DIVERSION OF INCIPIENT STREAMS. 101 



through a rugged barranca only a few miles away. A special feature 

 soon caught attention : On looking across the flood it was seen from the 

 movement of waves and flotsam that the rate of flow was generally uni- 

 form, a little more sluggish about the mesquite clumps, a little swifter 

 over the interspaces; but now and then a part of the sheet (usually 

 between and below mesquite clumps or slight elevations by which the 

 current was made to converge) began to move more rapidly, when almost 

 immediately the flotsam would shoot forward at twice or thrice the ordi- 

 nary rate, the flood surface would sink toward the upper end and swell 

 toward the lower part of the rush line, while the roar would rise above 

 the rustling tumult of the more sluggish waters ; within 50 feet or 50 

 yards the swelling rush would be churned into foam and rise several 

 inches above the general level of the flood, and then the waters would 

 diverge and slacken and quickly mingle with the general sheet ; some- 

 times the crest of a delta or fan would show through or above the water 

 at the lower end of the rush, and would push up stream through growth 

 chiefly at its upper end ; but in any event the whole process of the gath- 

 ering and respreading of the waters commonly lasted but a few seconds, 

 or perhaps a minute or two, and left but a faint trace in unusual rippling 

 of the flood. So common were these rushes that two or three or even 

 half a dozen might be within the field of vision at the same, time — some 

 just starting, some dying away. For perhaps five minutes the sheet- 

 flood maintained its vigor, and even seemed to augment in volume ; the 

 next five minutes it held its own in the interior, though the advance of 

 the frontal wave slackened and at length ceased ; then the torrent began 

 to disappear at the margin, the flow grew feeble in the interior, the water 

 shrank and vanished from the margin up the slope nearly as rapidly as 

 it had advanced, and in half an hour from the advent of the flood the 

 ground was again whitening in the sun, save in a few depressions where 

 muddy puddles still lingered. 



The after effects of the flood were not conspicuous, though significant. 

 The most striking effect was the accumulation of flotsam, chiefly twigs 

 and branches, against the upper sides of clumps of shrubbery, ant-hills, 

 ground-squirrel mounds, and other elevations ; and from these the ex- 

 tent of the flood could be traced even beyond the limits of vision, show- 

 ing that its wddth was at least one and perhaps two or three miles, and 

 that it nearly blended with other similar floods emerging from neighbor- 

 ing barrancas in the mountains and arroyos in the foot-slopes. A less 

 striking effect was the accumulation of a nearly continuous film of sedi- 

 ment, chiefly fine sand or silt, hardly distinguishable after drying from 

 the general surface deposit, with which it undoubtedly soon blended. 

 This film was usually an inch or less in thickness, though sometimes it 



