the north-westward,
with thick weather and a descent more than usually rapid in the marine
barometer, a stretch off shore should immediately be made, to prepare for
a south-west gale. A look-out must be kept for an island lying to the
west-south-west of Cape Northumberland; it was seen by Mr. Turnbull,
commander of the Britannia, south whaler, but the weather being thick,
its situation was not well ascertained. According to the best information
I could procure, this island lies in 381/4 deg. south, and about 1391/2 deg. or 3/4
east longitude.

From Kangaroo Island, a straight course may be made for the southernmost
part of the Archipelago of the Recherche; but should the winds come from
the westward and not blow a gale, or be light and unsteady, I would steer
more northward, nearer to the land, in the hope of having them more
favourable. From the Archipelago to Cape Leeuwin it seems best to keep at
a distance from the land, unless under the necessity of stopping in that
neighbourhood; for the current runs strong near the shore, and with the
advantage of an offing of twenty or thirty leagues, a ship may lie clear
of the cape with a wind which might otherwise keep her beating for many
days.

There appears to be no place of shelter against western gales, between
Bass' Strait and Kangaroo Island; but there are then, besides various
anchorages under that island, the bays and coves at the entrance of
Spencer's Gulph; and further westward, Coffin's Bay, Petrel Bay in the
island St. Francis, and Fowler's Bay near the head of the Great Bight.
Afterwards come Goose-Island Bay, Thistle's Cove, and the lee of
Observatory Island, all in the Archipelago of the Recherche; the cove
cannot be entered in a gale, but when once secured in the south-west
corner, a ship will be safe; the other two places afford very indifferent
shelter from strong winds, and are indeed fit only for a temporary
anchorage in moderate weather. Doubtful-Island Bay and King George's
Sound afford complete shelter against western gales; but some little time
would be lost in getting out of them, if a ship waited until an eastern
wind set in. Some account of all these places will be found in the
preceding pages of this volume; with the exception of Observatory Island,
for which D'Entrecasteaux's voyage may be consulted.

APPENDIX.

Account of the observations by which the Longitudes of places on the
south coast of Terra Australis have been settled.

The lunar