To the West of Cape Denis is Cape North, the extremity of some low islands, which appear to be the westernmost land. Beyond this is a line of coral reefs, which were called the Lusançay Reefs. They extend to a very considerable distance; and indeed are more or less connected with a range of coral barrier reefs to the meridian of 150° 40', where D'Entrecasteaux again encountered the land. This extensive line runs nearly on a parallel, and partakes of all the dangerous characteristics of the coral reefs-perpendicular borders, and detached outlying shoals, one of which D'Entrecasteaux particularly notices.

The COAST of NEW GUINEA.-The coast to the North was very imperfectly seen and delineated by D'Entrecasteaux. The first point more particularly noticed was Cape Longuerue, the South extremity of Huon Gulf. It is in lat. 7° 22′, and lon. 147° 23′ 45" (D'Entrecasteaux). To the N.W. is a group of low islands. The mountains of New Guinea are here prodigiously high, affording a magnificent spectacle. Cape Cretin is the North point of Huon Gulf. At 5 miles S.E. of it are three small low islands; the northernmost of these is in lat. 6° 47′ 45′′, lon. 147° 50′ 4′′.†

RICHE ISLAND lies to the N.W. of the land first laid down on this exploration, which is a portion of this great island. It is of considerable size, and tolerably high. Its North point is in lat. 8° 2', lon. 147° 57' 40" (D'Entrecasteaux). CAPE KING WILLIAM is the West point of Dampier's Strait, and is elsewhere described.*

NEW IRELAND.

The N.E. side of this island was discovered by Le Maire and Schouten in 1616, and was again seen by Tasman in 1643; but these navigators supposed it to form a portion of the large Island of New Guinea. This supposition was disproved by Dampier, who sailed through the strait now bearing his name in 1700; but all the land to the eastward retained the name of Nova Britannia till 1767, when Carteret found that the bay, called by Dampier St. George's Bay, was in reality a strait separating it into two islands; the land to the eastward then received the name of New Ireland, that to the westward retaining the title of New Britain. Its native name is Tombara.

The island is very imperfectly known with the exception of two or three points which have been more recently visited and better determined, we are obliged to resort to the accounts of the original discoverers, necessarily very imperfect authorities. According to these, the island extends 60 leagues in a N.W. and S.E. direction; the North side remains nearly unknown. The South side has been imperfectly examined by Carteret, D'Entrecasteaux, Hunter, and Duperrey.

CAPE ST. GEORGE is the southernmost point; and, from the observations of Duperrey at Port Praslin, it is in lon. 152° 48′ 14"; D'Entrecasteaux places it in lat. 4° 51' S.

CAPE SANTA MARIA is the easternmost point of New Ireland; and hence the coast assumes a new direction, trending away to W.N.W. The land between

D'Entrecasteaux's Voyage, by Rossel, vol. i. chap. xix. p. 403, et seq. The positions are corrected by D'Urville's observations, as detailed in M. Vincendon-Dumoulin's charts.

the two capes last named is high, mountainous, and wooded; numerous points jut out, forming as many fine bays.*

ST. JOHN'S ISLAND lies off Cape Santa Maria, and was so named by Tasman from the day it was first seen. Bougainville calls it Bournand Island. Dampier says it is 9 or 10 leagues in circumference, and well covered with large trees. Much cultivation was visible, and abundance of cocoa-nut trees. The people came off, and appeared friends.

ANTHONY KAAN ISLAND of Tasman is probably the Oraison Island of Bougainville, who places it in 3° 30′ S. Both these navigators speak but of one, but Dampier (whose accuracy is proverbial) says:-"On the S.E. part of it are three or four small woody islands, one high and peaked, the other low and flat; all bedecked with cocoa-nut trees and other wood." The island itself is 4 or 5 leagues in circumference, high, well wooded and well cultivated, having abundance of cocoa-nut trees. On the North side of it is another island of moderate height, and rather larger than the high island. Dampier passed between them.† Off these islands to the eastward is a group called the Fead Islands.

FEAD ISLANDS, or ABGARRIS.-This group was discovered by Capt. Renneck, of the Lyra, eight days after the discovery of the shoal so named. They occupy a space of 9 leagues in a N.W. and S.E. direction, and are composed of a chain of low islands and sand-banks, surrounded by a reef, the North part of which is in lat. 3° 9' S., and lon. 154° 22′ E.

GOODMAN ISLAND is the southernmost of the Fead Islands. It is separated from the rest by a channel which Capt. Renneck could not examine, so that its safety is doubtful. To the South of this island is a detached shoal, with a reef, in lat. 3° 33' S., and lon. 154° 37' E. The greater part of the islands are inhabited, and they abound in cocoa-nut trees.

GERRIT DENYS ISLAND, another discovery of Tasman, is called by Bougainville Ile du Bouchage. Dampier says it is of an irregular figure, projecting points forming sandy bays in all parts. It is 14 or 15 leagues in circumference, high and mountainous, and very well wooded. In the bays are abundance of cocoa-nut trees. The timber appeared to be fine. The island was very populous, but the natives were considered to be treacherous and warlike. They will come off in canoes with cocoa-nuts, taro, plantains, &c., for barter to passing ships (1839).

