New York AGRICULTURAL EXPERIMENT STATION. 189 
objectionable, as the adhesion of the soil to the inner wall of the 
cylinder became so great that it was found almost impossible to 
settle the latter through the last few inches of its descent. To 
‘avoid this difficulty in the longer cylinder, a hoop of bar-iron was 
made, of which the inside diameter was one-fourth of an inch 
less than that of the cylinder. This was attached to the end of 
the cylinder that was to enter the soil, by means of small iron 
brackets riveted to thethoop and fastened to the ends of the staves 
by strong screws. This hoop, being slightly smaller in diameter 
than the cylinder, regulated the size of the column of soil and 
prevented friction against the inner wall. It also furnished a 
convenient guide in dressing down the outside of the column. 
With this arrangement no additional weight was needed 
to force the cylinder down as fast as the excavation progressed. 
The cylinder, having been sunk to the desired depth, was 
inclined to one side sufficiently to break the column of soil at the 
lower end, when the edge of the plank bottom was inserted into 
the fissure thus formed. By means of jackscrews set between the 
cylinder and the back of the pit, the cylinder was then crowded 
over upon its bottom, the latter being held in place by bracing it 
from the opposite bank. This accomplished, the bottom was 
temporarily fastened in place by screwing strong angles of band- 
iron to the sides of the cylinder, the lower arm of the angle being 
first driven beneath the bottom. A temporary head was now put 
on the top to keep the soil from spilling out during the removal, 
after which the cylinder was tipped until its top lay against the 
bank of the excavation. It was then lifted with jackscrews until 
by means of skids, a rolling chain hitch and a team, it could be 
rolled out of the excavation. It was next loaded upon a truck- 
wagon and hauled to the pit prepared for it, where it was unloaded 
with the end intended to enter the soil upward. The plank bot- 
tom was then removed, and six inches of the earth taken out to 
make room for the layer of sand, shown at B in the drawing, upon 
which the column of soil rests. This accomplished, the drain 
tube C was inserted through a hole in the side of the cylinder and 
securely soldered to its place. The sand, which had previously 
been washed and dried, was then poured in, and the top care- 
fully leveled, after which the copper bottom D was inserted and 
