Page:Africa by Élisée Reclus, Volume 1.djvu/78

 lake, which rises slowly during the floods, and falls imperceptibly during the dry season. The yearly discrepancy between the levels of the lake scarcely exceeds forty inches.

The Abai, its largest affluent, rises at Gish Abai, near the north-east foot of Mount Denguiya, some 60 miles from the lake. The Portuguese colony settled in this region towards the end of the sixteenth century certainly visited the sources of the Abai; but they were first described by the Jesuit Paez, who tells us that the water, oozing from a marshy field, is collected in a limpid lake, supposed by the natives to be "unfathomable" because they cannot reach the bottom with their spears. Thence trickles a rivulet, whose course can be traced only by a surface growth of waving grasses, but which over a mile lower down emerges in the open. This is the brook to which both the Portuguese and Bruce gave the name of the Nile. The fiery exhalations often seen flitting about its source, doubtless will-o'-the-wisps, have earned for the Abai the veneration of the natives, who still sacrifice animals to the local river genius. The stream has a width of over 30 feet where it reaches the south-west inlet of the lake, and where its turbid waters have developed an alluvial delta of considerable size. But the outlet, which retains the name of Abai, is a limpid blue current fully entitled to its Arabic designation of Bahr-el-Azraq. Like most other rivers which are at once affluents and emissaries of lacustrine basins, the Abai is constantly said to traverse lake Tsana without mingling with its water. But although such a phenomenon is well-nigh impossible, a perceptible current certainly appears to set steadily from the mouth of the affluent to that of the outflow.

Tsana cannot be compared for size to the great equatorial lakes. According to Stecker's survey, it has a superficial area of scarcely 1,200 square miles, or less than the twentieth part of Victoria Nyanza. But it must have formerly been more extensive than at present, as is evident from some alluvial plains found especially on the north side. It has the general form of a crater, except towards the south, where it develops into a gulf in the direction of its outlet. Hence the hypothesis advanced by several authors that it may have originally been a vast volcanic cone, and certainly some of the rounded islets in the neighbouring waters look like extinct craters, while the surrounding shores are diversified with bold basaltic headlands. The central part of the basin is probably very deep, for even in the southern inlet Stecker recorded a depth of 240 feet. The water is extremely pure, and as pleasant to the taste as that of the Nile. Towards the south-west the shore is fringed with dense masses of a long light reed (arundo donax), with which the natives construct their tankuas, frail skiffs or rafts propelled by two or four oars, and provided with raised benches to keep the cargo dry. But very little traffic is carried on from coast to coast. Through the foliage which encircles this lovely sheet of water, little is visible except the distant hills and the conic islets rising above the sparkling surface. Herds of hippopotami are often seen on the shores, but there are no crocodiles in the lake, although the Abai below the cataract is infested by these reptiles. Nor has any European traveller seen the aila, a small species of manatee said by the natives to inhabit its waters; which, however, abound