0 degrees 17 minutes West of Swan River.** The reef continues along the
eastern side of Barrow's Island, extending off three miles; our anchorage
was consequently little more than that distance from the shore. We
examined the northern and eastern sides; the former is composed of red
sandstone cliffy projections, separated by sandy bays, fronted for nearly
two miles by a coral reef, partly dry at low-water; but the south part of
the eastern side becomes very low; and where the cliffs end there is a
remarkable valley trending westwards. There were recent marks of the sea
many feet above the ordinary reach of the tides, bespeaking occasional
strong south-east winds. A number of stony-topped hills, from 150 to 200
feet in height, were scattered over the northern parts of the island. In
the valleys was a little sandy soil, nourishing the spinifex, and a
stunted kind of wood sufficiently large for fuel.

(*Footnote. Lowendal Island, bearing east, leads into it.)

(**Footnote. We recognised them from a sketch furnished by the Admiralty,
and made in 1719 by a Dutch sloop sent in search of them from Batavia.
They placed them eight degrees west from the coast of New Holland. If we
take leagues instead of degrees it would bring them near their actual
distance from the shore. Van Keulen says they were seen in the ship
Vaderland Getrouw, and found to be in 20 degrees 30 seconds south. In
1777 they were seen by Captain Joss, of the Danish ship Frederisberg
Castel, who places them in 20 degrees 40 minutes South. It was by his
description that I recognised them beyond a doubt, although his longitude
would place them thirteen degrees more to the westward, and near the
position they have occupied for years in the charts. The centre of them
bears North by East five miles and a quarter from Cape Dupuis, the
north-west point of Barrow's Island.)