NA TURE 



\^Fcb. II, 1875 



the natives, Capt. Nares had given a passage to eleven 

 Api men, who had been employed for a three-years' term 

 in Fiji under the arrangement which exists there for the 

 regulation of Polynesian labour. Two or three of us, 

 with an armed party, took the returned labourers ashore ; 

 and as the natives, although they appeared somewhat 

 mistrustful, and were all armed with clubs and spears and 

 bows with sheaves of poisoned arrows, were sufficiently 

 friendly, nearly all the officers landed and spent a few 

 hours rambling about the shore. It was not thought 

 prudent to go far into the forest, which was very dense 

 and luxuriant, and came close down to the beach. 



The natives were almost entirely naked, and certainly 

 bore a very savage and forbidding aspect. One of them 

 was manifestly greatly superior to the others, and ap- 

 peared to exercise a considerable influence over them. 

 He wore trousers and a shirt and a felt hat, and could 

 speak English fairly. He recognised me at once as 

 having seen me at the sugar plantation in Queensland, 

 where he had been for the usual three-years' engagement, 

 and showed me, with great pride, a note from his former 

 employer, saying that the bearer was anxious to return to 

 his service, and that he would willingly pay his passage 

 money and all expenses in case of his being given a pas- 

 sage to Brisbane. I had been paying some attention to 

 the South Sea labour question, and had formed a very 

 strong opinion of the value to the inhabitar.ts of these 

 islands of the opportunity given them by this demand for 

 labour, of testing their capacity to enter into and mix 

 with the general current of working men, and thereby 

 possibly avoid extermination ; and 1 was greatly pleased 

 to see the result in this instance. 



From the island of Api we shaped our course to the 

 north-westward towards Raine Island in a breach of the 

 great barrier reef not far from the entrance of Torres 

 .Strait. On the 19th of August we sounded, lat. 16" 47'S., 

 long. 165° 20' E., at a depth of 2,650 fathoms, with a 

 bottom of " red clay," and abottom-tempciature of v"] C. 

 (35° F.) A serial temperature-sounding was taken to the 

 depth of 1,500 fathoms, and it was found that the mini- 

 mum temperature (i'7 C.) was reached at a depth of 

 1,300 fathoms, and that consequently a stratum of water 

 at that uniform temperature extended from that depth to 

 the bottom. 



Serial temperature-soundings were taken on the 2isf, 

 the 24th, the 25th, the 27th, and the 2Sth of August, in 

 2,325, 2,450, 2,440, 2,275, 'i^'i 1,700 fathoms respectively ; 

 and in each case the minimum temperature of i°7C., 

 or a temperature so near it as to leave the difierence 

 within the limit of instrumental or personal error of 

 observation, extended in a uniform layer, averaging 7,000 

 feet in thickness, from the depth of 1,300 fathoms 

 to the bottom. 



It will be seen by reference to the chart that on our 

 course from Api to Raine Island we traversed for a 

 distance of 1,400 miles a sea included within a broken 

 barrier, consisting of the continent of Australia to the 

 west ; the Louisiade Archipelago, the Solomon Islands, 

 and a small part of New Guinea to the north ; the New 

 Hebrides to the east ; and New Caledonia and the line of 

 shoals and reels which connect that island with Australia 

 to the south. The obvious explanation of this peculiar 

 distribution of temperatures within this area, which we 

 have called for convenience of reference the '' Melanesian 

 Sea," is that there is no free communication between this 

 sea and the outer ocean to a greater depth than 1,300 

 fathoms, the encircling barrier being complete up to that 

 point. 



