Next occurs a succession of rivers and roadsteads, seven in number, which being passed we are conducted to the +Puralaän+ Islands, and what is called a canal or channel (διώρυξ). These islands are not mentioned elsewhere. They can readily be identified with the two called +Manda+ and +Lamou+, which are situate at the mouths of large rivers, and are separated from the mainland and from each other by a narrow channel. Vincent would assign a Greek origin to the name of these islands. “With a very slight alteration,” he says, “of the reading, the Puralian Islands (Πῦρ ἁλιον, _marine fire_,) are the islands of the Fiery Ocean, and nothing seems more consonant to reason than for a Greek to apply the name of the Fiery Ocean to a spot which was the centre of the Torrid Zone and subject to the perpendicular rays of an equinoctial sun.” [The Juba islands run along the coast from Juba to about Lat. 1° 50´ S., and Manda bay and island is in Lat. 2° 12´ S.]
Beyond these islands occurs, after a voyage of two days and two nights, the island of +Menouthias+ or +Menouthesias+, which it has been found difficult to identify with any certainty. “It is,” says Vincent, “the _Eitenediommenouthesias_ of the _Periplûs_, a term egregiously strange and corrupted, but out of which the commentators unanimously collect Menoothias, whatever may be the fate of the remaining syllables. That this Menoothias,” he continues, “must have been one of the Zangibar islands is indubitable; for the distance from the coast of all three, Pemba, Zangibar, and Momfia, affords a character which is indelible; a character applicable to no other island from Guardafui to Madagascar.” He then identifies it with the island of Zangibar, lat. 6° 5´ S., in preference to Pemba, 5° 6´ S., which lay too far out of the course, and in preference to Momfia, 7° 50´ S. (though more doubtfully), because of its being by no means conspicuous, whereas Zangibar was so prominent and obvious above the other two, that it might well attract the particular attention of navigators, and its distance from the mainland is at the same time so nearly in accordance with that given in the _Periplûs_ as to counterbalance all other objections. A writer in Smith’s _Classical Geography_, who seems to have overlooked the indications of the distances both of Ptolemy and the _Periplûs_, assigns it a position much further to the north than is reconcilable with these distances. He places it about a degree south from the mouth of the River Juba or Govind, just where an opening in the coral-reefs is now found. “The coasting voyage,” he says, “steering S. W., reached the island on the east side—a proof that it was close to the main.... It is true the navigator says it was 300 stadia from the mainland; but as there is no reason to suppose that he surveyed the island, this distance must be taken to signify the estimated width of the northern inlet separating the island from the main, and this estimate is probably much exaggerated. The mode of fishing with baskets is still practised in the Juba islands and along this coast. The formation of the coast of E. Africa in these latitudes—where the hills or downs upon the coast are all formed of a coral conglomerate comprising fragments of madrepore, shell and sand, renders it likely that the island which was close to the main 16 or 17 centuries ago, should now be united to it. Granting this theory of gradual transformation of the coast-line, the +Menouthias+ of the _Periplûs_ may be supposed to have stood in what is now the rich garden-land of +Shamba+, where the rivers carrying down mud to mingle with the marine deposit of coral drift covered the choked-up estuary with a rich soil.”