-GEOGRAPHY.] INDIA 735 narrow maritime tract and the Irawadi runs a backbone of lofty ranges. These ranges, known as the Yoma (Roma) mountains, are covered with dense forests, and both his torically and geographically separate the Irawadi valley from the strip of coast. The Yoina (Roma) ranges have peaks exceeding 4000 feet, and culminate in the Blue Mountain (7100 feet). They are crossed bypasses, one of which, the An or Aeng, rises to 4668 feet above the sea- level. A thousand creeks indent the seaboard ; and the whole of the level country, both on the coast and in the Irawadi valley, forms one vast rice-field. The river floats down an abundant supply of teak and bamboos from the north. Tobacco, of an excellent quality, supplies the little cigars which all Burmese (men, women, and children) smoke. Arakan and Pegu, or the provinces of the coast strip and the Irawadi valley, contain mineral oil-springs. Teuasserim forms a long narrow maritime province, which runs from the mouths of the Irawadi southward to Point Victoria, where the British territory adjoins Siam. It is rich in tin mines, and contains iron-ores equal to the finest Swedish, besides gold and copper in smaller quantities, and a very pure limestone. Rice and timber form the staple exports of Burmah ; and rice is also the universal food of the people. British Burmah, with Tenasserim, has an area of 88,556 square miles, and had a population, in 1876, of just under 3 millions of persons. GEOLOGY. For geological purposes India may be mapped out into the three geographical divisions of the Himalayan region, the Indo-Gangetic plain, and Peninsular India. na- The Himalayan Region. The geology of this district is far more complex and less fully known than that of the Peninsular area. Until the ground has been carefully gone over by the Geological Survey, many points must remain doubtful; probably even then the problems will not be fully solved, as large areas of the Himalayas (Nepal and Bhutan) are at present inaccessible to Europeans. The eiss. oldest rock of the Himalayas is gneiss, but its age is quite unknown. It generally differs in character from the gneiss of the Peninsula, and also from that of Assam and Burmah. The Himalayan gneiss is usually white and grey, its felspars being orthochse and albite ; it contains much mica and mica schist, and is generally much more uniform in character than the gneiss of the Peninsula, The latter is usually pink, its felspar being orthoclase and oligoclase ; it contains little mica schist, but often has quartzite and hornblendic rock. Hornblende occurs in the syenitic gneiss of the Northern Himalayan (or Ladakh) range. The itral Central Himalayan region may be roughly described as ! ic consisting of two gneissic axes, with a trough or synclinal valley between them, in which fossiliferous beds have been deposited and are now preserved. The gneiss of the southern or main axis (the " central gneiss " of Dr Stoliczka) is the oldest ; that of the northern or Ladakh axis come.s next in age. The gneiss of the Ladakh axis is generally syenitic, or is that variety of the Himalayan gneiss already described as containing hornblende. It is probably an extremely altered condition of ordinary marine sediment. The gneiss of the central axis is the ordinary kind ; it is penetrated by granite, which ranges along some of the highest peaks. Between these two gneissic axes occurs the basin-shaped valley, or the Hundes and Zanskar synclinal. In this valley fossiliferous rocks are preserved, giving representatives of the Silurian, Carboniferous, Trias- .sic, Jurassic, and Cretaceous formations. All these seem there to have followed each other without important breaks or unconformities; but after the deposition of the Cretaceous rocks of the Himalayan region, there appear to have been important changes in physical geography. The Nummulitic (Eocene) strata were laid down on the eroded edges of some of the older beds, and in a long trough within the Silurian gneiss of the Ladakh axis. On the south of this true Himalayan region there is a band of country known as the Lower Himalaya, in which the beds are often greatly Lower disturbed, and even completely inverted, over great areas, Hima- the old gneiss apparently overlying the sedimentary rocks. layas- This Lower Himalayan region is about 50 miles wide, and consists of irregular ridges, varying from 5000 to 8000 feet in height, and sometimes reaching 12,000 feet. Resting upon the gneiss, but often through inversion apparently underlying it, in the neighbourhood of Simla, is a series of unfossiliferous beds (schists, quartzites, sand stones, shales, limestones, &c.) known in descending order as the Kr61, Infra-Kro l, Blaini, and Infra-Blaini beds. In the Kr6l beds is a massive limestone (Kr61 limestone) probably representing the limestone of the Pir Panjal range, which is most likely of Carboniferous age. The Blaini and Infra-Blaini beds are probably Silurian. The Lower Himalayan range ends at the Sutlej valley, west of which the continuation of the central range is followed immediately by the third or Sub-Himalayan Sub range. This occurs almost always on the south of the ilira ^- Lower Himalayas; it is composed of later Tertiary - as< rocks (Siwaliks, &c.), which range parallel with the main chain. Generally the Sub-Himalayas consist of two ranges, separated by a broad flat valley (" dun " or " doon ") ; the southern slope, overlooking the great Indo-Gangetic plain, is usually the steepest. Below Naini Till and Darjiling (Darjeeling), the sub-Himalayan range is wanting ; on the Bhutan frontier the whole range is occasionally absent, and then the great plain slopes up to the base of the Lower Himalayan region. It is within the Sub-Himalayan range that the famous Siwalik Siwdlika. beds occur, long since known for their vast stores of extinct mammalia. Of about the same age are the Manchhar beds of Sind, which also contain a rich mammalian fauna. The Lower Manchhars probably correspond to the Nahan beds, the lowest of the Siwaliks ; they rest upon the Gaj beds, which are probably Upper Miocene. From this it would seem that the lowest Siwaliks are not older than Upper Miocene. The higher Siwalik beds are considered by Mr W. T. Blanford to be Pliocene, and to this later period he also refers the mammalian beds of Pikermi in Greece. These have a large number of fossils in common with the Siwaliks ; but they contain, at their base, a marine band with Pliocene shells, The Manchhar and Siwalik beds are chiefly of freshwater origin. The Salt Range in the north-west of the Punjab has, in Salt addition to its economic value, a special geological im- Range, portance ; and from that point of view it is one of the most interesting districts in India. Representatives of most of the great European formations of Silurian and later epochs are found there; and throughout all the vast length of time represented by these formations there is here no direct evidence of any important break in succession, or unconformity. The lowest beds (salt marl, probably Silurian) and the highest (Siwaliks) are found throughout the range. But the others cannot all be traced con tinuously throughout ; some occur well developed in one place, some in another. All the principal fossiliferons beds of the Jurassic, Triassic, and Carboniferous formations are confined to the western part of the range. The Indo-Gangetic Plain covers an area of about 300,000 i m lo- square miles, and varies in width from 90 to nearly 300 Gangetic miles. It rises very gradually from the sea at either end ; IJlain - the lowest point of the watershed between the Punjab rivers and the Ganges is about 924 feet above the sea. This
point, by a line measured down the valley, but not fol-
