The scenery varies extremely at different parts of the surface. Towards the flat portion, where the English reside, the aspect is as bleak and inhospitable as can be imagined: a thin stratum of marshy or sandy soil covers a tabular mass of cold red sandstone; and there is not a tree, and scarcely a shrub to be seen, except occasional clumps of Pandanus. The low white bungalows are few in number, and very scattered, some of them being a mile asunder, enclosed with stone walls and shrubs; and a small white church, disused on account of the damp, stands lonely in the centre of all.

The views from the margins of this plateau are magnificent: 4000 feet below are bay-like valleys, carpetted as with green velvet, from which rise tall palms, tree-ferns with spreading crowns, and rattans shooting their pointed heads, surrounded with feathery foliage, as with ostrich plumes, far above the great trees. Beyond are the Jheels, looking like a broad shallow sea with the tide half out, bounded in the blue distance by the low-hills of Tipperah. To the right and left are the scarped red rocks and roaring waterfalls, shooting far over the cliff's, and then arching their necks as they expand in feathery foam, over which rainbows float, forming and dissolving as the wind sways the curtains of spray from side to side.

To the south of Churra the lime and coal measures rise abruptly in flat-topped craggy hills, covered with brushwood and small trees. Similar hills are seen far westward across the intervening valleys in the Garrow country, rising in a series of steep isolated ranges, 300 to 400 feet above the general level of the country, and always skirting the south face of the mountains. Considerable caverns penetrate the limestone, the broken surface of which rock presents many picturesque and beautiful spots, like the same rocks in England.

Westward the plateau becomes very hilly, bare, and grassy, with the streams broad and full, but superficial and rocky, precipitating themselves in low cascades over tabular masses of sand-stone. At Mamloo their beds are deeper, and full of brushwood, and a splendid valley and amphitheatre of red cliffs and cascades, rivalling those of Moosmai (chapter xxvii), bursts suddenly into view. Mamloo is a large village, on the top of a spur, to the westward: it is buried in a small forest, particularly rich in plants, and is defended by a stone wall behind: the only road is tunnelled through the sandstone rock, under the wall; and the spur on either side dips precipitously, so that the place is almost impregnable if properly defended. A sanguinary conflict took place here between the British and the Khasias, which terminated in the latter being driven over the precipices, beneath which many of them were shot. The fan-palm, Chamaerops Khasiana ("Pakha," Khas.), grows on the cliff's near Mamloo: it may be seen on looking over the edge of the plateau, its long curved trunk rising out of the naked rocks, but its site is generally inaccessible;* [This species is very closely allied to, if not identical with P. Martiana of Nepal; which ascends to 8000 feet in the western Himalaya, where it is annually covered with snow: it is not found in Sikkim, but an allied species occurs in Affghanistan, called P. Ritcheana: the dwarf palm of southern Europe is a fourth species.] while near it grows the Saxifragis ciliaris of our English gardens, a common plant in the north-west Himalaya, but extremely scarce in Sikkim and the Khasia mountains.

Illustration—MAMLOO CASCADES.

The descent of the Mamloo spur is by steps, alternating with pebbly flats, for 1500 feet, to a saddle which connects the Churra hills with those of Lisouplang to the westward. The rise is along a very steep narrow ridge to a broad long grassy hill, 3,500 feet high, whence an extremely steep descent leads to the valley of the Boga-panee, and the great mart of Chela, which is at the embouchure of that river. The transverse valley thus formed by the Mamloo spur, is full of orange groves, whose brilliant green is particularly conspicuous from above. At the saddle below Mamloo are some jasper rocks, which are the sandstone altered by basalt. Fossil shells are recorded to have been found by Dr. M'Lelland* [See a paper on the geology of the Khasia mountains by Dr. M'Lelland in the "Bengal Asiatic Society's Journal.">[ on some of the flats, which he considers to be raised beaches: but we sought in vain for any evidence of this theory beyond the pebbles, whose rounding we attributed to the action of superficial streams.

It is extremely difficult to give within the limits of this narrative any idea of the Khasia flora, which is, in extent and number of fine plants, the richest in India, and probably in all Asia. We collected upwards of 2000 flowering plants within ten miles of the station of Churra, besides 150 ferns, and a profusion of mosses, lichens, and fungi. This extraordinary exuberance of species is not so much attributable to the elevation, for the whole Sikkim Himalaya (three times more elevated) does not contain 500 more flowering plants, and far fewer ferns, etc.; but to the variety of exposures; namely, 1. the Jheels, 2. the tropical jungles, both in deep, hot, and wet valleys, and on drier slopes; 3. the rocks; 4. the bleak table-lands and stony soils; 5. the moor-like uplands, naked and exposed, where many species and genera appear at 5000 to 6000 feet, which are not found on the outer ranges of Sikkim under 10,000.* [As Thalictrum, Anemone, primrose, cowslip, Tofieldia, Yew, Pine, Saxifrage, Delphinium, Pedicularis.] In fact, strange as it may appear, owing to this last cause, the temperate flora descends fully 4000 feet lower in the latitude of Khasia (25 degrees N.) than in that of Sikkim (27 degrees N.), though the former is two degrees nearer the equator.

The Pandanus alone forms a conspicuous feature in the immediate vicinity of Churra; while the small woods about Mamloo, Moosmai, and the coal-pits, are composed of Symplocos, laurels, brambles, and jasmines, mixed with small oaks and Photinia, and many tropical genera of trees and shrubs.

Orchideae are, perhaps, the largest natural order in the Khasia, where fully 250 kinds grow, chiefly on trees and rocks, but many are terrestrial, inhabiting damp woods and grassy slopes. I doubt whether in any other part of the globe the species of orchids outnumber those of any other natural order, or form so large a proportion of the flora. Balsams are next in relative abundance (about twenty-five), both tropical and temperate kinds, of great beauty and variety in colour, form, and size of blossom. Palms amount to fourteen, of which the Chamaerops and Arenga are the only genera not found in Sikkim. Of bamboos there are also fifteen, and of other grasses 150, which is an immense proportion, considering that the Indian flora (including those of Ceylon, Kashmir, and all the Himalaya), hardly contains 400. Scitamineae also are abundant, and extremely beautiful; we collected thirty-seven kinds.

No rhododendron grows at Churra, but several species occur a little further north: there is but one pine (P. Khasiana) besides the yew, (and two Podocarpi), and that is only found in the drier interior regions. Singular to say, it is a species not seen in the Himalaya or elsewhere, but very nearly allied to Pinua longifolia,* [Cone-bearing pines with long leaves, like the common Scotch fir, are found in Asia, and as far south as the Equator (in Borneo) and also inhabit Arracan, the Malay Peninsula, Sumatra, and South China. It is a very remarkable fact that no Gymnospermous tree inhabits the Peninsula of India; not even the genus Podocarpus, which includes most of the tropical Gymnosperms, and is technically coniferous, and has glandular woody fibre; though like the yew it bears berries. Two species of this genus are found in the Khasia, and one advances as far west as Nepal. The absence of oaks and of the above genera (Podocarpus and Pinus) is one of the most characteristic differences between the botany of the east and west shores of the Bay of Bengal.] though more closely resembling the Scotch fir than that tree does.