LYRA SHOAL.-This reef was discovered, February 8, 1826, by Capt. Renneck, commanding the ship Lyra, belonging to the well-known and enterprising merchants, Messrs. Enderby. It is a narrow belt of rock, extending 11 miles in a N.W. and S.E. direction, or from 14° 8' to 15° 9' S., the longitude being 153° 28'. The ship crossed the middle of this reef, and the coral rocks beneath were plainly seen, the depth being estimated at 4 or 5 fathoms.

DAMPIER ISLANDS.-Some islands were seen by Dampier, March 2, 1699. He does not give any detailed account of them, but states that to the N.E. of the larger island were two others, one small but woody, the other a league long, inhabited, and full of cocoa-nut trees. The latter was so narrow, that there was no shelter on the East side of it. The large island was thickly inhabited.

Dampier, vol. iii. p. 207.

+ Ibid. vol. iii. p. 204.

WISHART OF FISHER'S ISLAND of Le Maire and Schouten, 1616, is called Suzannet Island by Bougainville. Its South extremity is 10 leagues W. by N. of Gerrit Denys Island, and consequently is in lat. 2° 55′ S. The island lies North and South, and its North point is in lat. 2° 32′.

Maurelle mentions several other islands on the N.W. of New Ireland, but describes them very superficially. They are named San Francisco, San Josef, and San Antonio, but their existence and character require confirmation.

The COAST of New Ireland at this part is high and mountainous, and covered with fine trees. The sides of the hills are cleared and cultivated in many parts, and this with numerous smokes indicate a dense population.

SLINGER'S BAY is in this part; it is a point where Dampier, seeking anchorage, was deterred by the distance from the sea and the hostile appearance of the natives, who, on his returning, attacked him by slinging stones, but they were quieted by firing a gun.

CAPE BYRON is the N.W. extremity of New Ireland, and was thus named by Carteret, September 12, 1767. He determined the separation between New Ireland and New Hanover, calling the channel Byron Strait. It probably is not navigable; for Carteret says, there are several small islands in it, upon one of which is a remarkable peak, to which he gave the name of Byron Island; D'Entrecasteaux calls it Mausolée Island, and says the channel is filled with islets and rocks.

NEW HANOVER, the island to the westward, so called by Carteret, is stated by him to be high, finely covered with trees, among which are many plantations, the whole having a beautiful appearance. The South point was called Queen Charlotte's Foreland, in honour of the queen. The cape and the land in its vicinity a reremarkable for numerous small hummocks. The island is still very imperfectly known. Tasman says that the North coast runs East and West, 10 or 12 leagues; his description in some degree identifies the channel, separating it from New Ireland (Byron Strait). Sailing to the westward, August 7, 1643, he passed four low islands, and then three others near noon; in the afternoon he came to a low point, before which lay two islands, which he named Point Salomon Sweert, after a member of the Council of the Indies. The coast then trended to the southward, so that it must have been the N.W. point of New Hanover.

The PORTLAND ISLES, which are alluded to by Tasman, were thus named by Carteret. He says they are six or seven in number, two of them being tolerably large. They were seen at a great distance.

SANDWICH ISLAND, to the South of the West end of New Ireland, was first distinguished as an island and so named by Carteret. Admiral Hunter describes it as of moderate height, and well covered with wood. It was passed on the South side. Its general direction was about E.N.E. and W.S.W., and in that direction is about 7 leagues (D'Entrecasteaux says 14 miles). It appeared to Hunter to be of considerable breadth at its eastern end, and narrow towards its western, where it terminates in a narrow point, off which lies a small woody island, with a narrow passage between it and the main island, to which it appears to be connected by a reef.

The channel between Sandwich Island and New Hanover is 7 miles broad.

Carteret says, on the North part of it is a remarkable peak, like a sugar-loaf, and opposite to it, on the coast of New Ireland, there is another; they are distant from each other about 5 leagues, according to Carteret (D'Entrecasteaux says, only half that distance), in a S. by E. & E. and N. by W. W. direction. The peak on Sandwich Island, according to D'Entrecasteaux, is in lat. 2° 55′ S., lon. 150° 44′ E. Carteret says the island is pleasant, and well inhabited.*

ST. GEORGE'S CHANNEL separates New Ireland from New Britain; its true character was first determined by Carteret, in September, 1767. It was previously marked as a bay from Dampier. It is 6 or 7 leagues broad, and contains several islands, but would appear to offer a safe and convenient passage for ships passing to the North or South.

