eter.

The passage between 2 and 3 is safe, and has seven and eight fathoms: the
north-west side of 3 is of rocky approach, but the opposite side of the
strait is bold to; the anchorage is tolerably good. The Mermaid drove,
but it was not considered to be caused by the nature of the bottom, which
is of soft sand, and free from rocks.

The channel between 1 and 2 appeared to be very rocky, and shoal: between
1 and the reef r there is probably a clear channel of about a mile wide:
the north-east end of 1 has a reef which extends off it for half a mile.

(*Footnote. Many shoals, partly dry, occupy the space to the northward
and eastward of Howick's Group. Roe manuscript.)

All the islands are low and wooded, and surrounded by a coral reef of
small extent.

4 has a small islet off its west end.

5, 8, and 9 did not appear to have any reefs projecting from them. 7 is
probably two islands, with a reef extending for half a mile on its
western side. 6 is of larger size than the generality of the low islands
hereabout, Number 1 excepted: its centre is in latitude 14 degrees 28
minutes, and longitude 144 degrees 45 minutes. The position of Number 10
was not correctly ascertained.

The peak of CAPE BOWEN is in latitude 14 degrees 34 minutes, and
longitude 144 degrees 35 minutes 40 seconds.

NOBLE ISLAND is a rock, having a sandy, or a coral beach at its
north-west end; although small it is very conspicuous; and, when first
seen from the southward, has the appearance of a rock with a double
rounded top.

The REEFS s, t, and u are unconnected; the north end of s, lying six
miles and a half due east from Point Barrow, was dry for a considerable
extent; t, one mile to the north, was covered; but there is a dry sandy
key on u, bearing from Point Barrow, North 32 degrees East, six miles:
some rocks showed themselves above the water off its south end.

v and w may possibly be connected; the former was noticed to extend for
three miles, and the latter for nearly ten miles; there was, however, a
space of three miles between them, where a channel may possibly exist.
The channels between t and u, and between v and w, appeared to be clear
and deep.

The REEFS x, y, and Z, are probably parts of the barrier reefs, for the
sea was breaking very heavily upon their outer edge; there were, however,
considerable spaces where no breakers appeared, some of which, being
three or four miles wide, may possibly be as many outlets to sea.

N