et in length.

I proceeded to obtain this result without delay. The process was simple
and easy. Laying myself horizontally, I planted my feet against one of
the great ribs of the ship, and rested the end of the stick between
them. I now stretched myself out at full length, and guiding the rod so
as to keep it parallel to the axis of my body, I brought it across my
forehead, and beyond. With my fingers I could tell the point that was
opposite the crown of my head, and carefully marking this point, I
afterwards notched it with the knife. I now possessed a four-foot rule,
exact enough for my purpose.

But there were difficulties yet to be encountered. With a four-foot
rule, I was but little advanced towards my computation. I might make a
nearer approach to the measurement of my diameters, but that would not
avail. I must know them _exactly_. I must know them in inches, and
even fractions of inches; for, as I have said, an error of half an inch
in some of my _data_ would make a difference of gallons in the result.
How, then, was I to divide a four-foot stick into inches, and register
the inches upon its edge? How was this to be done?

It seems simple enough. The half of my four feet--already ascertained--
would give me two feet; and the half of that again would reduce the
standard to a foot. This again notched in the middle would make two
lengths of six inches each. Then I could subdivide those into lengths
of three inches, which, if not small enough for my calculation, could be
still further subdivided into three equal parts, each of which would be
the desired minimum of an inch.

Yes, all this seems easy enough in theory, but how was it to be put in
practice upon a piece of plain straight stick, and in the midst of as
perfect darkness as that which surrounds a blind man? How was I to find
the exact middle--for it must be exact--of even the four feet, much less
divide and subdivide till I got down to the inches?

I confess that I was puzzled for awhile, and had to pause and reflect.

Not very long, however; I was soon able to get over this trifling
obstacle.

The plan that first suggested itself was to cut a third piece of stick
of a little over two feet in length, which I could easily guess at
within a few inches. This I could apply alongside of my four-foot rule,
beginning at the end, and proceeding as if I was measuring the latter
with the former. Of course, on the first application, two lengths