HYDROLOGY OF N EW YORK 



831 



The consumption of water from a navigable canal may be taken 

 to include the following items : 



1) For filling the prism, in case it is emptied for any reason. 

 '2 i Quantity required for lockages. 



3) The supply for replacing water lost by evaporation. This 

 head may be also taken to include the loss by percolation and 

 absorption by subsoil and aquatic plants. 



4) Loss by leakage as at aqueducts, culverts, lock gates, 

 valves, etc. 



5) Loss by wastage at spillways. 



6) Water required for power to operate lock gates and for 

 flushing out boats, barges and timber rafts, as well as for power 

 to operate electric lights at the locks during the nights. 



7) Quantity required for industrial and agricultural use. 



8) Losses by evaporation, percolation, etc. along the feeder. 

 This latter quantity, if the feeder is of considerable length, may 

 be large and can not be safely neglected in an estimate as to 

 water supply. 



There is no specific rule for determining water supplies for 

 canals. One chief source of loss is percolation, the determination 

 of which, in any particular case, is a matter of judgment, based 

 on experience. In any case we may assume much less loss with 

 good construction than with poor. The safest way to proceed is to 

 apply information derived from well attested experiments. 



Table No. 96 gives measurements and estimates of loss of 

 water from canals in Xew York State by evaporation, percola- 

 tion, waste, etc. Many of these measurements have been re- 

 ferred to in the preceding. 



In connection with the barge canal work, a number of gagings 

 were made at various points along the Erie canal, as at Lock- 

 port. Boonville, Glens Falls and Rochester. Current meter and 

 rod observations were also made at Cornell L^niversity. It is 

 stated in the barge canal report: 



Much time was spent in attempting to find a number of fair 

 comparisons in the results of the canal gagings made last sum- 

 mer (1900), but unfortunately the geological and topographical 

 conditions of the levels, or sections, were not sufficiently similar 

 to justify the acceptance of any expressions deduced therefrom. 



