Old Croton (1899) clean | 122.8 | 79,400,000| .0133 ” ” ordinary | | | condition; | | | not clean | .. | 73,300,000| ” ” not clean | .. | 85,600,000| Dorchester Bay tunnel | | | ” ” | .. | .. | .014 Wachusetts, new; probably | | | clean (approx.) | | | Sudbury, clean | | | ” ” | | | ” ” | | | ” ” | | | ” ” | | | --------------------------+----------+-----------+-------- † From report by J. R. Freeman to B. S. Coler, 1899. ‡ From report of New York Aqueduct Commission. =190. Flow of Water through Large Closed Pipes.=—The masonry conduits to which consideration has been given in the preceding paragraphs carry water precisely as in an open canal, but the closed conduits or pipes of steel plates and cast-iron, like the Hemlock Lake conduit at Rochester and the East Jersey conduit of the Newark Water-works, are of an entirely different type, as they carry water under pressure. Hence the slope or sine of inclination _s_ belongs to the hydraulic gradient rather than to the grade of the pipe itself. Where the pipe-line is a long one its average grade frequently does not differ much from the hydraulic gradient, but the latter quantity must always be used. As in the case of the masonry conduits, the coefficient _c_ in Chezy’s formula will vary considerably with the degree of roughness of the interior surface of the pipe, with the slope _s_, and with the mean radius _r_. An important distinction must be made between riveted steel pipes and those of cast-iron, for the reason that the rivet-heads on the inside of the former exert an appreciable influence upon the coefficient _c_. The rivet-heads add to the roughness or unevenness of the interior of the pipe. Table XV gives the elements of the flow or discharge in the two pipe-lines which have been taken as types, as determined by actual measurements; it also exhibits similar elements for timber-stave pipes, to which reference will be made later. [Illustration: CROTON AQUEDUCT IN EARTH.]