SECOND VOYAGE THROUGH TORRES STRAIT 115 



to be made by the Indians to their friends on Dungeness Island, expressive, as was 

 thought, of grief and consternation. 



" No arrows fell on board the ' Providence* but three men were wounded in the 

 ' Assistant? and one of them afterwards died. The depth to which the arrows 

 penetrated into the deck and sides of the brig was reported to be truly astonishing. 



" The vessels passed between Dungeness and Warriour's Islands [BASILISK PASS. 

 R. L. J.] . . . and anchored at 4 o'clock under the lee of DUNGENESS ISLAND and reef. 

 The passage to the westward then appeared clearer, three high islands bearing from 

 S. 60 W., 3 leagues, to N. 76 W., 5 leagues, forming the sole visible obstructions. 



" nth September. . . . Course WNW., and passed two islands, to which the 

 descriptive names of TURTLE-BACKED ISLAND and THE CAP were given, and soon, 

 after noon, the vessels anchored in 7 fathoms. . . . Lat. 9 43', long. 142 40'. . . . 

 Besides the islands above mentioned, there was in sight a mountainous land, to which 

 the name of BANKS was given, bearing S. 43 W., 12 or 13 leagues; also BURKE'S 

 ISLAND, S. I3W., 8 or 10 leagues, and MOUNT CORNWALLIS, on another island, 

 N. 29 W., 6 or 8 leagues ; and from behind this last, to N. 7 W., there extended a 

 level land, which was supposed to be a part of the coast of NEW GUINEA. 



" 12th September. The vessels followed the boats to the westward reefs, 1 and 

 anchored before noon. Lat. 9 41' S., long. 142 24' E. Two other islands were 

 then in sight. A low one, named TURN-AGAIN ISLAND, bore N. 53 W., about 4 leagues, 

 and JERVIS ISLAND, which is rather high, S. 48 W., 9 leagues. . . ." 



Here the ships lay at anchor for three days, the boats sounding, 

 while a fresh gale blew from the south-east. 



" 1 6th September. The vessels passed to windward of the southern [Orman. 

 R. L. J.] reef, and steered south-westward, as it trended, . . . until half-past noon, 

 when they anchored in lat. 10 3' and long. 142 14'.* The sole direction in which 

 the eye could range without being obstructed was that whence the vessels had come ; 

 everywhere else the view was arrested by rocks, banks and islands. The most extensive 

 of these was BANKS ISLAND, extending from S. 14 E. to 62 W., 2 or 3 leagues, with 

 a little hill upon it, named MOUNT AUGUSTUS, which bore S. 14 E. 3 Another large 

 ISLAND, named MULGRAVE'S, extended from behind the last to a cluster of rocks whose 

 extreme bore W. 5 N. The nearest land, bearing S. 24 E., i miles, was the 

 north-westernmost of three small isles, and to this the Second- Lieutenant was sent for 

 the purpose of taking possession of all the islands seen in the strait for His Britannic 

 Majesty George III, with the ceremonies used on such occasions. The name bestowed 

 upon the whole was CLARENCE ARCHIPELAGO. [The name has not been adopted by 

 cartographers. R. L. J.] 



" NORTH POSSESSION ISLAND was found to be little else than a mass of rocks 

 surrounded by a reef, but it was covered with a variety of trees and shrubs. Amongst 

 them was a cluster of coco-nut trees, bearing a small but delicious fruit. . . . 

 There did not appear to be any fixed inhabitants upon Possession Island, but from 

 a fire which had been recently extinguished, and the shells and bones of turtle scattered 

 about, it was supposed to have been visited not many days before. . . . 



" ijth September. The boats led to the westward, steering for a passage between 

 MULGRAVE'S and JERVIS' ISLANDS ; but seeing it full of rocks and shoals the vessels 



1 Charted as the " Orman Reef, 1816." 



2 There is an error of 4 minutes in the longitude if the latitude and bearing to 

 Mt. Augustus are right. R. L. J. 



3 Note by Flinders: " This mountain, in lat. 10 12' S., long. 142 13' E., was seen 

 by Captain Bligh from the 'Bounty's ' launch, and marked in his chart (Voyage, &c., 

 p. 22o)k It appears to be the same island, indistinctly laid down by Captain Cook in lat. 

 10 10', long. 141 14', and is also one of those to which the term HOGE LANDT is applied 

 in Thevenot's Chart of 1663." (The " little hill " is 1,310 feet high, according to modern 

 charts. R. L. J.). 



