9° 15' 20" east longitude. The watch at the same time gave 9° 36' 45".
Soon after the variation was found to be 13° 10' west. It is nearly in
this situation that Mr Bouvet had 1° east. I cannot suppose that the
variation has altered so much since that time; but rather think he had
made some mistake in his observations. That there could be none in ours
was certain, from the uniformity for some time past. Besides, we found
12° 8' west, variation, nearly under this meridian, in January 1773.
During the night the wind veered round by the N.W. to N.N.E. and blew a
fresh gale.

At eight in the morning of the 19th, we saw the appearance of land in
the direction of E. by S., or that of our course; but it proved a mere
fog-bank, and soon after dispersed. We continued to steer E. by S. and
S.E., till seven o'clock in the evening, when being in the latitude of
54° 42' S., longitude 13° 3' E., and the wind having veered to N.E., we
tacked and stood to N.W. under close-reefed topsails and courses; having
a very strong gale attended with snow-showers.

At four o'clock next morning, being in the latitude of 54° 30' S.,
longitude 12° 33'. east, we tacked and stretched to N.E. with a fresh
gale at S.W., attended with snow-showers and sleet. At noon, being in
the latitude of 54° 8' S., longitude 12° 59' E., with a fresh gale at W.
by N., and tolerably clear weather, we steered east till ten o'clock in
the evening, when we brought-to, lest we might pass any land in the
night, of which we however had not the least signs.

At day-break, having made sail, we bore away E., and at noon observed in
latitude 54° 16' S., longitude 16° 13' east, which is 5° to the east of
the longitude in which Cape Circumcision is said to lie; so that we
began to think there was no such land in existence. I however continued