CLIMATE AND PLANT GROWTH. 21 



The evaporation from the white spherical atmometer and the free 

 water surface and the average daily relative humidity are shown in 

 figure 5. In order to determine which of the evaporation curves is 

 the most reliable for the periods under consideration, so far as rela- 

 tive humidity is concerned, note was taken of the number of cases in 

 which the evaporation curves slope in a direction opposite to the 

 corresponding humidit}^ curve. As the relative humidity decreases, 

 evaporation, other things being equal, would increase; hence it would 

 be expected that the graph of instrumental evaporation values would 

 show an opposite slope direction to the graph of relative humidity. 

 For the graph representing the free water surface evaporation the 

 slope is opposite on 19 days out of the 30, or in 63 per cent of the 

 cases. The graph of evaporation from the spherical atmometer, on 

 the other hand, slopes o^^posite to the relative humidit}^ graph in T3 

 per cent of the total number of cases. In deriving these percentages, 

 in the case of both evaporation" graphs, it was deemed advisable to 

 consider slopes as opposite in those cases in which they came very 

 near being so, as well as when they were actually opposite. Since the 

 evaporation curve for the spherical atmometer shows more cases of 

 slope opposite to the relative humidity curve than does the evapora- 

 tion curve for the free water surface, it may be considered that the 

 atmometer is somewhat more reliable than the free water surface in 

 determining evaporation for dail}^ periods, in so far as evaporation 

 is determined by relative humidity. 



A comparison of the evaporation values obtained from the two 

 instruments with the daily mean temperature is presented in figure 6. 

 In this instance it would, of course, be expected that the evapora- 

 tion curves would show agi'eement in slope with the curve represent- 

 ing the mean temperature. Examination of the graphs shows for 

 the atmometer 67 per cent agreement (20 i)eriods out of 30) with 

 the temperature curve; and for the free water surface only 47 per 

 cent (14 periods out of 30). If only slight disagreements between 

 evaporation and temperature are considered, the per cent of agree- 

 ment in the case of the atmometer record is even greater than that 

 from the free water surface. It is interesting to note that the evap- 

 oration record obtained from the free water surface commonly lags 

 about one day behind that of the temperature ; that is to say, if the 

 evaporation from the free water surface is compared with the tem- 

 perature for the preceding period there is a much closer agreement 

 than when the comparison is made for the same day. In order to 

 obtain an evajDoration record which is comparable with the trans- 

 piration of the plant for short periods, the instrument with which the 

 evaporation is measured should respond quicklj^ to temperature 

 changes in a manner similar to the transpiration of the plant itself, 



