From the western extremity of this island, we steered N. 82 W. and at noon on the 16th, were in latitude 14 deg.28’S. longitude 156 deg.23’W. the variation being 7 deg.40’E.  The wind was now easterly, and we had again the same mountainous swell from the southward that we had before we made the Islands of Direction, and which, from that time to this day, we had lost:  When we lost that swell, and for some days before, we saw vast flocks of birds, which we observed always took their flight to the southward when evening was coming on.[38] These appearances persuaded me that there was land in the same direction, and I am of opinion, that if the winds had not failed me in the higher latitudes, I should have fallen in with it:  I would indeed at this time have hauled away to the southward, and attempted the discovery, if our people had been healthy, for having observed that all the islands we had seen were full of inhabitants, I was still more confirmed in my opinion; as I could account for their being peopled only by supposing a chain of islands reaching to a continent; but the sickness of the crews, in both ships, was an insuperable impediment.

[Footnote 38:  No doubt to the Navigators’ Islands, so called by Bougainville.  Captain Wallis touched at one of them, and named them Boscawen’s and Keppel’s Islands.  Peyrouse has given a very curious, but not a pleasing account of their inhabitants.  To the south of them again are the Friendly Islands.—­E.]

The next day we again saw many birds of various sorts about the ship, and therefore supposed that some other island was not far distant, for the swell continuing, I concluded that the land was not of very great extent:  I proceeded, however, with caution, for the islands in this part of the ocean render the navigation very dangerous, they being so low, that a ship may be close in with them before they are seen.  We saw nothing, however, on the 18th, the 19th, nor the 20th, during which we continued to steer the same course, though the birds still continued about the vessel in great numbers.  Our latitude was now 12 deg.33’S. longitude 167 deg.47’W.  The Prince of Wales’s Island was distant, 313 leagues, and the variation of the needle was 9 deg.15’E.  The next morning about seven o’clock, we discovered a most dangerous reef of breakers, bearing S.S.W. and not farther distant than a single league.  In about half an hour afterwards, land was seen from the mast-head, bearing W.N.W. and distant about eight leagues; it had the appearance of three islands, with rocks and broken ground between them.  The south-east side of these islands lies N.E. by N. and S.W. by S. and is about three leagues in length between the extreme points, from both which a reef runs out, upon which the sea breaks to a tremendous height.  We sailed round the north end, and upon the north-west and west side, saw innumerable rocks and shoals, which stretched near two leagues into the sea, and were extremely dangerous.  The islands