86 
WATER REQUIREMENT OF PLANTS. 
tion was ascribed to the transpiration of the crop. This, divided by 
the dry matter produced, gave the water requirement. The crops 
grown were heavy, amounting to 5,000 pounds or more per acre in the 
case of the cereals. About 50 square feet were harvested for deter- 
mining the dry matter produced. The area of this sample included 
the point where the initial moisture determination was taken. Mois- 
ture samples were taken to a depth of 9 feet. Table LXXIV shows 
the results of Leather's field determinations of the moisture require- 
ment and includes also some determinations of the water requirement 
of wheat grown under irrigation at Cawnpore, India. 
Table LXXIV. — Water requirement of crops in India grown under field conditions, 
according to Leather (1911, pp. 234-279). 
Crop. 
Station. 
Year. 
Rainfall 
during 
experi- 
ment. 
Water require- 
ment. 
No ma- 
nure. 
Manure. 
Wheat 
/Pusa 
1908-9 
1909^10 
1908-9 
1908-9 
1909-10 
1909-10 
1908-9 
1907-8 
1908-9 
1909-10 
1908-9 
1909-10 
1908-9 
1908-9 
1909-10 
Inches. 
0.38 
.22 
.25 
.25 
1.81 
1.81 
.38 
2.8 
.38 
.22 
.38 
.22 
.38 
.32 
.16 
415 
\....do 
(Cawnpore 
j....do 
i....do 
[... .do 
237 
442 
550 
407 
302 
430 
411 
331 
313 
341 
1 do 
J ...do 
Oats 
230 
1. ...do 
/....do 
\....do 
.do . . 
282 
693 
706 
314 
471 
477 
718 
940 
505 
Mustard 
/....do 
\....do 
298 
614 
Unfortunately, Leather's experiments were, for the most part, 
conducted upon a single plat hi each case. It is consequently not 
possible to form any definite conception of the probable error of the 
experiments. Results from duplicate plats are, however, available 
in the case of the irrigated wheat grown at Cawnpore, and the probable 
error for a single determination is, in the case of these results, 12 per 
cent and 17 per cent, respectively, of the observed water requirement. 
It is therefore evident that a relatively large observational error is 
to be expected in this method of determining the water requirement. 
Some of the anomalies appearing in the results, notably the higher 
water requirement obtained in some cases when manure is used, are 
probably to be ascribed to experimental error. From investigations 
now in progress regarding the loss of water from fallow lands hi the 
Great Plains, the writers are inclined to believe that the direct evapo- 
ration from the soil of the fallow land is considerably greater than the 
corn spending loss from the soil of the cropped plats. The soil of the 
cropped plats is not only shaded by the growing crop, but is also 
considerably protected from wind, so that the humidity of the air is 
285 
