ed, and in a measure realized, that
during the first moments of the operations, when the level of
the water in the dock was above the center of the runner of
the pumps, that the discharge would be proportioned to the
work done, where no effort was necessary to maintain a free
and full flow through the suction pipes; but as the level
passed lower and farther away from the center there was no
apparent diminution of the flow, and no noticeable addition
to the load imposed on the engine. The variation in piston
speed, noted during the trial, was probably due to the
variation of the boiler pressure, as it was difficult to
preserve an equal pressure, as it rose in spite of great
care, owing to the powerful draught and easy steaming
qualities of the boilers.

"After the trial of the second pump had been completed the
dock was again filled through the caisson, and as both pumps
were to be tried, the water was admitted to a level with the
23d altar, containing 7,317,779 gallons, which was seven feet
above the center of the pumps; this was in favor of the pumps
for the reasons before stated. In this case all the boilers
were used.

"Everything moved most admirably, and the performance of
these immense machines was almost startling. By watching the
water in the dock it could be seen to lower bodily, and so
rapidly that it could be detected by the eye without
reference to any fixed point.

"The well which communicates with the suction tunnel was
open, and the water would rise and fall, full of rapid swirls
and eddies, though far above the entrance of these tunnels.
Through the man hole in the discharge culvert the issuance
from the pipes could be seen, and its volume was beyond
conception. It flowed rapidly through the culvert, and its
outfall was a solid prism of water, the full size of the
tunnel, projecting far into the river.

"During a pumping period of 55 minutes, the dock had been
emptied from the twenty-third to two inches above the sixth
altar, containing 6,210,698 gallons, an average throughout of
112,922 gallons per minute. At one time, when the revolutions
were increased to 160 per minute, the discharge was 137,797
gallons per minute. This is almost a river, and is hardly
conceivable. After the pumps were stopped, on