River engineering has been faced for many years with the dilemma that, on the one hand, its accumulated experience is sometimes inadequate for reliable prediction of river behavior, while, on the other hand, there has been little attempt at systematic quantitative evaluations to “discover identity in difference”—all the above referring particularly to the form of river‐channels, in cross‐section, plan, and profile. The purpose of this paper is to review certain principles which afford a rational derivation of the profile of river‐beds and to test the results on several American rivers.

Fluvial morphology may be defined as the science of the forms of river‐beds, their shapes in plan and in profile. There seems to be no satisfactory explanation of the cause of lateral migration of streams, that is, of the phenomenon of serpentining or meandering; and this is also true of the rhythm of change of the profile as seen in the downstream‐progression of its pools and concomitant shoals. Extant knowledge is generally qualitative; if quantitative, it is usually of empiric origin and applicable.only to a particular stream. Still there are some universal quantitative principles of fluvial morphology that have been developed theoretically. The purpose of this paper is to present, first, a very brief statement of some of the qualitative observations and, second, a more detailed statement of several quantitative laws and their use.

Recent studies [see 1 of references at end of paper] seem to indicate that there is no definite oasis for selection of a design formula for a non‐silting and non‐scouring channel. One reason for this lack is the failure or inability, except in the case of the Lacey formula, to include the effect of the size or mechanical composition of the bed‐material and the solids load, as well as the magnitude of the load. In connection with certain studies dealing with the Imperial Desilting Works of the All‐American Canal, it was necessary to determine the maximum size of material that might be delivered to the desilting works by the Colorado River and also what sizes could be carried in the channels beyond the canal‐intake. Various formulas or experience data were considered, but only those judged after careful analysis to be relevant, usable, and most reliable were employed. Since, as will be shown immediately below, these formulas yield a definite particle‐size criterion for a stable channel in erodible material, in that they correlate the maximum transportable size and certain hydraulic channel characteristics, they are presented here together with a comparison of the results obtainable with them.

Long‐range prediction or precipitation and its value In hydraulic engineering Is exemplified by its application to the drainage‐area of Telgitsch Creek in Styria, Austria. Its drainage‐area at the Langmann Dam is 170 sq km (42,000 acres) and the yearly mean flow is 316 cu m/sec (127 cusecs).

No abstract