DUKE OF YORK ISLAND, or AMACATA, was discovered and named by Carteret. He describes it as level, and having a delightful appearance. Inland it is covered with lofty woods, and near the sea-shore are the houses of the natives, which stand not far from each other, among groves of cocoa-nut trees, forming a beautiful and romantic prospect. Carteret passed to eastward of it, the channel being about 8 leagues broad. Capt. Hunter gives a long account of the island and its inhabitants. He judged it to be about 10 miles long, in a S.S.W. and N.N.E. direction; not high, nor yet to be called low land, well covered with wood. From the observations made while on it, it was pronounced to be a perfect garden, as far as it could with propriety be called so in the hands of an uncultivated people. The soil, for richness, was beyond anything they had seen, exceeding Norfolk Island in that respect. It produces cocoa-nuts and the usual tropical fruits; hogs, poultry, &c., are kept by the natives. They were friendly at first, but afterwards inclined to be hostile. The bay in which he anchored is on the N.W. part of the island, and was found to be convenient and safe at that season (May, 1791). Anchorage, in any part of it, in 25 to 15 fathoms; the shoalest water has the worst ground. The watering place is on the East side, from some rivulets; but as the tide flows up to the place whence the fresh water (which is very soft) is procured, it is best to fill from half-flood to half-ebb. The tide rises 5 or 6 feet. The bay was called Port Hunter, and lies in lat. 4° 7′ 30′′, lon. 152° 22' E., according to Duperrey.

CARTERET'S HARBOUR is on the eastern side of St. George's Channel, towards the S.E. end of New Ireland. Carteret anchored here in September, 1767. It is formed by two islands and the main; the largest, which is to the N.W., was called Cocoa-nut Island, and the other, to the S.E., was called Leigh's Island. The channel is here 23 miles wide. Between the two islands is shoal water, and each of them forms an entrance into the harbour: the S.E. or weather entrance is formed by Leigh's Island, and in this there is a rock above water, named the Booby Rock; the passage is between the rock and the island; the rock has deep water all round it. The N.W. or lee entrance is formed by Cocoa-nut Island, and this is the best, because there is good anchorage in it, the water in the other being too deep. Carteret entered by the S.E. passage, and

Carteret, in Hawkesworth, vol. i. p. 598; Hunter's Historical Journal, p. 238; D'Entrecasteaux, par Rossel, tome i. pp. 129-30. + Hunter's Historical Journal, pp. 226–238.

quitted by the N.W. At the S.E. end of the harbour is a large cove, secure from all winds, and fit to haul a ship into. In the N.W. part of the harbour is another cove, from which very good water was procured; this, too, is fit to haul a ship into, and is very convenient to wood and water in; she may lie in from 30 to 5 fathoms, soft muddy bottom. The harbour runs about S.E. by S. and N.W. by N., and is about 3 miles long and 4 cables' length broad.* D'Urville watered here, and began his examination of New Britain opposite to it. The N.E. point of Cocoa-nut Island is in lat. 4° 42′ 0′′, lon. 152° 44′ 30′′.

GOWER'S HARBOUR, or PORT PRASLIN, is to the S.E. of Carteret Harbour, according to Carteret, 4 leagues distant, but Capt. Hunter, in 1791, was surprised to find it only 2 leagues. The first name was given to it by Carteret, the discoverer; the second by Bougainville, who anchored here, and observed an eclipse of the sun, July 13, 1768. He gives the following account of it. It is formed by a larger island, Wallis or Marteaux Island, and a smaller, Green Island, to the South of it.

"In entering, a small island (Wallis Island) and an islet (Green Island) will be left to port on the West; they are half a league from the coast; a point which projects opposite the islet forms within it an excellent harbour, sheltered from all winds, the bottom throughout of white sand, the depth from 30 to 13 fathoms. Off the East point is a breaker, but it is visible, and does not extend far off. On the North side of the bay there are two other banks, which uncover at low water. On the edge of the reefs there are 11 fathoms. The entrance of this port is very easy; the only attention required is to hug the East point close-to with plenty of sail, because as soon as it is doubled, you will be becalmed, and then you must proceed with the headway.

"There is excellent opportunity for watering here at four rivulets, all within 400 paces. The landing is excellent, without a rock or any swell. Wood may (or might) be had on the shore, excellent as firewood, and much of it beautiful for carpentry. There were no inhabitants, and this gave all the quietude desired; neither cocoa-nut, nor banana, nor other vegetable could be procured, but fish was abundant.”+

Capt. R. L. Hunter says that a ship being in the neighbourhood of New Ireland or the Salomon Islands, and in want of water, firewood, or spars, cannot obtain these with greater facility or safety than in Gower's Harbour. The southern entrance to this harbour is about 2 miles from Cape St. George; either entrance will, however, be easily seen by a ship at a moderate distance in the offing. There is water in all the coves, but the northernmost has the finest stream, and either that or the southern cove will be found the most convenient. The most expeditious method is to drop anchor in 16 or 18 fathoms, then run a line to the shore, and haul your stern in. There is anchorage also on the North side of the East point of Wallis Island, in 28 or 30 athoms, about 40 yards from the shore. You are so completely sheltered with the high land above you, in every direction, that a ship, in case of necessity, may select many places suitable for heaving out.

* Carteret, in Hawkesworth's Collection, vol. i. p. 594.

+ Bougainville, pp. 273-4.