Having completed our water, and got on board every other thing we wanted at Teneriffe, we weighed anchor on the 4th of August, and proceeded on our voyage, with a fine gale at N.E. At nine o'clock in the evening on the 10th,[79] we saw the island of Bonavista bearing south, distant little more than a league; though, at this time, we thought ourselves much farther off: But this proved a mistake. For, after hauling to the eastward till twelve o'clock, to clear the sunken rocks that lie about a league from the S.E. point of the island, we found ourselves, at that time, close upon them, and did but just weather the breakers. Our situation, for a few minutes, was very alarming. I did not choose to sound, as that might have heightened the danger, without any possibility of lessening it. I make the north end of the island of Bonavista to lie in the latitude of 16° 17' N., and in the longitude of 22° 59' W. [Footnote 79: As a proof of Captain Cook's attention, both to the discipline and to the health of his ship's company, it may be worth while to observe here, that it appears from his log-book, he exercised them at great guns and small arms, and cleaned and smoked the ship betwixt decks, twice in the interval between the 4th and the 10th of August.--D.] As soon as we were clear of the rocks, we steered S.S.W., till day-break next morning, and then hauled to the westward, to go between Bonavista and the isle of Mayo, intending to look into Port Praya for the Discovery, as I had told Captain Clerke that I should touch there, and did not know how soon he might sail after me. At one in the afternoon, we saw the rocks that lie on the S.W. side of Bonavista, bearing S.E., distant three or four leagues. Next morning, at six o'clock, the isle of Mayo bore S.S.E., distant about five leagues. In this situation we sounded, and found ground at sixty fathoms. At the same time the variation, by the mean of several azimuths taken with three different compasses, was 9° 32 1/2' W. At eleven o'clock, one extreme of Mayo bore E. by N., and the other S.E. by S. In this position, two roundish hills appeared near its N.E. part; farther on, a large and higher hill; and, at about two-thirds of its length, a single one that is peaked. At the distance we now saw this island, which was three or four miles, there was not the least appearance of vegetation, nor any relief to the eye from that lifeless brown which prevails in countries under the Torrid Zone that are unwooded.