Page:Africa by Élisée Reclus, Volume 2.djvu/138

 110 NORTH-WEST AFRICA. evaporation, whence the brackish character of its waters. The course of the two streams across the plain does not appear to have been perceptibly modified within the historic period, for the old Roman highway runs directly through this alluvial basin, at intervals touching the windings along the left bank. At the confluence of the Beja descending from the north, the Mejerda enters the series of narrow tortuous gorges by which it has forced its way through the surrounding chalk formation. But at the junction of the Zerga it encountered harder rocks, which it was unable to pierce. Hence it is here abruptly deflected southwards to its confluence with the Siliana, where it finds an easier outlet towards the north-east. Below the barrage constructed in 1622 by Dutch engi- neers the Mejerda flows by the west foot of the Jebel Ahmor to the alluvial plain through which it discharges into the shallow El-Bahira (Ghar-el-Melah) lagoon. This basin, which in the seventeenth century was " the finest harbour in Barbary," and which still communicates with the sea through a small channel accessible to fishing-smacks, has been gradually filled in by the alluvia of the Mejerda. Its depth, which now nowhere exceeds 5 or 6 feet, appears to have been diminished by 30 feet during the last hundred years. It will probably disappear altogether bt^fore the end of the century, just as the older Gulf of Utica in the same delta has been converted during the lust one thousand six hundred years into the marshy depression of Mabtuha. The Er-Ruan sebkha and other lagoons in this district are also being slowly effaced, while the shore line between Cape Sidi Ali-el-Mekki and the hills of Carthage is continually advancing seawards. According to Tissot, the land has here encroached on the sea to a probable extent of 100 square miles in the course of the last two thousand one hundred years. During the historic period the Mejerda has often shifted its bed, and by the aid of the old writers and a careful survey of its lower course, it might be possible to reconstruct the map of its delta at different epochs. In the time of the Carthaginians, the Makarath or Bagrada skirted the north foot of the Jebel Ahmor, leaving on the left a ridge of insular rocks from 100 to 150 feet high, and reaching the sea at a point just north of Cape Carthage. The old bed can still be traced by the sands and gravel, in which now grow a few oleander bushes. Subsequently two other beds were excavated farther north, both of which had also their origin in the gorge at the north foot of the Jebel Ahmor. But the present channel runs due north along the depression of the old Lake of Utica, terminating just south of the headland at the northern extremity of the Gulf of Tunis. South of the Mejerda there are no permanent rivers, their mouths being closed by a bank of sand for at least several months in the year. But communication with the sea is effected by one lagoon, the Lake of Tunis, a second Bahira, similar to that into which the Mejerda falls ; it is somewhat larger, however, and attains a greater depth, being some 6 feet in the deepest parts. Its approach is formed by an artificial canal, which has replaced a natural channel farther south, and which will admit vessels drawing over 4 feet of water ; but its waters are rendered impure by the sewage of Tunis, and hence the banks are unhealthy. Like those of the Mejerda delta, this lake, which was formerly crowded by the Roman and I