97 t '"^2 ] 
The inquiry, liow much water a canal requires in skuations and 
seasons exposed to those causes of evaporation and soakage which 
are active, and, in a degree, uncertain, is answered with no lif.tle 
difficulty, even from experience. Take, for exain[)le, the five most 
complete and noble canals in France, the country whose climate 
most resembles our own; and take, also, Eric canal, in its middle 
and most unexceptionable portion, we find tliey dififer from each other, 
probably being through a rough country, ii» some instimccs, and con- 
structed with more or less precaution. We know of no better rule 
than the average of these canals, viz. 
De Briare, 6 ra. 40ft. wide, 4 ft. deep, see 1'28 it. usee 62.60 pr. m, pr. niin. 
' 375- 
' 215 
232 
' 375 
' 136 
Lanci-uedoc, 69 
64 
« 
7i 
Du Centre, 69 
47 
«( 
H 
St. Qiiintin 32 
54 
et 
^ 
Narbonae 61 
64 
(C 
7^ 
iJl-ie, 141 
40 
<c 
4 
lock;l^'e5.80 56 80 
35.90 1 
ockHg-e 17 18 90 
55.60 " 
9.74 45.86 
70.70 " 
•<• 10.52 60, 5S 
68.80 " 
" 17 51.80 
121 
" 6.16 114.84 
414 
6)348.78 
69 
6 
pi- m. p 
„.^- ro 1"? evap. or 
rmni. 58. lo > ,' 
Eor evaporation, soatage and lockage, 69 feet. 
But we do not know how much was used in the lockage, except in 
the instance of Languedoc, on which the average of six years was 
1920 boats, its locks being 100 f. 20 f wide, 8 feet dcep= 16000 f x 
1920 = 30,720,000 cubic feet-T-365 = 84l 644 per day, 12 hours is 
J 169 per minute-T-69 miles= 17 feet per mile per minute nearly. JNow, 
assuming that the others have proportional lockage, and that the 
size of their locks are in proportion to their size, or section, the de- 
duction must be made as above for their lockage, respectively, leaving 
the nett expense of water for evaportion and filtration; the average 
of which is seen to be 58.13 per mile per minute; this average is 
more favorable than the experience of Erie alone, as it has been stat- 
ed by one of the Engineers on that work, we do not know from what 
data, or with what correctness, not having had yet an opportunity of 
personal verification thereof. 
That 61 miles from Rochester to Seneca receives 6000 feeti 
1 1 " the Camillus Level, *' - * 2000 " 
69 " the Rome Summit *« - 9000 « 
141 17000 
Or per mile per minute 120 
.57 
Applying this experience to the second section and the size of the 
locks in contemplation there, and assuming the summit and adjoining 
levels to be from the forks of the Conemaugh to the Tories of the Ju^ 
niatta 28 miles, the calculation would stand tiius: 28 x 58.82=1627. 
64 ]per minute per mile, for evaporation and filtration, the locks 80 
13 
