420 TRANSACTIONS OF THE 



There is, of course, very much connected with the construction 

 and action of tile drains tliat could not be comprised within the 

 scope of a single article. My object has been merely to show the 

 fallacy of supposing tliat water enters the tile mainl}^ from above, 

 and that it is thereftn-e necessary to place loose material, such as 

 small stones aud leaves above tiiem. In making stona-drains, 

 which are verj^ liable to obstruction, it is often well to do so, but 

 with tile (properly laid) it never is necessary. Let us have a 

 reform in this matter, and thus cheapen the construction and 

 extend the use of tile- drains. 



Dr. Waterbui;y — In 1851 I purchased a meadow of about six 

 acres, consisting of two very different kinds of land. The upper 

 part of it was composed of what geologists term drift or loose 

 stone, with their corners worn off by attrition against each other 

 and deposited in a direction and inclination nearly uniform 

 during some uncertain ancient period. The lower part which 

 was not quite so large was composed of swampy ground under- 

 laid by clay, and was very much the most productive. A moun- 

 tain brook that crossed a corner of the upper part suggested to 

 me tlie idea of converting the dry hill side into a wet land like 

 the lower part, and thus rendering it equally fertile. Accord- 

 ingly, by means of sluices from the stream taken along the side 

 of the hill, at a downward inclination of about the half of one 

 degree, I managed to obtain a sufficient supply of water, but 

 when applied instead of wetting the soil generally I found it to 

 percolate almost directly down until it met the impervious lower 

 strata, running along which it made its appearance as nvunerous 

 springs at the upper edge of the naturally wet part of the meadow. 

 During the same and the subsequent season I had occasion to 

 subject some fifty acres or more of meadow land of clayey soil to 

 irrigation with in every case a beneficial result, in some cases the 

 annual growth of grass being more than quadrupled. From these 

 experiments I drew the conclusion that irrigation to be of prac- 

 tical value must be practiced on soils not too open, but which 

 have enough plasticity of composition to prevent too rapid filtra- 

 tion through them, and that when practiced on such soils as 

 nature dresses with water it is one of the cheapest and most effec- 

 tive means of improving them in fertility. 



The water that was supplied to this hill side during the two 

 years in which the experiment was conducted was like all surface 

 water, roilly, that is more or less charged with organic matter, 

 and yet after the filtration, when it made its appearance in the 

 springs, it was not only quite free from any such taste but it had 



