NARRATIVE OF THE CRUISE. 
559 
The Brother Islands and Kanalur are inhabited, for cocoanut trees were seen grow- 
ing on them, and boats pulling from one island to the other, besides smoke rising 
from Kanalur. It could not be ascertained whether the Tionfolokker Islands were 
inhabited or not. The officers forbore naming the various islands of that group, as 
doubtless they have already received names from the Malays, and it would have been 
a pity to introduce others. 
All these islands appeared to be of coral formation. Even Kanalur, 1600 feet high, 
looked like a succession of . raised reefs, but whether they really are all of coral formation 
it was impossible to determine, as landing was not effected on any of them. 
At 5 p.m. Pulo Ta bore S. 32° W., and Kanalur peak N. 44° W. From this posi- 
tion the ship proceeded under steam for Bird Island, in order to fix its position, which 
appeared to be very uncertain, the chart and Horsburgh’s Directory disagreeing con- 
siderably. 
At daylight on the 28th Bird Island was sighted, and at 9 a.m. a sounding and 
temperatures were taken in 2800 fathoms (see Sheet 31). The temperature from 900 
fathoms to the bottom remained the same, showing that the Banda Sea is separated 
from the Arafura Sea and Pacific Ocean by a ridge, the greatest depth over which cannot 
exceed 900 fathoms. At noon the ship proceeded to the westward, and at 2.45 p.m., 
when Bird Island bore south (true) sights for longitude and a meridian altitude of Venus 
were obtained, which enabled the position of the island to be fixed accurately. The 
result was considerably different from the position assigned it on the chart. At 6 p.m. 
no' bottom was obtained at 600 fathoms, and the course then altered for the Banda 
Islands. 
Bird Island, in lat. 5° 33^' S., long. 130° 18' E., is about one mile long, N.N.W. and 
S.S.E., and half a mile wide. In shape it is somewhat like a small prism, and rises at an 
angle of 30° from the sea to the summit, which attains an elevation of 980 feet. There 
are trees on the island, but they do not completely cover it, as the southern part of the 
summit and the western side of its slopes were bare. Smoke was seen on shore, but 
whether this was due to volcanic agency, or merely to a fire lit by natives, could not be 
determined. 
On the 29th, at daylight, the Banda Islands were seen ahead, Pulo, Rhun bearing 
N.N.W. f W., and Pulo Rozengain N.N.E. ^ E. Steering for the passage between 
Banda and Rozengain Islands, the islands were found placed incorrectly relatively to 
each other on the chart, so a running survey of them was made. At 10 a.m. no bottom 
was obtained with 300 fathoms, the left extremity of Pulo Waii, in line with the left 
extremity of Banda N. 76^ W., and the right extremity of Banda N. 14° W. Proceeding 
a short distance farther to the northward, a sounding and dredging were obtained in 200 
fathoms with the left extremity of Banda S. 65° W., the peak of Rozengain S. 67° E., 
and the right extremity of Bancla N. 29° W. ; and again in 360 fathoms, with Gunung 
(narr. chall. exp. — vol, i. — 1885.) 71 
