In previous experiments this year the ironclad was attacked on the port
side, which had been specially strengthened for the occasion, and the
result was a victory for the defense. On June 13 the starboard side was
selected for attack, in order that a comparison might be instituted with
the effects produced under different conditions by a similar experiment.

Last year in the latter case the double bottom was filled with coal; and
after the charge, which was lashed against the ship in the same way, had
been exploded, it was found that the bilge keel had been shivered for a
length of 20 ft., while the lower plating had been much bulged above the
bilge keel. Four strakes of the skin plating extending up to the armor
shelf had also been forced inward and fractured where they crossed the
longitudinal frames. They had parted in the middle for a distance of 8
ft., while some of the butts had been opened so that gashes 2 in. or 3
in. wide appeared between them. The coal had been pulverized and
scattered in all directions, and other internal damage inflicted.
Nevertheless, the watertight bulkheads remained intact, and by confining
the influx of water to a single compartment so much buoyancy was
preserved that, though the ship heeled over to starboard and was maimed,
she remained afloat, and might have continued to fight her guns,
provided always that no injury had been sustained by her machinery, a
point which these experiments do not touch. Crippled, however, as she
was, it was thought at the time (and the probability was strengthened by
subsequent examination of the ship in dock) that the coal, instead of
being a protection to the double bottom, had in reality proved a source
of weakness by receiving the energy of the explosion from the outer
plating and communicating it to the inner plating, and so distributing
it throughout the submerged portions of the hulk.

The question was sufficiently important to demand an experimental