This noble conception of a great circumfluent River was doubtless founded upon reports of two classes which had reached Homer. One class would be reports of streams flowing from some great outer water into the _Thalassa_, and seeming to feed it. The other class might be formed by reports of waters outside the _Thalassa_, and not known to communicate with it, which Homer would at once very naturally reckon as portions of his great world-embracing Stream. With the former class we have already dealt largely in discussing the Ocean-mouth. To the latter one, Phœnician sailors might contribute reports of the Atlantic and German Oceans. And particularly in the east, I think, we cannot doubt that, along with the rumours and traditions of Arabians, Ethiopians, Persians, and Cimmerians, Homer cannot but have received other vague rumours of waters as well as lands; of waters exterior to his _Thalassa_ (which included the Mediterranean and the Euxine), waters of which two would clearly be the Caspian Sea, and the Persian Gulf. On these two I wish to fix attention; and indeed the only other water he was likely to have heard of would probably be the Red Sea. Now it will be observed upon any map, 1. that the Caspian lies north and south; 2. that a line prolonged from N. to S. down the Caspian will strike the Persian Gulf. In conjunction with this, let the reader observe the course of Ulysses. Quitting the Euxine at the Ocean-mouth, or Straits of Yenikalè, he turns round to the right by the Sea of Azof, enlarged so as to join the Caspian. In the interval between them there is still a low salt valley, which may in Homer’s time have been a water-way[645]. He is thus in a condition to proceed southward towards the dwelling of Persephone, which I have already shown some cause for placing in the east and to the south. Now the provision of wind, which Homer has made for his hero, is precisely that which this hypothesis requires[646]: [645] Voyages de Pallas, vol. i. p. 320, Paris 1805. [646] Od. x. 507. τὴν δέ κέ τοι πνοιὴ Βορέαο φέρῃσιν. In other words, from Homer’s use of Boreas in this place it appears that he meant to describe the course of his Ocean-stream at this quarter as from south to north, or thereabouts; and this is the line actually formed by the junction of the Persian gulf and the Caspian, which I submit that we may accordingly with propriety consider as genuine fragments of geography, incorporated into his fabulous conception of the Ocean-stream.