July 30, 1917 Evaporation from Water and River-Bed Materials 249 



evaporation loss. Nor is it thought that such an investigation would 

 have been of greater value than the one carried out. Typical sands to be 

 used in the study were selected, No. 10 and 11 having a mean analysis 

 curve, No. 1 2 representing a fine material, and No. i a coarse sand. Then 

 in addition laboratory soil was used, this being of a still finer texture than 

 No. 12. 



EQUIPMENT 



All records from water-surface evaporation tanks were taken, and all 

 meteorological observations were made in connection with water- 

 surface evaporation. No special apparatus was required for that. The 

 equipment for the sand work consisted mainly of the Fortier type of 

 water-jacketed tank (10, 1 1) shown as D, figure 3. Plate 38, A , shows the 

 installation of these. Net losses from evaporation were determined by 

 weight, the tanks being lifted to the platform scales in the manner shown 

 in Plate 38, B. This motor-driven hoist runs on wood rails; its capacity 

 is 2,500 pounds. Since the water levels were below the slirface, a well 

 (tin pipe i inch in diameter) was placed in each tank and the top closed 

 with a cork stopper (Pi. 38, .4). A description of the tanks is given in 



Table XVIII. 



Table XVIII. — Description of tanks used 



« Labaratory soil. 



There were available 15 hoods for use in time of storm. These were 

 used on tanks for which the correction necessary for the addition of 

 storm water was difficult to make. 



