MADAGASCAR 169 Tolcauic >elt of the interior, the chief rivers now to the west and north west sides of the island. The eastern streams are all less in size, except the Mangoro, which flows for some distance parallel with the coast. Few of them, therefore, are of much service for navigation, except for the light-draught native canoes, and almost all of them are more or less closed at their outlets by sand-bars. Commencing at the southern point and going northward, on the eastern side, the principal rivers are the Mananara, Manambava, Matitanana, Mananjara, Oniv6 and Mangoro, Maningory, and the Anjahanambo at the head of Antongil Bay, besides numerous smaller streams. On the north-west coast, going southward, are the Sofia and Mahajamba, falling into Majambo Bay, the B6tsib6ka with the Ikiopa, the great drains of the northern central provinces, and forming unitedly the second largest river of the island and falling into Bembatoka Bay, the Manjaray, Manambolo, Tsiribihina or Onimainty, the largest river of Madagascar, draining by its tributaries the Kitsamby, Mahajilo, and Mania the central parts of the island, the Morondava, Maharivo,Mang6ky, the third largest river, the Mahanomby, Fiherdnana, and Onilahy. Of these western rivers the Betsiboka could be ascended by steamers of light draught for about 90 miles, and the Tsiribihina is also navigable for a considerable distance. The former is about 300 miles long ; the latter is somewhat less, but by its affluents spreads over a greater extent of country. It brings down so large a body of water that the sea is said to be fresh 3 miles from the land. But owing to the height of the interior of Madagascar there is no uninterrupted water-communication with it from the sea by any of the rivers, which are all crossed by rocky bars, and in some cases by grand waterfalls, as on the Mania. The eastern rivers cut their way through the ramparts of the high land by magnificent gorges, amidst dense forest, and descend by a succession of rapids and cataracts. The Matitanana, whose falls were first seen by the present writer in 1876, descends at one plunge some 400 or 500 feet. On the eastern side of Madagascar the contest between the fresh water of the rivers and the sea has caused the formation of a long chain of lagoons for nearly three hundred miles. In many parts these look like a river following the coast-line, but frequently they spread out into extensive sheets of water. So short is the distance between these that, by cutting about 30 miles of canal to connect them, a continuous water-way could be formed for 2GO miles along the coast. This will doubtless be accomplished at some future time, with great benefit to the commerce of the country. Besides these lagoons, there are few lakes of any size in Madagascar, although there were probably some very extensive ones in a recent geological epoch. Of one of the largest of these the Alaotra Lake in the Antsihanaka plain is the relic ; it is about 25 miles long. Next comes Itasy, in western Ime rina, about 8 miles long; and a large lake is reported by the natives to be formed by an expansion of the river Mangoky. Two salt lakes are said to exist near the south-west coast. Among the many new facts brought to light by recent research in Madagascar is the evidence of very widespread and powerful subterranean action throughout a great part of the island, apparently extending almost unbroken from the south-east to the north-west and extreme north. This volcanic belt is part of a line which has its northern extremity in the Comoro Islands, all of which are volcanic in origin, and where, in Great Comoro, there is a still active vent. There is now no active volcano in Madagascar, but a large number of extinct cones have been observed in various parts of the country. In the central province of Tm6rina, within an arc of about 90 miles round the mass of Ankcaratra, Dr Mullens counted a hundred craters. Others are found farther north, in the Antsihanaka province and the Mandritsara valley, and on the north-west coast and its islands, the great mountain Ambohitra being an old volcano. Others have been observed towards the southern extremity of the higher region of the island, as well as columnar basalt, and beds of lava rock, pumice, and ash. Slight shocks of earthquake are felt every year in Madagascar, and other signs of subterranean action are evident in the hot springs which occur at several places in the central and eastern provinces. Several of these are sulphurous and medicinal, and have been found efficacious in skin diseases. The geology of Madagascar has as yet been very imper- Geology, fectly investigated, for few travellers have possessed the special scientific knowledge requisite to give much value to their observations ; and hardly anything has yet been done towards making collections of fossils, or in procuring specimens of rocks and minerals. There are, however, a few facts of a general character which are easily recogniz able. In the first place, the upper region of the island already mentioned appears to consist chiefly of Primary and unstratified rocks granite, gneiss, and basalt which form the highest points of the hills, and present most varied and picturesque outlines, resembling titanic castles, cathedrals, domes, pyramids, and spires. The general face of the country consist^ of bare rolling moors, with a great amount of bright red and light brown clays, while the valleys have a rich vegetable soil of bluish-black alluvium. No stratified or fossiliferous rocks have yet been discovered in this upper part of the island, which appears to be very ancient land, and during portions of the Secondary and Tertiary periods probably formed the entire island, then about from a third to two-fifths of its present size, while the extensive southern and western plains were again and again submerged. The lower portions of Madagascar do not, as far as is Fossils yet known, much exceed from 300 to 600 feet in height and above the sea-level (except, of course, the three chains of minera l s - hills in the south-west). They appear in several localities to consist of strata of the Secondary period, with fossils of the Neocomian age belonging to the genera Nerinea, Turritdla, Ammonites, Terebratula, Khynconella, Neritina, and Echinoderms, and also Foraminifera of the genera Alveolina, Orbitoides, Triloculina, &c. There are also beds of a much later age, containing fossils of recently extinct gigantic tortoises, hippopotami, and struthious birds. In addition to the rocks already mentioned as found in the higher portions of the island, there are also slate, mica schist, grey wacke, chert, pink and white quartz, and an unstratified limestone deposited by hot springs. Iron exists in great abundance in the central parts of the country, and copper and silver are said to have been found in small quantities, but are not worked. Antimony seems to be plentiful in the north, and rock-salt, iron pyrites, plumbago, and various ochres and coloured earths are among the mineral products. On the north-west coast thin beds of lignite, suitable for steam coal, occur, but no true coal has yet been discovered. The climate varies very much in different parts of the Climate, country. In the high interior districts it resembles that of the temperate zones, with no intense heat, and is quite old during the nights in winter. These parts of the island are therefore tolerably healthy for Europeans. But the coasts are much hotter, especially on the western side ; and, from the large amount of marsh and lagoon, malarial fever is prevalent, and frequently fatal both to Europeans and to natives from the interior. The seasons are two the hot and rainy season from November to April, and the cool and dry season during the rest of the year. Rain indeed falls almost all the year round on the eastern coast, which is exposed to the vapour-laden south-east trade
XV. 22