 



Slope 



Slope 



FIGURE 60. Curves of capacity in relation to slope for a mixture of 

 five grades, (CDEFO). Comparison of mixture curves for three 

 discharges, and of mixture curve with curves for component grades. 



In the mixture of five grades the proportions 

 were so arranged as to approximate a natural 

 combination. The components and their per- 

 centages are shown by writing (C^D^E^FA)- 

 The curves of capacity in relation to slope are 

 shown, for the three discharges, in the left- 



