The country continued a grassy level, with marshes and rice cultivation, to the first range of hills, beyond which the river is unnavigable; there also a forest commences, of oaks, figs, and the common trees of east Bengal. The road hence was a good one, cut by Sepoys across the dividing ranges, the first of which is not 500 feet high. On the ascent bamboos abound, of the kind called Tuldah or Dulloah, which has long very thin-walled joints; it attains no great size, but is remarkably gregarious. On the east side of the range, the road runs through soft shales and beds of clay, and conglomerates, descending to a broad valley covered with gigantic scattered timber-trees of jarool, acacia, Diospyros, Urticeae, and Bauhiniae, rearing their enormous trunks above the bamboo jungle: immense rattan-canes wound through the forest, and in the gullies were groves of two kinds of tree-fern, two of Areca, Wallichia palm, screw-pine, and Dracaena. Wild rice grew abundantly in the marshes, with tal1 grasses; and Cardiopteris* [A remarkable plant of unknown affinity; see Brown and Bennett, "Flora Java:" it is found in the Assam valley and Chittagong.] covered the trees for upwards of sixty feet, like hops, with a mass of pale-green foliage, and dry white glistening seed-vessels. This forest differed from those of the Silhet and Khasia mountains, especially in the abundance of bamboo jungle, which is, I believe, the prevalent feature of the low hills in Birmah, Ava, and Munnipore; also in the gigantic size of the rattans, 1arger palms, and different forest trees, and in the scanty undergrowth of herbs and bushes. I only saw, however, the skirts of the forest; the mountains further east, which I am told rise several thousand feet in limestone cliffs, are doubtless richer in herbaceous plants.

The climate of Cachar partakes of that of the Jheels in its damp equable character: during our stay the weather was fine, and dense fogs formed in the morning: the mean maximum was 80 degrees, minimum 58.4 degrees.* [The temperature does not rise above 90 degrees in summer, nor sink below 45 degrees or 50 degrees in January: forty-seven comparative observations with Calcutta showed the mean temperature to be 1.8 degrees lower at Silchar, and the air damper, the saturation point being, at Calcutta 0.3791, at Silchar 0.4379.]

The annual rain-fall in 1850 was 111.60 inches, according to a register kindly given me by Captain Verner. There are few mosquitos, which is one of the most curious facts in the geographical distribution of these capricious bloodsuckers; for the locality is surrounded by swamps, and they swarm at Silhet, and on the river lower down. Both on the passage up and down, we were tormented in our canoes by them for eighty or ninety miles above Silhet, and thence onwards to Cachar we were free.

On the 30th of November, we were preparing for our return to Silhet, and our canoes were loading, when we were surprised by a loud rushing noise, and saw a high wave coming down the river, swamping every boat that remained on its banks, whilst most of those that pushed out into the stream, escaped with a violent rocking. It was caused by a slip of the bank three quarters of a mile up the stream, of no great size, but which propagated a high wave. This appeared to move on at about the rate of a mile in three or four minutes, giving plenty of time for our boatmen to push out from the land on hearing the shouts of those first overtaken by the calamity; but they were too timid, and consequently one of our canoes, full of papers, instruments, and clothes, was swamped. Happily our dried collections were not embarked, and the hot sun repaired much of the damage.

We left in the evening of the 2nd of December, and proceeded to Silhet, where we were kindly received by Mr. Stainforth, the district judge. Silhet, the capital of the district of the same name, is a large Mahometan town, occupying a slightly raised part of the Jheels, where many of the Teelas seem joined together by beds of gravel and sand. In the rains it, is surrounded by water, and all communication with other parts is by boats: in winter, Jynteapore and Pundua may be reached by land, crossing creeks innumerable on the way. Mr. Stainforth's house, like those of most of the other Europeans, occupies the top of one of the Teelas, 150 feet high, and is surrounded by fine spreading oaks,* [It is not generally known that oaks are often very tropical plants; not only abounding at low elevations in the mountains, but descending in abundance to the level of the sea. Though unknown in Ceylon, the Peninsula of India, tropical Africa, or South America, they abound in the hot valleys of the Eastern Himalaya, East Bengal, Malay Peninsula, and Indian islands; where perhaps more species grow than in any other part of the world. Such facts as this disturb our preconceived notions of the geographical distribution of the most familiar tribes of plants, and throw great doubt on the conclusions which fossil plants are supposed to indicate.] Garcinia, and Diospyros trees. The rock of which the hill is composed, is a slag-like ochreous sandstone, covered in most places with a shrubbery of rose-flowered Melastoma, and some peculiar plants.* [Gelonium, Adelia, Moacurra, Linostoma, Justicia, Trophis, Connarus, Ixora, Congea, Dalhousiea, Grewia, Myrsine, Buttneria; and on the shady exposures a Calamus, Briedelia, and various ferns.]

