Hi 



intake by the plant, the "balance being water freed from 

 the water-film system because of its lowered capillary 

 tension. The practical effect of this is that readings 

 of the water loss from the reservoir do not represent 

 precisely the actual water intake of the plant if the 

 temperature lias changed between the initial and final 

 reading. If the temperature has fallen, as, for instance, 

 in the evening, the reading will be too high. If it has 

 risen, as in the morning, the reading will be too low. 

 If should be noted that the error occurs only in the case 

 of change of temperature. If the temperature is con- 

 stant, the reading of water loss represents accurately 

 the water intake by the plant, regardless of what the 

 temperature actually is.. Similarly, there is no error 

 if the temperature of the initial and final readings be 

 the same, regardless of how the temperature has varied 

 in the meantime. Thus a twenty- four hour period usually 

 has about the same initial and final temperatures, espec- 

 ially if the readings are made relatively early in the 

 morning or relatively late in the afternoon. According- 

 ly a reading of water-loss over such a twenty- four hour 

 period is not likely to vary widely from the actual in- 

 take of water by the plant. 31ow seasonal changes of 



1. f course, the changes must be slow enough 

 to permit constant maintenance of equilibrium. 



