Off the entrance of Van Diemen's Gulf (Dundas Strait) we passed through large quantities of sea sawdust, some of which was put into a bottle; and when the process of putrefaction had taken place the substance sunk to the bottom and coloured the water with a crimson tinge. 1820. September 3. After passing the meridian of Cape Van Diemen our course was directed towards Captain Baudin's Banc des Holothuries near Cape Bougainville; but being impeded by calms and light winds did not reach it until the 3rd of September, when we passed between its south-east extremity and Troughton Island. Before dark we passed over the north extremity of the long reef to the westward of Cape Bougainville. September 5. The following day at noon we were near Condillac Island, after which a sea-breeze from the westward enabled us to pass Cape Voltaire, at which point our last year's survey terminated. When we were within the Cape we found an ebb-tide setting out of a bight, which trended deeply in to the southward and appeared to be studded with rocky islands. This adverse tide continued to run all the evening and prevented our reaching the bottom; so that at sunset we dropped the anchor a few miles to the south of Cape Voltaire. To the westward of this position we counted twenty-three islands, the northernmost of which were supposed to be the Montalivet Isles of Baudin. The whole have an uninteresting and rocky appearance but are not altogether destitute of vegetation: a greenish tinge upon the nearest islet saved them from being condemned as absolutely sterile. September 6. The next morning a boat visited the outer north-easternmost islet, named in the chart Water Island, which was found to be as rocky in reality as it was in appearance. It is formed of a hard granular quartzose sandstone, of a bluish-gray colour; the basis is disposed in horizontal strata but the surface is covered with large amorphous rocks of the same character that have evidently been detached and heaped together by some convulsion of nature: over these a shallow soil is sprinkled, which nourishes our old acquaintance spinifex, and a variety of plants of which Mr. Cunningham collected more than twenty distinct known genera. The exposed surfaces of the rocks are coloured by the oxide of iron, which is so generally the case upon the northern and north-western coasts that the name of Red Coast might with some degree of propriety be applied to a great portion of this continent. Mr. Hunter found a large quantity of bulbous-rooted plants; they proved to be a liliaceous plant of the same species as those which we had before found upon Sims' Island, the islands of Flinders' Group on the eastern coast, and at Percy Island.