TERRA AUSTRALTS. 
223 
Melville Bay . ] 
of the rocks which form the south-east side of this snug little T ,\ sos - 
place, Harbour Rock ; and the sandy point at the entrance of the bay Sunday is. 
is named Point Dundas. After the seine had been hauled witli good 
success, I walked to the extremity of the point ; and from a hillock 
of sand a little way back, took a set of bearings to commence the 
survey, in which was included the bluff north end of Mount Saund- 
ers at N. 74° 55' E. Many foot-marks of men, dogs, and small kangu- 
roos were observed on the beach, but neither natives nor quadrupeds 
were seen. 
Early next morning a party of men was sent to cut wood, Monday u. 
and the botanical gentlemen landed on Point Dundas upon their pur- 
suits ; I went to examine the north-eastern part of the bay, where 
the water extended two miles above the ship; but the depth in it pre- 
sently diminished to 2^ fathoms, and to 1 near the end. Beyond a 
low isthmus there, a piece of water was seen communicating with 
the south-eastern part of the bay, and making a peninsula of the high 
rocky land named Drimmie Head; at high water, indeed, it is an 
island, for the tide then flows over some parts of the isthmus. After 
taking two sets of bearings, I rowed southward along the shore of 
Drinnnie Head ; and from a hill near the south-west extremity ob- 
tained a good view of the bay, and saw the western coast as far 
northward as a cliffy cape which was named after William Wilber - 
force, Esq., the worthy representative of Yorkshire. The principal 
bearings from hence were, 
Cape Wilberforce, highest part, - N. 25* 40' W. 
Mount Bonner, - - - N. 51 55 W. 
Point Dundas, distant 2 miles, - - N. 52 30 W. 
Leaving Drimmie Head, I steered over to the middlemost of 
the three rocks in the bay, with a depth of water from 3 to 6 ~ fathoms, 
on muddy ground. These rocks lie nearly due south from Point 
Dundas, and I proposed to observe the latitude on both sides from 
thence, whilst lieutenant Flinders did the same at the point, that a 
base line for the survey might be obtained from the difference ; but 
