the North-East, which, after passing over us, returned again from the
westward with more rain but less wind.

October 22.

At daylight (22nd) we saw the Island of Rottee, but instead of being,
according to our account, to windward of it, we were very little to the
eastward of its south-west end; having been set forty-three miles to the
westward since yesterday noon. During the day, as the wind was at
South-East, we endeavoured to pass round its windward side, but the
current was setting with such strength to the westward that, finding we
had lost ground during the night, we bore up the next morning for the
island of Savu, a proceeding which, if we should succeed in procuring
refreshments and fresh water there, would be more advantageous than going
to Timor: for in the first place there was less chance of incurring
sickness among the crew; and secondly we should be farther advanced on
our voyage back. Captain Cook, on his visit to Savu in 1770, found a
Dutch resident there; and I recollected having been assured by Mr.
Hazaart, the Resident at Timor, that the people were well-disposed
towards the English: Captain Horsburgh also mentions in his description
of Savu that the Dutch have residents on all these islands; and, as a
corroboration of these accounts, I had been informed by the master of a
merchant schooner at Port Jackson, who had lately been among these
islands, that abundance of good water could be procured there. Opposed to
this last report, Captain Cook says, "We were upon the coast at the
latter end of the dry season (September), when there had been no rain for
seven months, and we were told, that when the dry season continues so
long, there is no running stream of fresh water upon the whole island,
but only small springs, which are at a considerable distance from the sea
side:"* this conflicting account was discouraging; but as we had lately