A STUDY OF COBBLE ABRASION 



513 



It is to be noted that severity of wear is nearly proportional to 

 the number of cobbles up to a certain point and then remains at 

 an almost constant value. This point, beyond which further 

 additions make little change in rate, is reached when the cobbles 

 are rolling close-packed across the full width of the drum and 

 additional cobbles roll down on top, forming a second layer. 

 Future work will advantageously be done with a fihing equal to or 

 greater than 100X4 1.7 grams to ehminate to a large extent the 

 critical effect of small changes in amount of mixture. The run 

 shown in Figure 5 was started at 144 cobbles, reduced by stages 

 to 14 cobbles, and then in- 

 creased in an attempt to 

 duplicate the curve. The 

 form is nicely duplicated, 

 while the change in absolute 

 value is due to the diminu- 

 tion in the meantime of the 

 test pieces and their conse- 

 quent higher rate (see Fig- 

 ure 2) ; therefore the greater 

 ordinates of the upper and 

 later curve. 



4. Relation of wear to an- 

 gularity. — Not yet studied. 



5. Relation of wear to 

 violence of motion. -^-Not yet 

 studied. 



6. Relation of wear to kind of rock.- — Not yet studied. 



7. Relation of change of shape to distance traieled. — In stream- 

 transported materials the change in shape, provided the rock is 

 homogeneous, is an approach to sphericity. In studying this 

 change in shape it is necessary to recognize varying degrees 

 of roundness, in other words, to have a numerical answer to the 

 question of how round a given piece is. At least three criteria 

 of roundness readily occur to one in considering the question. 

 These are (i) the ratio of surface area to volume, (2) the average 

 deviation of diameters from a mean diameter, (3) the average 



Fig. 6. — Gauge used in measuring curva- 

 ture. Figures give radii in millimeters. 



