48 Quantity of Water Consumed in Irrigation. 



APPENDIX D. 



Calculation of the Probable Depths op Water Lost by 

 Percolation in Various Soils. 



Taking the figures given in App. B, the average depth lost over the 

 whole wetted surface of canal, distributaries and watercourses (case 

 A) was "3638 ft. — or, more probably, the canal not being included 

 (case B), '6064 ft. — over wetted surface of distributaries and water- 

 courses only. 



In light sandy soil the duty obtained from the water (see par. 9) 

 was 72 acres, making the consumption per acre = aq -f- 72 = 183,000 

 c. ft. Since the quantity used in the fields in sandy soil has been 

 taken at 72,000 c. ft. per acre (see par. 13), that lost by percolation 

 would be 183,000 - 72,000 = 111,000 c. ft. per acre. It may be 

 assumed that the depth percolating through a light sandy soil would 

 bear the same ratio to the average depth lost in the canal as the loss 

 per acre in the light soil (111,000 c. ft.) bears to the average loss per 

 acre (51,000 c. ft. — see qa, App. B). Then — 



As 51,000 : lli'obo :: I «£. : -792 £t -case A. 



( -6064 ft. : 1-320 ft. — case B. 

 Now, since the loss per acre in light soil is based on the results 

 obtained in the Anupshahr division of the Ganges canal, in which 

 sandy soil largely predominates, though there is also some clayey soil, 

 it is fair to assume that, had the soil been entirely sandy, the loss 

 would have been greater. Assuming case B as the correct one (see 

 par. 7), it will probably not be thought too much, after reading the 

 instances of loss mentioned in the paper (pars. 18 to 21) and in 

 Appendix C, to take the loss, in what in App. A (class 2) is termed an 

 excessively sandy soil, at lj? feet per diem. This result may be con- 

 sidered as of general application in India or elsewhere. 



The loss in clayey soil is not so easily estimated. The Cawnpore 

 division of the Ganges canal, as a type of a mixed sand and clay soil, 

 shows a duty of 149 acres per c. ft. per second (see par. 9), making the 

 consumption per acre == aq -h 149 = 88,000 c. ft. Calculating as 

 above (in the case of a sandy soil), the depth would be *523 ft. in case 

 B ; but, as far as present information goes, this is too vague to be of 

 much use. The loss in pure clay is practically nil ; but a slight 

 admixture of clay reduces materially the percolation through a sandy 

 soil. On the whole, perhaps, the empirical scale given in App. A is as 

 near an approach to a correct estimate as is possible at present (see 

 par. 30). It will at any rate be better than taking one depth for any 

 mixture of sand and clay, irrespective of the proportions. If further 

 inquiries are carried on as recommended in the paper (par. 33), more 

 definite results may be obtained in time. 



