Now we are provided with an important argument, drawn, like some preceding ones, from what we may fairly call Homer’s experience, and tending to fix the site of Ogygia in the north or north-east. It is derived from the route taken by Mercury, when he carries the message of the Immortals to Calypso, which in another point of view we have already had to examine[612]: [612] Od. v. 50. Πιερίην δ’ ἐπιβὰς, ἐξ αἰθέρος ἔμπεσε πόντῳ. We are obliged to suppose, as has been observed, that Mercury, who does not march, but flies like a bird wont to hunt for fish[613], must move in a direct line towards his object. But Pieria is a district stretching along the shore of Macedonia; it begins in the south, to the eastward of Olympus, and then extends due north of it. Its limits are variously defined[614]; but the only question about it could be, whether it verges, not to the westward, but to the eastward of North. Again, from the route of Juno in the Fourteenth Iliad[615], no question can arise, except what would tend to give Pieria an eastward turn. [613] Ibid. 51-3. [614] Cramer’s Greece, i. 204. [615] Il. xiv. 226. A line drawn from Olympus over the centre of Pieria would carry Mercury to the North. It might, consistently with the condition of crossing Pieria, diverge a little either to the east or the west of due North, but only a little. Consequently the island of Calypso may be affirmed to be, according to the intention of Homer, in the North, and not very far from due North. This conclusion is confirmed by two other arguments; which are both of the class which I have described as legitimate, because they are founded on Homer’s physical knowledge of the direction of the winds. After the storm has destroyed the ship of Ulysses to the south of Thrinacie, Notus, a wind of decidedly southerly character, carries him back again to Scylla, Od. xii. 426: and again, when he has passed it, he proceeds thus[616]: [616] Od. xii. 447. ἔνθεν δ’ ἐννῆμαρ φερόμην, δεκάτῃ δέ με νυκτὶ νῆσον ἐς Ὠγυγίην πέλασαν θεοί. Now there is no mention between these two passages either of any change of wind, or of any particular wind. Consequently it seems rational to assume that Homer meant us to understand a continuance of the wind just named, namely Notus. Even independently of this collocation, we should be thrown back upon the general rule of the Wanderings, which is that southerly winds blow Ulysses away from home, while northerly ones bring him back again.