A?7>* erly wind prevails as far as io Or ia° S. and the wefterly as
>-—.--- 1 far as 6 or 8” ; in the intermediate fpace the winds are vari-
Fnday IJ' able, and the air, I believe, always unwholefome; it certainly
aggravated the difeafes which we brought with us
from Batavia, and particularly the flux, which was not in
the Ieaft degree checked by any medicine, fo that whoever
was feized with it, confidered himfelf as a dead man; but
we had no fooner got into the trade-wind, than we began to
feel its falutary effects; we buried indeed feveral of our
people afterwards, but they were fuch as had been taken
on board in a ftate fo low and feeble that there was fcarcely
a poflibility o f their recovery. At firft we fufpefted that this
dreadful diforder might have been brought upon us by the
water that we took on board at Prince’s Ifland, or even by
the turtle that we bought there ; but there is not the leaf!
reafon to believe that this fufpicion was well grounded, for
all the fhips that came from Batavia at the fame feafon, fuf-
fered in the fame degree, and fome of them even more-
feverely, though none of them touched at Prince’s Ifland in
their way.
A few days after we left Java, we faw boobies about the;
fhip for feveral nights fucceflively, and as thefe birds are
known to rooft every night on fhore, we thought them an
indication that fome ifland was not far diftant ; perhaps it
might be the ifland of Selam, which, in different charts, is
very differently laid down both in name and fituacion.
The variation of the compafs off the weft coaft of Java is
about 30 W. and fo it continued without any fenfible variation,
in the common track of fhips to the longitude of 288°:
W. latitude 22 S. after which it increafed apace,’ fo; that in
longitude 295°, latitude 230, the variation was io° 20' W.: in
feven degrees more of longitude, and one of latitude, it increafed
creafed two degrees ;■ in the fame fpace, farther to the weft, '77'■
it increafed five degrees : in latitude 28°, longitude 3140, it _
was 24?, 20', in latitude 29° longitude ,317°,■ it was 26° io', Fnday 15‘
and was then flationary for the fpacé of about ten degrees
farther to the weft; but in latitude 340, longitude 3330, we
obferved it twice to be 28°i W. and this was its greateft variation,
for,in latitude 3s ° i, longitude 3370, it was 24°, and
continued gradually to decreafe; fo that off Cape Anguillas
it was 220 30', and in Table Bay 20° 30' W.
As to currents it did not appear that they were at all con-
fiderable, till we came within a little diftance of the meridian
of Madagafcar ; for after we had made 328 of longitude from
Java Head, we found, by obfervation, that our error in longitude
was only two degrees, and it was the fame when we
had made only nineteen. This error might be owing partly
to a current fetting to the weftward, partly to our not making
proper allowances for the fetting of the fea before which we
run, and perhaps to an error in the affumed longitude of
Java Head. If that longitude is erroneous, the error muft be
imputed to the imperfection of the charts of which I made
ufe in reducing the longitude from Batavia, to that place,
for there can be no doubt but that the longitude of Batavia
is well determined. After we had palfed the longitude of
307°, the effects of the wefterly currents began to be con-
fiderable} for in three days, our error in longitude was
1° 5': the velocity of the current kept increafing, as we-proceeded
to the weftward, in fo much that for five days fucceflively
after we made the land, we were driven to the S.W.
or S.W. by W. not lefs than twenty leagues a day; and this
continued till we were within fixty or feventy leagues of thé
Cape, where the current fet. fometimes one way, and fome-
times the other, though inclining rather to the. weftward. •
3 After