the list we may place the two quartos published by de Brosses in 1756, containing a complete collection of all the known voyages to the South Lands-(p. 575); the first volume of which contained the Dutch voyages en Australasie, with a chapter (xxvi) on the Dutch discoveries in New Holland. The charts published with each volume showed the position and extent of Nouvelle Hollande as it was then known, and were no doubt consulted with peculiar interest as the Endeavour neared its eastern coast. When he was leaving it in September, 1770, Cook mentioned them in his journal :-" The charts with which I compared such parts of this coast as I visited, are bound up with a French work entitled Histoire des Navigations aux Terres Australes, which was published in 1756, and I found them tolerably exact." Looking at one of these charts, we observe that there is nothing to indicate the existence of the straits between the mainland and Van Diemen's Land; but the passage now known as Torres Straits is distinctly shown, although in the text the author repeatedly expresses a doubt whether the mainland touched New Guinea or not.

Why this doubt should have been expressed by de Brosses when the position of the straits is shown so clearly in his charts, is a question not easily answered. The discovery of the fact that Torres had sailed through the straits in 1606 is attributed to Dalrymple, who made it known to the world in his Account of the Discoveries in the South Pacific Ocean previous to 1764, published in 1767-a work which we may safely assume had its place in the Endeavour's library-(p. 576). Flinders states in his introduction that "the existence of such a strait was generally unknown until 1770, when it was again discovered and passed by our great circumnavigator, Captain Cook." In making this statement, he seems to have repeated a remark made in the introduction (p. xvi) to Cook's Third Voyage, where the reader is told that "though the great sagacity and extensive reading of Mr. Dalrymple had discovered some traces of such a passage having been found before, yet those traces were so obscure and so little known in the present age that," among other things, "the President de Brosses had not been able to satisfy himself

about them."* But unless he had satisfied himself on the subject, why did he construct his maps of New Holland and New Guinea in such a manner as to show the straits? This is one of the many little puzzles connected with Australian geography of the last century which deserve the attention of those who are interested in it. The only answer to the question seems to be that de Brosses looked upon New Holland as an island, probably considering that fact established; but not having seen the Relation written by Torres of his passage through the straits, he thought that there was just room for a doubt on the subject. Nothing was known about Tasman's second voyage in his time.

Dalrymple's Historical Collection of Voyages and Discoveries in the South Pacific Ocean was another work of great authority at the time it was published-1770. It contained a chart of the South Pacific, "pointing out the discoveries made therein previous to 1764," which showed Torres' track in 1606 through the straits. The work made its appearance too late to form part of the Endeavour's library; but its influence on the geographical speculation of the age may be seen at a glance if we compare the introduction and the chapter entitled "Investigation of what may be farther expected in the South Sea," with the introduction to Cook's Voyage towards the South Pole. Dalrymple was an enthusiast on this subject, but he was not entitled to any credit for originality in his speculations; he merely revived the old theory of the southern continent, but he did it with so much force of argument and illustration that an expedition to determine the question was a natural result. It was perhaps unfortunate for us that his work was not published before the Endeavour sailed; because we may be allowed to suppose that if Cook had had an opportunity of reading it, his attention would have been directed to the name AUSTRALIA, from its frequent appearance in capital letters-suggesting the idea that the author intended to

*Cook stated in 1776 that "Torres seems to have been the first who sailed between New Holland and New Guinea"-Voyage towards the South Pole, introduction, p. xii; but he made no reference to him when he gave the name Endeavour Streights to the passage between the Prince of Wales's Islands and the mainland in 1770; Hawkesworth, vol. iii, p. 619.

point it out as the proper name for the country. By such an accident, the land discovered by Cook might, peradventure, have escaped the unsatisfactory name it has since borne.

Callander's translation of de Brosses appeared in 1766-68(p. 576); the second and third volumes being published only two months before the Endeavour sailed; but we may take it for granted that they were not left behind. The three volumes had the advantage of being published in a handy form; but the literary execution was slovenly, and it is manifest that Callander was not a geographer of much discrimination. He published two charts, reproductions from the French work, the larger one showing the outlines of New Holland and the discoveries of de Quiros. Let us suppose that, as soon as the Endeavour was steered westward from Cape Farewell in New Zealand, Cook and his companions read the following account of the country they were about to explore:

New Holland is that vast region which extends from the sixth to the 34th degree of south latitude, and from longtitude 124 degrees to 187. To the north it has the Molucca islands, or the sea of Lanchidol. To the west and south the Indian Ocean, and the Pacific to the east. But, in this immense stretch of land, we are acquainted only with some parts of the coast lying separated from each other, without being able to affirm whether they compose one continent or (as it is more likely) they are large islands separated from each other by canals or arms of the sea, the narrowest of which have been supposed by navigators to be the mouths of rivers. Neither are we yet assured if New Holland joins New Guinea on the north, or Van Diemen's Land to the south. Tasman has verified by his course that New Zealand, lying to the southeast, is totally separated by the sea from the continents and islands that lie nearer the equator.

The principal countries of New Holland we are as yet acquainted with are, Carpentaria to the north-east, the coast of which, forming a great bay, faces to the west. At the entry of this bay are the Molucca islands; to the north lie the lands of Arnheim and Diemen, which last is different from the Diemen of Abel Tasman: To the north-west lies the land of De Witte. Towards the west lie Endracht or Concordia, Edels, and Lewin. This last occupies the point which lies south-west. To the south lies the land of Peter Nuytz, and further south, but

trending eastwards, the land of Diemen, if indeed this last should be comprehended under the division we are now describing.

In running along the east coast of this country, back towards the Equator, we find the Terra Australis del Espiritu Santo, discovered by Quiros. But all this vast interval, lying behind Lewin and Quiros' discovery, is so little known that we cannot tell what part of it is land and what is sea. This tract extends from latitude 43 degrees south to latitude 19 degrees, and has not hitherto been visited, at least as far as we know.

The last paragraph shows that Callander, following de Brosses, imagined that the land discovered by de Quiros formed part of the mainland, as shown on the chart. But this error was detected by Cook before he passed out of the reefs into the open sea. How correctly he had judged the matter may be seen from his statement on the 13th August, 1770, when he wrote:

The islands which were discovered by Quiros, and called Australia del Espiritu Santo, lie in this parallel; but how far to the eastward cannot now be ascertained in most charts they are placed in the same longitude with this country which, as appears by the account of his voyage that has been published, he never saw; for that places his discoveries no less than two and twenty degrees to the eastward of it.*

It is worth while to remember that Dampier intended, in 1699, to begin his discoveries "upon the Eastern and least known Side of the Terra Australis." He did not carry out that intention because he was afraid of "compassing the South of America in a very high Latitude in the depth of the Winter there"; and also for another reason which he stated in these words

For should it be ask'd why at my first making that Shore, I did not coast it to the Southward, and that way try to get round to the East of New Holland and New Guinea; I confess I was not for spending my Time more than was necessary in the higher Latitudes; as knowing that the Land there could not be so well worth the discovering, as the Parts that lay nearer the Line, and more directly under the Sun. †

*Hawkesworth, vol. iii, pp. 602-3.

+ Voyage to New Holland, vol. iii, pp. 124-5, ed. 1729.