t meant delay. The _St. Paul_ was therefore ordered to Key West,
instructions being telegraphed there to coal her full immediately on
arriving. She would there be as near the Windward Passage as Curacao
is, and yet able, in case of necessity, to proceed by the Yucatan
Passage or in any direction that might meanwhile become expedient. It
may be added that the _St. Paul_ reached Key West and was coaled ready
for sea by the evening of May 18th, four days from the time she left
Hampton Roads, a thousand miles distant.

While on her passage, the Department had entertained the purpose of
sending her to the Gulf of Venezuela and adding to her the _Harvard_
and the _Minneapolis_, the object being not only to find the enemy, if
there, but that one of the three should report him, while the other
two dogged his path until no doubt of his destination could remain.
Their great speed, considered relatively to that which the enemy had
so far shown, gave reasonable probability that thus his approach
could be communicated by them, and by cables, throughout the whole
field of operations, with such rapidity as to ensure cornering him at
once, which was the first great essential of our campaign. A cruiser
reporting at Cape Haitien was picked up and sent to the _Minneapolis_,
whose whereabouts was sufficiently known, because circumscribed, and
she received her orders; but they served only to develop the weakness
of that ship and of the _Columbia_, considered as cruisers. The coal
left after her rapid steaming to her cruising ground did not justify
the further sweep required, and her captain thought it imperative to
go first to St. Thomas to recoal,--a process which involved more delay
than on the surface appears. The bunkers of this ship and of her
sister, the _Columbia_, are minutely subdivided,--an arrangement very
suitable, even imperative, in a battleship, in order to localize
strictly any injury received in battle, but inconsequent and illogical
in a vessel meant primarily for speed. A moment's reflection upon the
services required of cruisers will show that their efficiency does not
depend merely upon rapid going through the water, but upon prompt
readiness to leave port, of which promptness quick coaling is a most
important factor. This is gravely retarded by bunkers much subdivided.
The design of these two ships, meant for speed, involves this lack of
facility for recoaling. There is, therefore, in them a grave failure
in that unity