Having hauled the wind to the starboard tack, and spent the night plying, on the 14th, at sun-rise, the island of Balabea bore S. 6 E., and the land seen the preceding night W., but the reef still trended N.W., along which we steered with a light breeze at E.S.E. At noon we observed in latitude 19° 28', longitude from Observatory Isle 27' W. We had now no sight of Balabea; and the other land, that is, the N.W. part of it, bore W. by S. 1/2 S., but we were not sure if this was one continued coast, or separate islands. For though some partitions were seen, from space to space, which made it look like the latter, a multitude of shoals rendered a nearer approach to it exceedingly dangerous, if not impracticable. In the afternoon, with a fine breeze at E.S.E., we ranged the outside of these shoals, which we found to trend in the direction of N.W. by W., N.W. by N., and N.N.E. At three o'clock we passed a low sandy isle, lying on the outer edge of the reef, in latitude 19° 25', and in the direction of N.E. from the north-westernmost land, six or seven leagues distant. So much as we could see of this space was strewed with shoals, seemingly detached from each other; and the channel leading in amongst them appeared to be on the S.E. side of the sandy isle; at least, there was a space where the sea did not break. At sun-set we could but just see the land, which bore S.W. by S., about ten leagues distant. A clear horizon produced the discovery of no land to the westward of this direction; the reef too trended away W. by N. 1/2 N., and seemed to terminate in a point which was seen from the mast-head. Thus every thing conspired to make us believe that we should soon get round these shoals; and with these flattering expectations we hauled the wind, which was at E.N.E., and spent the night making short boards.
Next morning at sun-rise, seeing neither land nor breakers, we bore away N.W. by W., and two hours after saw the reef extending N.W. farther than the eye could reach; no land was to be seen. It was therefore probable that we had passed its N.W. extremity; and, as we had seen from the hills of Balade its extent to the S.W., it was necessary to know how far it extended to the east or southeast, while it was in our power to recover the coast; for, by following the direction of the shoals, we might have been carried so far to leeward as not to be able to beat back without considerable loss of time. We were already far out of sight of land; and there was no knowing how much farther we might be carried, before we found an end to them. These considerations, together with the risk we must run in exploring a sea strewed with shoals, and where no anchorage, without them, is to be found, induced me to abandon the design of proceeding round by the N.W., and to ply up to the S.E., in which direction I knew there was a clear sea. With this view we tacked and stood to the S.E., with the wind at N.E. by E., a gentle breeze. At this time we were in the latitude of 19° 7' S., longitude 163° 57' E.
