On the 5th in the afternoon on our return to the vessel, after visiting the shore and landing upon the flat rock, which is merely a bare mass of granite, of about thirty yards in diameter, some natives were heard calling to us, and upon our pulling to the part whence the sound came, we found two men and a boy. After some time they were discovered to be three of our Oyster-Harbour friends, and therefore we made no hesitation of communicating with them, and of taking them on board, where they were regaled upon the flesh of the seals we had killed at the island.

Notwithstanding the friendly disposition of the inhabitants of this sound, I felt it necessary to act very cautiously in our communication with them, in order to avoid any misunderstanding. And that this might not even be accidentally done, I requested Mr. Cunningham to confine his walks to the vicinity of the vessel, and particularly to avoid any route that would take him towards their encampment. He was therefore prevented from visiting many parts near which he had promised himself much amusement and information in botanizing, particularly the neighbourhood of Bayonet Head, and the distant parts of Oyster Harbour. At our former visit to this place he had searched in vain for that curious little plant Cephalotus follicularis, Br.,* but on this occasion he was more fortunate, for he found it in the greatest profusion in the vicinity of the stream that empties itself over the beach of the outer bay where we watered. Of this he says: "The plants of cephalotus were all in a very weak state, and none in any stage of fructification: the ascidia, or pitchers, which are inserted on strong foot-stalks, and intermixed about the root with the leaves, all contained a quantity of discoloured water, and, in some, the drowned bodies of ants and other small insects. Whether this fluid can be considered a secretion of the plant, as appears really to be the fact with reference to the nepenthes, or pitcher-plant of India,** deposited by it through its vessels into the pitchers; or even a secretion of the ascidia themselves; or whether it is not simply rainwater lodged in these reservoirs, as a provision from which the plant might derive support in seasons of protracted drought, when those marshy lands (in which this vegetable is alone to be found) are partially dried of the moisture that is indispensable to its existence, may perhaps be presumed by the following observations. The opercula, shaped like some species of oyster, or escalop-shells, I found in some pitchers to be very closely shut upon their orifices, although their cavities, upon examination, contained but very little water, and the state of the weather was exceedingly cloudy, and at intervals showery; if, therefore, the appendages are really cisterns, to receive an elemental fluid for the nourishment of the plant in times of drought, it is natural to suppose that this circumstance would operate upon the ramified vessels of the lids, so as to draw them up, and allow the rain to replenish the pitchers. Mr. Brown also, who had an opportunity in 1801 of examining plants fully grown, supposes it probable that the vertical or horizontal positions in which the opercula were remarked, are determined by the state of the atmosphere, at the same time that he thinks it possible that the fluid may be a secretion of the plant. The several dead insects that were observed within the vases of cephalotus were very possibly deposited there by an insect of prey, since I detected a slender-bodied fly (ichneumon) within a closed pitcher, having evidently forced its passage under the lid to the interior, where an abundant store of putrescent insects were collected. Whilst, therefore, these pitchers are answering the double purpose, of being a reservoir to retain a fluid, however produced, for the nourishment of the plant in the exigency of a dry season, as also a repository of food for rapacious insects, as in sarracenia, or the American pitcher-plant; it is also probable that the air, disengaged by these drowned ants, may be important and beneficial to the life of the Australian plant, as Sir James E. Smith has suggested, in respect to the last-mentioned genus, wild in the swamp of Georgia and Carolina.

(*Footnote. Flinders volume 1 page 64 and Brown's General Remarks in Flinders volume 2 page 601 et seq.)

(**Footnote. Smith's Introduction to Botany page 150.)

"I spent much time in a fruitless search for flowering specimens of cephalotus; all the plants were very small and weak, and showed no disposition to produce flowers at the season, and none had more than three or four ascidia."*

(*Footnote. Cunningham manuscripts.)

The only edible plants that Mr. Cunningham found were a creeping parsley (Apium prostratum, Labil.) and a species of orach (Atriplex halimus, Brown) the latter was used by us every day, boiled with salt provisions, and proved a tolerable substitute for spinach or greens. During our visit we caught but very few fish, and only a few oysters were obtained, on account of the banks being seldom uncovered, and the presence of the natives which prevented my trusting the people out of my sight for fear of a quarrel. Shellfish of other sorts were obtained at Mistaken Island in abundance, of which the most common were a patella and an haliotis; the inhabitant of the former made a coarse, although a savoury dish. There were also varieties of the following genera: namely, lepas, chiton, cardium, pinna, nerita, two or three species of ostrea, a small mytilus, and a small buccinum of great beauty; that covered the rocks and at low water might be collected in abundance.

CHAPTER 4.

Leave King George the Third's Sound, and commence the survey of the West Coast at Rottnest Island.
Another remarkable effect of mirage.
Anchor under, and land upon Rottnest Island.
Break an anchor.
Examine the coast to the northward.
Cape Leschenault.
Lancelin Island.
Jurien Bay.
Houtman's Abrolhos.
Moresby's Flat-topped Range.
Red Point.
Anchor in Dirk Hartog's Road, at the entrance of Shark's Bay.
Occurrences there.
Examination of the coast to the North-west Cape.
Barrow Island.
Heavy gale off the Montebello Isles.
Rowley's Shoals.
Cape Leveque.
Dangerous situation of the brig among the islands of Buccaneer's Archipelago.
Examination and description of Cygnet Bay.
Lose an anchor, and leave the coast.
Adele Island.
Return to Port Jackson.