The " Melanesian Sea" is in the belt of the S.E. trade- 

 winds, and the general course of a drift-current which 

 traverses its long axis at an average rate of half a knot 

 an hour is to the westward ; evaporation is, as it is usually 

 throughout the course of the trade- wmds, greatly in excess 

 of precipitation, so that a large amount of the surface- 



v/ater is removed. This must, of course, be replaced, and 

 it is so by an indraught of ocean-water over the lowest 

 part of the barrier at the proper temperature for that 

 depth. We had previously found a temperature of i^'i C. 

 at a depth of 1,300 fathoms on the i6th, the 19th, and the 

 2ist of June between Australia and New Zealand, on the 

 17th of July in lat. 25° 5' S., long. 172° 56' W., and earlier 

 on the loth of March in lat. 47° 25' S. The bottoni within 

 the Melanesian Sea may be described generally as " red 

 clay," with a small but varying proportion of the shells of 

 Foraminifera, sometimes whole but more usually much 

 broken up and decomposed. In one or two soundings 

 the tube showed curiously interstratified deposits, differ- 

 ing markedly in colour and in composition. The trawl 

 was sent down on the 25th of August to a depth of 2,440 

 fathoms. The animals procured were few in numbtr — 

 some spicules of Hyaloncma, a dead example of Fiingia 

 syniinctriia, two living specimens of a species of Umbel- 

 lularia, which appears to differ in some respects from the 

 Atlantic form, and a very fine and peifect Crisinga, also 

 living. The existence of animal life is therefore not im- 

 possible in the still bottom-water of such an enclosed 

 sea ; but, as we have already seen in the Mediterranean, 

 the conditions do not appear to be favourable to its deve- 

 lopment. On the 29th of August we trawled in 1,400 

 fathoms, about 75 miles to the east of Raine Island, with 

 somewhat greater success. This might have been antici- 

 pated, as the depth was not much greater than that at 

 which the free interchange of water was taking place, and 

 diftusion and intermixture was no doubt much more rapid 

 than at the bottom. 



On the 31st of August we visited Raine Island, which we 

 found to correspond in every respect to Jukes's description 

 in the " ^'oyage of the Fly.'' We observed and collected 

 the species of birds which were breeding there. In the- 

 afternoon we dredged off the island in 155 fathoms with 

 small success, and proceeded towards Port Albany, Cape 

 York, where we arrived on the ist of September. 



We left Somerset on the Sth, and proceeded across the 

 Arafura Sea to the Aril Islands, reaching Dobbo on the 

 island of Wamma on the i6th. We found no depth in 

 the Arafura Sea greater than 50 fathoms, and the average 

 depth was from 25 to 30 fathoms. The bottom was a 

 greenish mud, due apparently in a great degree to the 

 deposit from the great rivers of New Guinea and the 

 rivers falling into the Gulf of Carpentaria. Animal life 

 was not abundant. Many of the animals seemed dwarfed, 

 and the fauna had somewhat the character of that of a 

 harbour or estuary. The specific gi'avity of the surface- 

 water was unusually low, falling on the 23rd off Dobbo 

 Harbour to ro25o5 ; the temperature reduced to I5°'S C, 

 distilled water at 4^ C. = i. 



After spending a few days shooting Paradise Birds and 

 getting an idea of the natural history of the island of 

 Wokaw, we left Dobbo on the 23rd and proceeded to Kd 

 Doulan, the principal village in the Kd group. We then 

 went on to the island of Banda, where we remained a 

 couple of days, and thence to Amboina, which we reached 

 on the 4th of October. 



On the 20th of September, after leaving the K^ islands, 

 we sounded and trawled in 129 fathoms. The trawl 

 brought up a wonderful assemblage of things, including, 

 with a large number of MoUusca, Crustacea, and Echino- 

 derms of more ordinary forms, several fine examples of 

 undescribed Hexactincliid sponges, and several very per- 

 fect specimens of two new species of Pentacrinus. Tem- , 

 perature-soundings were taken on the 2Sth of September 

 and on the 3rd of October at depths of 2,Sco and 1,420 

 fathoms respectively, and on both occasions the minimum 

 tem.perature (3° C.) was reached at a depth of 900 fathoms, 

 indicating that the lowest part of a barrier inclosing the 

 Banda Sea, bounded by Taliabo, Buru, and Ceram on 

 the north, the Arii islands on the east, Timor and the 

 Serwatty islands on the south, and Celebes and the shoals 



