250 



TRANSPORTATION OF DEBRIS BY RUNNING WATER. 



place ill vertical planes parallel to the axis. 

 The movements in a vertical plane may be 

 exhibited by giving to the trough a glass side 

 and by giving to the water in that plane ex- 

 clusive illumination. When one of our labora- 

 tory currents, being illuminated from above, was 

 viewed from the side, small particles in suspen- 

 sion were seen to be conspicuous, so conspicu- 

 ous, in fact, as to resemble the motes in a sun- 

 beam. Some of the particles were shreds of 

 wood fiber worn from the trough, and these 

 could be more easily followed by the eye 

 because of their distinctive forms. Impressed 

 by this phenomenon, I am confident that water 

 movements in a vertical plane can be effec- 

 tively revealed by giving to the water a suitable 

 amount of suitable suspended material and by 

 giving to the selected plane a brilliant and ex- 

 clusive illumination. 



With the aid of such simple arrangements 

 much may be seen, but measurement will be 

 difficult; and vortical movements may not be 

 easily discriminated from those which are 

 merely sinuous. It is believed, however, that 

 both these results may be achieved by aid of 

 the moving field. In a very simple form this 

 device was employed by us in the study of 

 processes of traction (p. 27), and despite the 

 crudity of the apparatus it was found to be 

 highly efficient. An experiment trough having 

 in its side a glass panel, A A in figure 4, bore a 

 sliding screen, B, in which was an opening, C. 

 Moving debris was watched through the open- 

 ing at the same time that the screen and open- 

 ing were moved in the direction of the current. 

 To the field of view was thus given a hori- 

 zontal motion, and that motion was, in effect, 

 subtracted from the motions of the objects 

 observed. The apparent motions were motions 

 in relation to the moving field, and not to the 

 fixed trough. No provision was made for de- 

 termining the velocity of the field, nor for 

 measuring the apparent motions of the objects 

 viewed; but even without these devices for 

 quantitative work, the possibilities of the 

 method of observation were sufficiently evi- 

 dent. With the necessary supplementary 

 devices, the moving field promises to measure 

 the horizontal and vertical components of 

 motion hi any part of the vertical section, and 



it thus makes possible the complete delineation 

 of velocities and directions of details of current, 

 so far as those details may be exhibited in 

 longitudinal vertical sections. By giving to 

 the field a suitable velocity it should be possi- 

 ble to see a traveling vortex which rotates in a 

 vertical plane, just as vortices of horizontal 

 rotation are seen on the surface of a stream. 



A notable defect in the Berkeley arrangement 

 was the requirement that the observer move 

 his head in unison with the moving field. This 

 interfered with steadiness and also limited nar- 

 rowly the space covered by an observation. It 

 could be remedied by substituting for the slide 

 a car which should carry both observer and 

 peephole at a determined rate. 



Another suggested arrangement places the 

 eye of the observer at a fixed telescope and 



A 



FIGURE 84. Diagrammatic plan of suggested moving-field apparatus. 



moves the field by means of a rotating mirror. 

 This is illustrated by figure 84, where the 

 trough T is shown in plan. The telescope E 

 views the trough through the mirror, which is 

 pivoted on a vertical axis at M. As the plane 

 of the mirror rotates from a to i the field com- 

 manded by the telescope moves from A to B. 

 It is evident that if the angular velocity of the 

 mirror is constant the linear motion of the field 

 along the trough will be relatively fast at A 

 and B and relatively slow at 0. The error thus 

 arising may be avoided by some device of the 

 nature of linkage. It will be corrected with 

 sufficient approximation if the axis of the mir- 

 ror be controlled by a rigidly attached arm 

 MD, which is in turn controlled by an arm 

 about one-third as long, the two having sliding 

 contact at D and the short arm revolving uni- 

 formly about a vertical axis at F. 



