a less period that quantity neither will, not is it desirable
that it should, filter through an agricultural soil)--the interval between
two pipes will be called upon to pass two-thirds of a tablespoonful of
water per minute, and no more. Inch pipes, lying at a small inclination,
and running only half-full, will discharge more than double this quantity
of water in forty-eight hours."

Tiles may be made of any desired form of section,--the usual forms are the
"horse-shoe," the "sole," the "double-sole," and the "round." The latter
may be used with collars, and they constitute the "pipes and collars,"
frequently referred to in English books on drainage.

[Illustration: Fig. 13 - HORSE-SHOE TILE.]

Fig. 13 - HORSE-SHOE TILE.

_Horse-shoe tiles_, Fig. 13, are condemned by all modern engineers. Mr.
Gisborne disposes of them by an argument of some length, the quotation of
which in these pages is probably advisable, because they form so much
better conduits than stones, and to that extent have been so successfully
employed, that they are still largely used in this country by "amateurs."

"We shall shock some and surprise many of our readers, when we
state confidently that, in average soils, and, still more, in
those which are inclined to be tender, horse shoe tiles form the
weakest and most failing conduit which has ever been used for a
deep drain. It is so, however; and a little thought, even if we
had no experience, will tell us that it must be so. A doggrel
song, quite destitute of humor, informs us that tiles of this sort
were used in 1760 at Grandesburg Hall, in Suffolk, by Mr. Charles
Lawrence, the owner of the estate. The earliest of which we had
experience were of large area and of weak form. Constant failures
resulted from their use, and the cause was investigated; many of
the tiles were found to be choked up with clay, and many to be
broken longitudinally through the crown. For the first evil, two
remedies were adopted; a sole of slate, of wood, or of its own
material, was sometimes placed under the tile, but the more usual
practice was to form them with club-feet. To meet the case of
longitudinal fracture, the tiles were reduced in size, and very
much thickened in proportion to their area. The first of these
remedies was founded on an entirely mistaken, and the second on no
conception at all of