Broad flat valleys divide the hills, and are beautifully clothed with a bright green jungle of small palms, and many kinds of ferns. In sandy places, blue-flowered Burmannia, Hypoxis, and other pretty tropical annuals, expand their blossoms, with an inconspicuous Stylidium, a plant belonging to a small natural family, whose limits are so confined to New Holland, that this is almost the only kind that does not grow in that continent. Where the ground is swampy, dwarf Pandanus abounds, with the gigantic nettle, Urtica crenulata ("Mealum-ma" of Sikkim, see chapter xxiv).

The most interesting botanical ramble about Silhet is to the tree-fern groves on the path to Jynteapore, following the bottoms of shallow valleys between the Teelas, and along clear streams, up whose beds we waded for some miles, under an arching canopy of tropical shrubs, trees, and climbers, tall grasses, screw-pines, and Aroideae. In the narrower parts of the valleys the tree-ferns are numerous on the slopes, rearing their slender brown trunks forty feet high, with feathery crowns of foliage, through which the sun-beams trembled on the broad shining foliage of the tropical herbage below.

Silhet, though hot and damp, is remarkably healthy, and does not differ materially in temperature from Silchar, though it is more equable and humid.* [During our stay of five days the mean maximum temperature was 74 degrees, minimum 64.8 degrees: that of thirty-two observations compared with Calcutta show that Silhet is only 1.7 degrees cooler, though Mr. Stainforth's house is upwards of 2 degrees further north, and 160 feet more elevated. A thermometer sunk two feet seven inches, stood at 73.5 degrees. The relative saturation-points were, Calcutta .633, Silhet .821.] It derives some interest from having been first brought into notice by the enterprise of one of the Lindsays of Balcarres, at a time when the pioneers of commerce in India encountered great hardships and much personal danger. Mr. Lindsay, a writer in the service of the East India Company, established a factory at Silhet, and commenced the lime trade with Calcutta,* [For an account of the early settlement of Silhet, see "Lives of the Lindsays," by Lord Lindsay.] reaping an enormous fortune himself, and laying the foundation of that prosperity amongst the people which has been much advanced by the exertions of the Inglis family, and has steadily progressed under the protecting rule of the Indian government.

From Silhet we took large boats to navigate the Burrampooter and Megna, to their embouchure in the Bay of Bengal at Noacolly, a distance of 250 miles, whence we were to proceed across the head of the bay to Chittagong, about 100 miles farther. We left on the 7th of December, and arrived at Chattuc on the 9th, where we met our Khasia collectors with large loads of plants, and paid them off. The river was now low, and presented a busy scene, from the numerous trading boats being confined to its fewer and deeper channels. Long grasses and sedges (Arundo, Saccharum and Scleria), were cut, and stacked along the water's edge, in huge brown piles, for export and thatching.

On the 13th December, we entered the broad stream of the Megna. Rice is cultivated along the mud flats left by the annual floods, and the banks are lower and less defined than in the Soormah, and support no long grasses or bushes. Enormous islets of living water-grasses (Oplismenus stagninus) and other plants, floated past, and birds became more numerous, especially martins and egrets. The sun was hot, but the weather otherwise cool and pleasant: the mean temperature was nearly that of Calcutta, 69.7 degrees, but the atmosphere was more humid.* [The river-water was greenish, and a little cooler (73.8 degrees) than that of the Soormah (74.3 degrees), which was brown and muddy. The barometer on the Soormah stood 0.028 inch higher than that of Calcutta (on the mean of thirty-eight observations), whereas on the Megna the pressure was 0.010 higher. As Calcutta is eighteen feet above the level of the Bay of Bengal, this shows that the Megna (which has no perceptible current) is at the level of the sea, and that either the Soormah is upwards of thirty feet above that level, or that the atmospheric pressure there, and at this season, is less than at Calcutta, which, as I have hinted at chapter xxvii, is probably the case.]