But although they form a chain that seems to indicate a past union of Sumatra and Burma, investigation proves that this is far from being the case. For soundings in this part of the ocean show that between the Nicobars and the group of islands adjacent to the north-east point of Sumatra--entering from the open sea to the west, and thence trending north between the Andamans and Malay Peninsula almost as far as the lat.i.tude of Narkondam--there runs a long tongue of depressed ocean bed with depths everywhere over 1000 fathoms. This fact, together with the shallowness of the sea-bottom around and connecting the Andamans with the Arakan Yoma Peninsula, suggests the inference that the former were at one time past the termination of a seaward extension from Cape Negrais of the Arakan Yoma Range--a conclusion that is in some degree emphasized by the zoological and botanical conditions common to the two.

The princ.i.p.al islands are, Great and Little Andaman, Rutland and the Labyrinths, the Archipelago, North Sentinel, Interview Island, Landfall Island, and the Cocos, but there are many smaller adjacent, while to the eastward are the off-lying volcanic islets of Narkondam and Barren Island. The total area of the group is 2508 square miles.[84]

Great Andaman--in which may be included Landfall and Rutland Islands, for the whole land ma.s.s is so compact and divided up by such narrow shallow straits that it appears to be one single island that has been broken up by subsidence and adjoining volcanic action--is 142 miles long, and 17 miles broad at its widest point.

There are generally stated to be two straits, but as one of them bifurcates, the Great Andaman proper is really cut into four parts.

Austin Strait, which divides North from Middle Andaman, is very narrow and intricate, and not to be traversed by boats at low tide; but the Andaman Strait,--generally 2 to 3 cables wide--which separates South from Middle Andaman at a spot where the hills are lower than elsewhere, although intricate, and possessing a bar at its eastern mouth with a depth of 9 or 10 feet at low water, has depths from 10 to 14 fathoms throughout the narrower part, and nowhere less than 3 fathoms at low water. The stream is never strong, and the R.I.M.S. _Investigator_ pa.s.sed through three times while surveying the islands in 1888.

Homfray Strait cuts off Baratang Island from Middle Andaman, and joins the Andaman at its western mouth. It is intricate and rocky, but has good depths, except at the eastern entrance, where there is a broad bar of 8 feet. The tidal stream is weak, and the narrowest part is 60 yards across.

The surface of Great Andaman is extremely irregular, and a central range of mountains runs from north to south, with an escarped face on the east, and a sloping declivity on the west, where marshy localities abound.

The highest point is Saddle Hill (2400 feet) in North Andaman: Mount Harriet (1200 feet) stands on the north sh.o.r.e of Port Blair; and Ford's Peak, in Rutland Island, rises 1400 feet; while there are half a dozen unnamed summits with heights between 1000 feet and 1700 feet.

Narkondam rises 2330 feet, from an oval-shaped base whose greatest diameter is 2 miles, and the crater walls of Barren Island, 2 square miles in area, attain an elevation of 1158 feet.

Little Andaman, some 25 miles south of Rutland, 23 miles long and 17 miles wide, with an area of about 220 square miles, is, on the contrary, level throughout, and gradually rises to a height of 600 feet in the centre. None of the other islands save Rutland attain this elevation.

Owing to its shape and conformation, there are no rivers and but few streams on Great Andaman, and during the dry season--January to April--there is some scarcity of water. Several creeks, however, are of sufficient depth to allow pa.s.sage of boats for some distance into the interior. In the South Andaman the greater part of the drainage runs into the creeks, which ultimately leads off to the eastern sh.o.r.e, and in the North and Middle Andaman the bulk of the drainage seems to flow through gaps in the eastern range.

Little Andaman is swampy in many parts, and possesses a few small creeks.

On the western side, in which direction Great Andaman slopes gradually, banks of coral occur at distances of 20 and 25 miles from land. There are three of these, varying in length from 9 to 25 miles, all composed of dead coral and sand, with here and there single bunches of live coral 1 or 2 feet high. The water, which is so clear that on a calm day 8 or 9 fathoms looks like 20 feet, varies in its least depths from 3-3/4 to 6 fathoms, and, judged from the appearance of the bottom and the absence of reef-building coral, it seems probable that the surface _debris_ of the banks is disturbed by the send of the sea, and that the rollers topple and break on the middle bank in the south-west monsoon, though they may not do so on the others.

