Oct. 15, 1914 



Water Requirement of Plants 



39 



The effect of time of planting is shown in the following determinations 

 of the water requirement of alfalfa : 



Water 

 Crop requirement 



Alfalfa, grown August 7 to November 6, following Kharkov 



wheat 954rt 16 



Alfalfa, grown August 7 to November 6, following Turkey wheat. 912 ±15 

 Alfalfa, grown September 3 to November 7, following Bluestem 

 wheat 473±i3 



Alfalfa planted during the dry, hot days of August required almost 

 twice as much water for the production of a unit of dry matter as it did 

 when planted in September. 



The seven varieties and strains given in Table XXIII were included in 

 these late-season experiments and were grown under comparable condi- 

 tions. The water requirement of Grimm alfalfa, Sudan grass, and Kursk 

 millet was also determined in midsummer (see column 3), so that it is pos- 

 sible to reduce the water-requirement measurement of the other late-season 

 crops to a midsummer basis. The seasonal water requirement of Grimm 

 alfalfa was 39 per cent higher than that of the late-season crop. Assum- 

 ing this ratio to hold for the yellow-flowered alfalfa, the seasonal water 

 requirement of the latter would be 865±i8. The midsummer water 

 requirement of Sudan grass and of Kursk millet (S. P. I. 34771) was in 

 each instance 16 per cent above the late-season crop, and this ratio has 

 been used in computing the other millets to a midsummer basis. The 

 computed values (o) are given in the last column of Table XXIII. 



Table XXIII. — Water requirement of late-season crops 



o Computed. 



Of these seven varieties the }ellow-llowered alfalfa (Medkaqo falcata) 

 gave a water requirement in practical accord with the Grimm selection 

 grown during the same period. Kursk millet (S. P. I. 34771) gave the 

 lowest water requirement and proved to be decidedly more cfllcient than 

 Black Voronezh proso and Turkestan millet. Sudan grass required 91 

 per cent more water than the Kursk millet, which is in exact accord with 

 the results obtained for these two crops from determinations based on a 

 whole season's growth. 



