146 CF, TOLMAN 
material of a single size is being transported; therefore m can be 
considered a constant. Both 7, the number of particles thrown up- 
ward, and d, the distance of throw, depend upon the velocity of 
subwhirl, which was assumed above (on account of lack of experi- 
mental data) to vary approximately directly as the average current 
velocity,’ or 
(2) Ecc? 
Assume that in the tumultuous advance of the flood sheet, the 
average throw of the subwhirls is equal to the depth of the stream 
(Fig. 2, A), then an increase in the velocity (due to some irregularity 
in its path) cannot increase d (Fig. 2, B), because the particles cannot 
LB. M 4. 
| 
i} 
KC ION \\ \\\\ \ 
Fic. 2.—A. The average upward throw of the bottom whirl is equal to the depth 
of the stream. B. The average upward velocity imparted by subwhirl is greater than 
that necessary to take the particle to the surface. 
MEE 
be thrown above the surface of the water. Therefore under these 
conditions 
(3) Tixcow: 
This shows only a very moderate Increase in carrying capacity, 
not at all comparable with the effect of current velocity on the power 
to roll material forward, which seems generally to have been assumed 
to govern the eroding power of a loaded stream on receiving an increase 
in velocity. 
In the above analysis, it was assumed that the material shot up 
by the whirls sinks downward undisturbed by other currents. As a 
t The Colorado River offers excellent opportunity to study the difference in 
velocity at various depths in times of flood when silt laden, and in times of clear low 
water. 
