CHAP. 36. (31.)—ISLANDS OF THE ÆTHIOPIAN SEA.

We learn from Ephorus, as well as Eudoxus and Timosthenes, that there are great numbers of islands scattered all over this sea; Clitarchus says that king Alexander was informed of an island so rich that the inhabitants gave a talent of gold for a horse, and of another[1] upon which there was found a sacred mountain, shaded with a grove, the trees of which emitted odours of wondrous sweetness; this last was situate over against the Persian Gulf. Cerne[2] is the name of an island situate opposite to Æthiopia, the size of which has not been ascertained, nor yet its distance from the main land: it is said that its inhabitants are exclusively Æthiopians. Ephorus states that those who sail from the Red Sea into the Æthiopian Ocean cannot get beyond the Columnæ[3] there, some little islands so called. Polybius says that Cerne is situate at the extremity of Mauritania, over against Mount Atlas, and at a distance of eight stadia from the land; while Cornelius Nepos states that it lies very nearly in the same meridian as Carthage, at a distance from the mainland of ten miles, and that it is not more than two miles in circumference. It is said also that there is another island situate over against Mount Atlas, being itself known by the name of Atlantis.[4] Five days' sail beyond it there are deserts, as far as the Æthiopian Hesperiæ and the promontory, which we have mentioned as being called Hesperu Ceras, a point at which the face of the land first takes a turn towards the west and the Atlantic Sea. Facing this promontory are also said to be the islands called the Gorgades,[5] the former abodes of the Gorgons, two days' sail from the mainland, according to Xenophon of Lampsacus. Hanno, a general of the Carthaginians, penetrated as far as these regions, and brought back an account that the bodies of the women were covered with hair, but that the men, through their swiftness of foot, made their escape; in proof of which singularity in their skin, and as evidence of a fact so miraculous, he placed the skins[6] of two of these females in the temple of Juno, which were to be seen there until the capture of Carthage. Beyond these even, are said to be the two islands of the Hesperides; but so uncertain are all the accounts relative to this subject, that Statius Sebosus says that it is forty days' sail, past the coast of the Atlas range, from the islands of the Gorgons to those of the Hesperides, and one day's sail from these to the Hesperu Ceras. Nor have we any more certain information relative to the islands of Mauritania. We only know, as a fact well-ascertained, that some few were discovered by Juba over against the country of the Autololes, upon which he established a manufactory of Gætulian purple.[7]

1. Marcus says that these islands are those called the "Two Sisters," situate to the west of the Isle of Socotra, on the coast of Africa. They are called by Ptolemy, Cocionati.

2. The position of this island has been much discussed by geographers, as being intimately connected with the subject of Hanno's voyage to the south of Africa. Gosselin, who carries that voyage no further south than Cape Non, in about 28° north lat., identifies Cerne with Fedallah, on the coast of Fez, which, however, is probably much too far to the north. Major Rennell places it as far south as Arguin, a little to the south of the southern Cape Blanco, in about 20° 5? North latitude. Heeren, Mannert, and others, adopt the intermediate portion of Agadir, or Souta Cruz, on the coast of Morocco, just below Cape Ghir, the termination of the main chain of the Atlas. If we are to trust to Pliny's statement, it is pretty clear that nothing certain was known about it in his day.

3. The "Pillars." Marcus thinks that these were some small islands near the Isle of Socotra.

4. Hardouin says that this is not the Atlantis rendered so famous by Plato, whose story is distantly referred to in B. ii. c. 92 of this work. It is difficult to say whether the Atlantis of Plato had any existence at all, except in the imagination.

5. Medusa and her sisters, the daughters of Phorcys and Ceto. The identity of their supposed islands seems not to have been ascertained. For the poetical aspect of their story, see Ovid's Met., B. iv.

6. It is not improbable that these were the skins of a species of uran- outang, or large monkey.

7. The Purpurariæ, or "Purple Islands," probably the Madeira group.