e steamer had disappeared among the
islands of the bay.

"No, sir; but Mr. Fluxion told him to make it east-north-east."

"Very well; but the masters should do this duty," added Mr. Lowington,
as he directed the instructor in mathematics to require the masters, to
whom belonged the navigation of the ship, to indicate the course.

William Foster was called, and sent into the after cabin with his
associates, to obtain the necessary sailing directions. The masters had
been furnished with a supply of charts, which they had studied daily, as
they were instructed in the theory of laying down the ship's course.
Foster unrolled the large chart of the North Atlantic Ocean upon the
dinner table, and with parallel ruler, pencil, and compasses, proceeded
to perform his duty.

"We want to go just south of Cape Sable," said he, placing his pencil
point on that part of the chart.

"How far south of it?" asked Harry Martyn.

"Say twenty nautical miles."

The first master dotted the point twenty miles south of Cape Sable,
which is the southern point of Nova Scotia, and also the ship's
position, with his pencil. He then placed one edge of the parallel
ruler on both of these points, thus connecting them with a straight
line.

A parallel ruler consists of two smaller rulers, each an inch in width
and a foot in length, connected together by two flat pieces of brass,
riveted into each ruler, acting as a kind of hinge. The parts, when
separated, are always parallel to each other.

Foster placed the edge of the ruler on the two points made with the
pencil, one indicating the ship's present position, the other the
position she was to obtain after sailing two or three days. Putting the
fingers of his left hand on the brass knob of the ruler, by which the
parts are moved, he pressed down and held its upper half, joining the
two points, firmly in its place. With the fingers of the right hand he
moved the lower half down, which, in its turn, he kept firmly in place,
while he slipped the upper half over the paper, thus preserving the
direction between the points. By this process the parallel ruler could
be moved all over the chart without losing the course from one point to
the other.

On every chart there are one or more diagrams of the compass, with lines
diverging from a centre, representing all the points. The parallel ruler
is worked over the chart to one of these diagrams, where the direction
to which it has been set nearly or exa