juiyso. I9I7 Evaporation from Water and River-Bed Materials 237 



A careful consideration of all factors mentioned, and the curve (fig. 4) 

 which is approaching the horizontal for the large areas, leads the writer 

 to present the following conclusions: 



(a) Evaporation figures from tanks 2 feet or greater in depth, preferably 

 circular, set in the ground so that but a narrow metal rim not over 3 

 inches wide projects above the ground and in which water is kept 

 approximately at the ground level, are most applicable for extension to 

 large open- water surfaces. 



{b) Data on such tanks may be quite safely extended to large open- 

 water surfaces under exactly the same conditions of wind, air tempera- 

 ture, and relative humidity by multiplying the evaporation depth 

 from a — 



2-foot tank by 0.77 

 4-foot tank by 0.84 

 6-foot tank by 0.90 



9-foot tank by 0.98 

 1 2 -foot tank by 0.99 



Factors for depths from other tanks used have been found : 



3by3by3feet,sitting2.75feet in ground (Fort Collins type) o. 80 



United States Geological Survey floating standard, 3 by 3 by 1.5 feet 91 



United States Weather Bureau for class A station, 4 feet in diameter, 10 inches 

 deep, above ground 66 



These figures are not in agreement except in parts with those derived 

 from previous work (8). However, it is quite certain that no previous 

 investigation has gone into the problem in the manner of the one under 

 discussion. The agreement between the results from a floating pan and a 

 land pan of the same size is quite good; however, in many cases the descrip- 

 tion of the land pan does not state whether it was above the ground or set 

 in the earth. It is quite probable that earlier figures obtained from the 

 floating pan have been reduced too much in applying them to the reservoir 

 itself. The tendency in comparing various records for floating and land 

 pans has been to disregard the size of these pans. There are no data to 

 show that the ratio existing between the figures from 3-foot land and 

 floating pans would be the same as that from 6-foot land and floating 

 pans. On the contrary, the great difference between the evaporation 

 from land pans of 3 and 6 feet tends to show that the land-floating 

 evaporation ratio is different for each size of pan. 



METEOROIvOGICAL OBSERVATIONS TAKEN IN CONNECTION WITH EVAPORA- 

 TION INVESTIGATIONS IN GENERAL 



In the course of the study the writer has, up to the present, referred 

 to records in the original publications if possible, of evaporation measure- 

 ments at 84 points in North America. It is the purpose to compile these 

 in one table and present them to the public in a concise form if they can 

 be correlated. In only a few instances have the figures been obtained 



