22 BURTON EDWARD LIVINGSTON 



ATMOSPHERIC EVAPORATING POWER AND ITS MEASUREMENT IN 



GENERAL 



Conditions Controlling Evaporation 



This is not the place for a detailed discussion of evaporation 

 and its controlling conditions, but enough space may be taken 

 to point out certain features of evaporation and certain re- 

 quirements of atmometric measurements in general. In the 

 first place it is to be emphasized that the water surface from 

 which evaporation proceeds is as truly a control of the rate of 

 water loss as are the atmospheric conditions. This consideration 

 at once frees us from the still prevalent idea that there is any- 

 thing fundamental and absolute in evaporation rates obtained 

 from a free water surface. To make this clear, assume a water 

 surface of known area, the water held in a pan or tank of known 

 volume, the surface being a known distance below the margin, and 

 the material and form of the container being also defined. Further 

 assume a certain complex of aerial conditions, wind, temperature, 

 humidity. Under such conditions it may be found that water is 

 lost from the exposed surface at a given rate per hour, which may 

 be reduced to an hourly rate from unit of surface. Now let the 

 water container be altered in size, all other conditions remaining 

 the same, and we shall find that the rate of loss per unit of area 

 is not the same as before. Let the area remain the same and let 

 the form or material of the container alone be altered, and another 

 change in rate of water loss occurs. Of course it is understood 

 that the changes here supposed must be of adequate magni- 

 tude, otherwise the resultant differences in rate might hot be 

 detectable by the extremely crude methods usually employed 

 for measuring the rate of water loss. All the details of the con- 

 struction of the water pan must have their influence upon the 

 rate of evaporation per unit of surface, under any given set of 

 surroundings. 



If, however, two instruments are assumed to be exactly alike, 

 and if they be exposed to the same complex of external condi- 

 tions, then the rates of loss should be the same. If, in this 

 case, the external conditions to which two like pans are ex- 



