22 Proceedings of the Royal Society of Victoria. 



for the Lead sluices to silt up. The cost of raising water by 

 steam power is very great, necessitating a large initial 

 outlay and a constant fixed expenditure for maintenance, 

 Avhich bears a large proportion to the profit of irrigation, 

 and in case of partial failure of the crops, absorbs probably 

 the whole. The use of windmills for pumping no doubt 

 minimises the cost, but there is an element of uncertainty in 

 their use which it is desirable to eliminate as far as possible, 

 and consequently the suppl}'' of water should, wherever 

 practicable, be by gravitation. The supply of water from 

 wells (jccasions great cost in lifting, as a rule, and the su])ply 

 is small in comjjarison to the quantity required, and is 

 seldom profitable except for garden crops, or particularly 

 remunerative ones, such as chilies in India. 



The direction therefore to which irrigation schemes sliould 

 turn, is the regulation of the supply of water by storing it 

 before it I'eaches the rivers and watercourses, thus reducing 

 the risk of floods, and providing against deficiency in times of 

 di'ought. The distribution of water so stored is a matter for 

 subsequent consideration. In India, the periods of drought 

 are ordinarily about six months. Here, they are usually 

 about the same number of weeks ; the proportion of area 

 to storage required here is consequently much less If 

 storage is provided for the whole of the ai'ea under command, 

 the construction of channels for distribution may to a great 

 extent be postponed, as in most cases only a small proportion 

 of the irrigable land is under cultivation ; and, although 

 distribution through the natural channels is in most cases 

 wasteful, this waste might at first be disregarded, and the 

 channels could be constructed as more land comes under 

 cultivation and the necessity for economy arises, while the 

 increased area cultivated would enable the necessary funds 

 for the work to be provided. It would very frequently be 

 found also, that the first cost of storing the water, and thus 

 securing a thoroughly satisfactory system of irrigation 

 would be less, and it would very seldom exceed that of 

 constructing weirs or anicuts on rivers of au}'^ size. It does 

 not follow that weirs or anicuts are never desirable, but 

 there is little doubt that in most cases storage is preferable, 

 and that endeavours should be made to adopt this system 

 wherever practicable. 



The first steps to be taken are, to note the various 

 catchment basins, and to establish rain gauges in suitable 

 places. Considerable care and judgment is requisite in 



