Stout—Vegetation of a Typical Wild Hay Meadow. 443 
experiments were taken up and lienee in view of the problems 
involved in the growth of more valuable forage plants on these 
marsh meadows, the following species were used for the tests: 
Alsike clover (T. hybridum) ; red-top (Agrostis alba) ; timothy 
(Phleum pratense); and Calamagrostis canadensis . The lat¬ 
ter as has been shown is the most important grass of the marsh 
meadow. The other species are more or less in use in the at¬ 
tempts to improve marsh meadows. 
Tor the experiments six galvanized iron cylinders three feet 
in diameter and four and one-half feet in height were selected 
from the soil cylinders in use in experimental work carried on 
in the greenhouse of the Department of Soils of the Agricul¬ 
tural College. These particular cylinders had previously been 
filled with muck soil obtained from marsh land and hence closely 
duplicated the soil conditions of the marsh meadow in question. 
Strips of galvanized iron sheeting were set edgewise into the 
soil in such a manner as to divide the surface area of each 
cylinder into four equal segments. A cylindrical basket 
made from eight-mesh galvanized wire screening was placed 
in the center of each of these areas. These baskets were 
six inchs in diameter and from two to two and one-half 
feet in length. In sinking the baskets the soil was carefully 
sifted and gently tamped in and around each basket to prevent 
checks or pockets. Thus there was provided a total of 24 bas¬ 
kets, 4 in each cylinder and each basket imbedded in the cen¬ 
ter of one quarter of the surface of the cylinder. 
The bottom of the cylinders were covered with several inches 
of broken tile which formed a sort of reservoir beneath the dirt. 
To regulate the height of the water table, glass tubes were 
fitted by means of rubber corks into a tap connection at the bot¬ 
tom of the cylinder. The glass tubes were arranged to stand 
erect at the side of the cylinders and hence to act as water 
gauges. The height of the tube attached to a cylinder deter¬ 
mined the position of the water table within that cylinder for 
any excess would flow from the top of the gauge until the level 
sank to the height of the gauge. 
The cylinders were prepared between Eeb. 7 and 14, 1910. 
