July 30. 1917 Evaporation from Water and River-Bed Materials 243 



If it is assumed that the bed of a stream is }4 mile wide, in a 2-mile 

 length there is a square mile of exposed sand. If it is assumed that 

 this is a water surface from which the evaporation is 40 inches a year, 

 the loss for that year would amount to over 2,100 acre-feet. Even if 

 the evaporation is only a small part of that, the loss is extensive over a 

 200 or 300 mile length of stream bed. 



This evaporation can not be prevented, but it can be estimated quite 

 accurately, provided basic data are available. These estimates are 

 necessary in order to answer fairly questions pertaining to equitable 

 division of water. So far as can be learned, only one previous investi- 

 gation has been carried on the results of which may be applied to this 

 problem.^ 



A study of this loss was made at the Denver Laboratory during the 

 season 191 6. In general, the method was the following: Typical stream- 

 bed materials were secured; these were placed in water-tight tanks; the 

 water table was held at certain fixed levels; the loss by evaporation was 

 measured; the final figures for this loss are given as a percentage of the 

 loss from a water surface. 



RIVER-BED MATERIALS USED 



At the beginning of the work samples of river-bed materials were 

 secured from the principal streams. The specifications for the collection 

 of these called for 



at least two samples, each of which will represent an average type of river bed 

 material. These samples should be taken in a vertical section extending from the 

 surface to the depth of 24 inches. 



Each lot consisted of 20 pounds. Materials from 27 points, covering 

 16 diff^erent streams, were obtained. These are given in Table XV, with 

 their numbers, location on the river, and the collector's name. This 

 number will be used throughout the discussion. 



' DiesEM, H. C. PREiiMiNARy REPORT, PLAxre VAiLEY INVESTIGATION. Unpublished. 



