88 C 32 ] 
ascertain its sufficiency, we must examine next wliat those losses may 
amount to. 
Tnc length of this section is 144 miles — supposing it 5 feet deep, 28 
feet broad at the bottom, and 44 Icet at the surface of the water, the 
prism of its capacity, will have a base of 20 cubic yards, on a length 
of l4f miles, equal to a cubic of 519,200 cube yai'ds. This will be 
filled in the first days of March, without deranging the economy of 
water which we luue just analysed. We have already observed, 
that Deep Creek may supply every month a cube neasly corres- 
ponding to this or 492,034 cubic yards, at the minimum rate, and 
lowest state of its flow: we must now examine whether this supply 
will suffice every month to the filtrations and evaporations of 14| miles 
of Canal. - 
Without entering into minute calculations, which pi-opcrly belong 
to the report accompanying the final project of the canal, we will 
state generally the most positive results which experience has given 
as to the joint amount of filtrations and evaporations. Having as- 
certained that no experiments of this nature have been tried on the 
Erie Canal, where the supply of water was found evidently more 
than sufficient, we were obliged to consult the results of those canals 
constructed in Europe, under a climate, which, in summer, comes 
nearest to our own. We have selected for tfiis purpose, the canal of 
Narbonne, in the south of France. Narbonne and Baltimore, com- 
pai'ed as to climate and rain, are as follows: 
Narbonne, lat. N. 43° 11' fmean, greatest heat, 95°") Mean quanti- 
(Irora observations made •< mean temperature, 60° I ty of rain, 
duriwg twenty years.) (.mean, greatest cold, 24° J 29^^ in. 
Baltimore, lat. N. 59° 17' f ^ . , , ^, ,^ o"^ t,, 
(from observations made' '"^an, greatest heat, 94 VV Mean quan- 
1817—18:2, by Mr.i '"*''*" *''"'f^^^^^"''^' ^^tVo J> tityofram, 
LeNvisBrantz,of]VId. [mean, greatest cold, 0-,V/j SS^VV i". 
Of all sucJi works, the canal of Narbonne has given most trouble to 
its Engineers, from its excessive filtrations and los,s of water in the 
gravelly soil through which it is run. It is a branch from the canal 
of Langiiedoc to the City of Narbonne, three miles in length. As soon 
as it was opened, in 1788, it lost the v^lue or contents of its prism in 
a few days, and overflowed the surrounding country; in 1789, it still 
lost the value of its prism in 6 days; and in 1800. it lostit in 18 days, 
or the value of its prism, and | every month; (16| times its contents 
in 10 months navigation.) This evaluation is the result of careful and 
accurate observations: and, considering the climate and soil through 
Avhich this canal runs, it may fairly be taken as a specimen of the 
maximum loss, which a canal can suffer, through filtrations and evapo- 
rations. 
The ground through which rujis tlie western branch of our middle 
section, is of a quality far superior to the country through wl)ich 
runs the Narbonne Canal. It is, for 6i miles, of an exrollent quality; 
the remaining 8i miles run througli a rugged aiul rockvsoil, i;ut clav 
