290 TKANSACTIONS OF THE 



fies the air, as it is well known that shallow, stagnant water, and water 

 mechanically sejjarated, as it is in the soil, is far more susceptible of evapo- 

 ration than when in motion, as a creek, or combined as a pond ; and that 

 evaporation is invariably accompanied with a diminution of the tempera- 

 ture from objects from which it takes place. When water is evaporated,, 

 each drop divides itself into a thousand parts, which are borne up by the 

 friction of atmosphere. Drainage keeps water in constant motion, and 

 removes it so far below the surface as to prevent the sun from acting upoQ 

 it, which of necessity prevents excessive evaporation with its usual reduc- 

 tion of temperature. 



And if ammonia and other gaseous substances, the result of vegetable 

 and animal decomposition, are, as they are believed to be, the origin of 

 miasma, and the source of fevers, their retenti&n near the surface will be 

 great in proportion to the amount of water. 



In my practice,, by way of experiment^ I have used cylindrical pipes 

 fourteen inches long,, of several sizes, adapted to the quantity of water to- 

 be removed, and joined together at the junction by a collar, which centres 

 each pair of pipes, and maintains the line of the bore. I have likewise 

 placed small stones in the bottom of the drain, and on these arranged 

 cylindrical pipe tiles of two inch bore. I have also tried the horse-shoe 

 tile with sole, and the pipe tile of two inch tube without collars, which lat- 

 ter would run a much larger quantity of water in a stated time than the 

 horse-shoe tile, and carry with it, to its terminus, mere earthy matter taken 

 in at the junctions. Pipes are one-third less weighty than tiles with soles,, 

 and cost less at the tilery. I would therefore give them the preference,, 

 next to stone, of all other materials for general drainage. 



Open or horse-shoe tiles should always be laid on flat tile soles or slate. 

 Pipe tiles of less than 1^ inches should be laid with collars^ and for large 

 aizes, scoop out the bottom of the drain to fit the periphery of the tile, or 

 an unobstructed run of water will not be secured. Tiles ought always to 

 be laid by a man in whom you can place unbounded confidence, with strict 

 orders on no account to hurry the work, but to lay them perfectly even and 

 firm. 



^^«t \\T^Vil: 



Drains 12 feet apart require for an acre • 



12 inch tilo. 15 inch. 



12 feet apart, 3,630 2,904 



18 " 2,420 1,938 



21 " 2,074 1,659 



30 " 1,452 1,161 



36 " 1,210 968 



