CHAPTER IX. EXPERIMENTS WITH MIXED GRADES. 



ADJUSTMENT AND NOTATION. 



The tractional load of a natural stream in- 

 cludes particles with great range in size. The 

 grades of debris used in the laboratory had nar- 

 rowly limited ranges. Although the reasons 

 for this limitation were believed to be adequate, 

 the possibility was recognized that the laws 

 discovered by the use of artificial grades might 

 not apply without modification to natural 

 grades; arid in view of this possibility a series 

 of experiments were arranged to bridge over 

 the interval between the artificial and the natu- 

 ral in this particular respect. The same appa- 

 ratus and the same general methods being used, 

 observations were made first on mixtures of two 

 grades, then on mixtures of three or more, and 

 finally on a natural combination of sizes. 



When work on mixtures was begun, need was 

 soon found for a modification of the experi- 

 mental procedure. During the automatic proc- 

 ess of adjusting the slope to the load the cur- 

 rent acted unequally on the components of the 

 mixture, carrying forward an undue share of 

 the finer part and depositing an undue share, of 

 the coarser. To escape the difficulties intro- 

 duced by this partial re-sorting, a run was in- 

 terrupted after it had gone far enough to indi- 

 cate the approximate slope, the debris was 

 taken from arrester and experiment trough and 

 was throughly remixed, and then the approxi- 

 mate slope was artificially constructed in the 

 trough, after which the run was continued. 



The observational data are contained in divi- 

 sions (J) and (K) of Table 4. The values of 

 capacity in relation to slope were adjusted in 

 the same manner as with individual grades, 

 except that each value of a for an adjusting 

 equation was derived from the data of the par- 

 ticular observational series, without influence 

 from related series. The adjusted capacities 

 are recorded in Table 60, and with them are 

 fractional capacities computed for the several 

 grades composing the mixtures. Table 17 con- 

 tains the constants of the adjusting equations 

 and also series of values of the index of relative 

 variation. 



In each set of experiments with mixture of 

 two grades the proportions of the components 

 were varied. The usual series of proportions 

 was approximately 4:1, 2:1, 1:1, 1:2, and 

 1 : 4. To denote these mixtures, a notation 

 has been adopted similar to one employed in 

 chemistry; for example, a mixture of grades 

 (B) and (F) in the proportion 4 : 1 is desig- 

 nated (B 4 F,). 



The proportions of some of the mixtures are 

 told accurately by the subscript figures; those 

 of others only approximately. The theoretic 

 or standard mode of apportionment was by 

 weight of dry material, but when this was not 

 convenient the material was weighed or meas- 

 ured in moist or saturated condition, and the 

 actual ratio for dry weight was afterward 

 learned by computation. The actual propor- 

 tions are given in percentages in Table 4 (J) . 



TABLE 60. Adjusted values of capacity in relation to slope, for mixtures of two or more grades of debris and for an unsorted 



natural alluvium, based on data of Table 4 (J) and (K). 



169 



