144 THE RAISING OF THE DRY-DOCK DEWEY. 
nation of the deck and side walls. ‘The leaks could then be easily found, 
due to the fact that water was running in them and could be plainly felt 
by the divers. All such leaks were accordingly made water-tight, after 
which the pumps had no difficulty in removing the water. The 10-inch 
centrifugal pump mounted on the first cofferdam could alone remove the 
water. The pump on the after cofferdam did not operate satisfactorily 
when it had to lift water a greater height than about 16 feet. About 2,000 
tons of water were removed from the machinery spaces, the machinery 
deck being bare of water at the high end and containing about 10 feet of 
water on it at the lowend. After this, the centrifugal pump lost its suction, 
but the 6-inch duplex pump on the forward cofferdam and the pump from 
the Piscataqua on the third cofferdam were sufficient to keep the water 
down but could not remove all of it. A duplex pump with a 6-inch suction 
and a powerful horizontal pump with an 8-inch suction were then placed 
below through the second cofferdam and mounted on the side wall stiffeners 
and on the deck respectively in the dynamo room. These two pumps 
removed sufficient of the remaining water to clear the dock’s own central 
centrifugal pump for operation. At this time there were only about 600 
tons of water remaining in the machinery spaces, the deck being bare to 
almost the middle of the dock. Steam was then connected to the dock’s 
forward pump and the dock’s middle pump through flexible piping from the 
Wompatuck into the main steam line of the dock which was blanked off 
just abaft the central pump. It was considered desirable and a safeguard 
to have two of the dock’s own pumps ready for operation, so that the work 
would not have to be stopped in case one pump should breakdown. The 
exhaust piping from the central pump was led up thorugh the cofferdam 
above it to the atmosphere. In the meantime, while this work was in 
progress, compressed air was being pumped into the after end pontoon 
through five manholes on the low side. The pumping of the air was con- 
tinued for 36 hours, at the expiration of which time the starboard after end 
had lifted about 3 feet and the port after end about 4 inches. The pumping 
of the compressed air was then stopped. This was 38 hours before the 
dock’s pump was started for finally lifting the dock. During these 38 
hours some of the air escaped through leaky vent pipes, but the amount 
was small compared to the total quantity introduced. The after end of 
the dock was therefore almost afloat at the time the dock’s own pump was 
started for raising. By this time the port side of the dock had sunk about 
4 feet farther into the mud than it was at the time the dock sank. The 
floating or lifting of the starboard side always forced the port side farther 
into the mud so that wood extensions had to be built on two of the coffer- 
