588 



SCIENCE. 



[N. S. Vol. XXII. No. 567. 



be required to pay for so many cubic feet 

 of water delivered. The government or the 

 association owning the canal will then have 

 nothing to do with the way in which the 

 water is employed, and self-interest will 

 force the farmer to exercise economy in 

 flooding his land. But even then precau- 

 tions must be taken to prevent him from 

 keeping his sluice open when it should be 

 shut. 



In Italy and in America water is gener- 

 ally charged by the module ; but in many 

 cases, where the country is very flat, the 

 water can not fall with a free drop out of 

 the sluice, and, as far as I know, no satis- 

 factory module has yet been invented for 

 delivering a constant discharge through a 

 sluice when the head of water in the chan- 

 nel of supply is subject to variation. These 

 are the conditions prevailing in the plains 

 of northern India, where there is a yearly 

 area of canal irrigation of about six mil- 

 lions of acres. The cultivator pays not in 

 proportion to the volume of water he uses, 

 but on the area he waters every crop, the 

 rate being higher or lower according as the 

 nature of the crop demands more or less 

 water. 



The procedure of charging for water is, 

 then, as follows: When the crop is nearly 

 ripe the canal watchman, with the village 

 accountant and the farmers interested, go 

 over the fields with a government official. 

 The watchman points out a field which he 

 says has been watered. The accountant, 

 who has a map and field-book of the village, 

 states the number and the area of the field 

 and its cultivator. These are recorded 

 along with the nature of the crop watered. 

 If the cultivator denies that he has received 

 water, evidence is heard and the case is 

 settled. A bill is then made out for each 

 cultivator, and the amount is recovered 

 with the taxes. 



This system is perfectly understood, and 

 works fairly well in practise. But it is not 



a satisfactory one. ,It holds out no induce- 

 ment to the cultivator to economize water, 

 and it leaves the door open to a great deal 

 of corruption among the canal watchmen 

 and the subordinate revenue officials. 



GOVERNMENT CONTROL OF V^ATER SUPPLY. 



Where the subject agricultural popula- 

 tion is unfitted for representative govern- 

 ment it is best that the government should 

 construct and manage the irrigation, on 

 rules carefully considered and rigorously 

 enforced, through the agency of officers 

 absolutely above suspicion of corruption or 

 unfair dealing. Such is the condition in 

 Egypt and in the British possessions in 

 India. Objections to it are evident enough. 

 Officials are apt to be formal and inelastic, 

 and they are often far removed from any 

 close touch with the cultivating classes. 

 But they are impartial and just, and I 

 know of no other system that has not still 

 greater defects. 



Even if the agricultural classes in India 

 were much better educated than they are, 

 it would still be best that the control of the 

 irrigation should rest with the government. 

 By common consent it is the government 

 alone that rules the army. Now the irri- 

 gation works form a great army, of which 

 the first duty is to fight the grim demon of 

 famine. Their control ought, therefore, to 

 rest with the government ; but the condi- 

 tions are very different when the agricul- 

 tural classes are well educated and well 

 fitted to manage their own affairs. 



Irrigation is too new and experimental 

 in America for us to look there for a well- 

 devised scheme of water control. The laws 

 and rules on the subject vary in different 

 states, and are often contradictory. It is 

 better to look at the system evolved after 

 long years in north Italy. 



THE ITALIAN SYSTEM. 



I have already alluded to the great 



