t was worm-eaten, and the natives gave him to
understand, that it had been driven ashore by the waves of the
sea; and we had their own express testimony, that they had got the
inconsiderable specimens of iron, found amongst them, from some place
to the eastward.

From this digression (if it can be called so) I return to the
observations made during our stay at Atooi; and some account must now
be given of their canoes. These, in general, are about twenty-four
feet long, and have the bottom, for the most part, formed of a single
piece or log of wood, hollowed out to the thickness of an inch, or an
inch and a half, and brought to a point at each end. The sides consist
of three boards, each about an inch thick, and neatly fitted and
lashed to the bottom part. The extremities, both at head and stern,
are a little raised, and both are made sharp, somewhat like a wedge;
but they flatten more abruptly; so that the two sideboards join each
other side by side, for more than a foot. As they are not more than
fifteen or eighteen inches broad, those that go single (for they
sometimes join them as at the other islands) have outriggers, which
are shaped and fitted with more judgment than any I had before seen.
They are rowed by paddles, such as we had generally met with; and
some of them have a light triangular sail, like those of the Friendly
Islands, extending to a mast and boom. The ropes used for their boats,
and the smaller cords for their fishing-tackle, are strong and well
made.

What we saw of their agriculture, furnished sufficient proofs that
they are not novices in that art. The vale ground has already been
mentioned as one continued plantation of _taro_, and a few other
things, which have all the appearance of being well attended to. The
potatoe fields, and spots of sugar-cane, or plantains on the higher
grounds, are planted with the same regularity; and always in some
determinate figure, generally as a square or oblong; but neither
these, nor the others, are enclosed with any kind of fence, unless
we reckon the ditches in the low grounds such, which, it is more
probable, are intended to convey water to the _taro_. The great
quantity and goodness of these articles may also, perhaps, be as much
attributed to skilful culture as to natural fertility of soil, which
seems better adapted to them than to bread-fruit and cocoa-nut trees;
the few which we saw of these latter not being in a thriving state,
which will sufficiently