This western coast is fully exposed to the south-west monsoon, and is by no means a desirable locality to be in at that season.

Dalrymple Bank, of the same nature, lies adjacent to Little Andaman, on the same side; but Invisible Bank, to the eastward, has depths of 17 to 50 fathoms, with a rock awash in the centre. This is of bluish-grey sandstone, so that the Bank, taking into account its irregular surface and the rapidly-increasing depths around, may be considered a submerged mountain-range, of the same formation as the oldest part of the Andamans--of which, Flat Rock, an isolated peak, rises alone above the sea. All these banks probably formed islands, or part of the Andamans, when the latter stood at a higher elevation than they do to-day.

Throughout the Archipelago the scenery is of exceeding beauty. The picturesquely undulating surface is clad everywhere, save where artificial clearings have been made, with the most luxuriant jungle, for, situated within the tropics, with a fertile soil, and a climate that for two-thirds of the year is somewhat moist, the islands are covered from hill-top to sea-beach with an unbroken mantle of dense vegetation, rendered almost impenetrable by cane-brakes and undergrowth of rattans and other creepers. All along the sh.o.r.es are either stretches of yellow sand or brilliant green mangroves, and the seas round the islands are of the clearest water imaginable.

The coast-line is everywhere deeply indented, and affords a number--most unusual for such a small group--of deep-water harbours and other anchorages, where complete shelter can be found for large ships in all weathers and seasons. The most known and the best--although Port Cornwallis is nearly as good, and has the advantage of being some twelve hours' steaming nearer to Calcutta and Rangoon--is Port Blair, where the Settlement has been placed; but on the same coast of Great Andaman are many others, the more important being, Macpherson's Straits, Shoal Bay, Port Meadows, Colebrooke Pa.s.sage, and Stewart's Sound. On the west coast are situated, Port Andaman, Kw.a.n.g-tung Harbour, Ports Campbell and Mouat, while in the Archipelago perfect anchorage is to be found either in Outram Harbour or in Charka-Juru,[85] Kw.a.n.g-tung Strait, or Tadma-Juru.

These are well distributed all along the coasts, and were the Andamans situated in some position of greater political or commercial importance, they would form an invaluable possession for this reason. As it is, the islands are exploited merely as a convict establishment--an Indian Botany Bay--and the only industry of any magnitude appertaining to them is that of timber, for which indeed the harbours are very convenient, as the forests worked are all in the neighbourhood of the seash.o.r.e, or are so placed that after the trees are felled the logs can be hauled by elephants to the many creeks, and floated down to where the vessels engaged in the business are anch.o.r.ed.

The geographical conditions, and more especially the Tertiary sandstone of which the large area of the islands consists, point to a former connection with Arakan, and, in accordance with these indications, it is found that the bulk of the flora is Burmese; but the forest trees are finer, being very lofty and straight, while not a few purely Malayan species find their northern limit in the Andamans. The flora is not related to that of Hindustan and India proper--a coincidence which can be partly explained by the insular climate and difference of soil.

The forests produce valuable woods, which can be used for many trade purposes--furniture; ship, boat and house building; railway carriages and sleepers; paving blocks, boxes, gun-carriages and stocks, pianos, etc.; and as profitable minor products, there are canes for furniture, rattans for walking-sticks, and gurjan oil. Some of the woods can be obtained in extremely large quant.i.ties; all possibly in sufficiency for any trade that may arise with the islands.

Palms abound; the banian and padouk, that resembles mahogany; marble-wood, of a black, mottled appearance; satin-wood; and the iron tree, which turns the edge of an axe, are all found in the forests, in beautiful confusion with cotton-trees, screw-pines, and arborescent euphorbias, and with large clumps of bamboos, 30 and 40 feet high; while all round the coasts, mangroves, the most satisfactory of firewoods, give shelter to lovely orchids.

A very conspicuous feature of the forests is the distribution (apart from the strictly littoral vegetation) into evergreen forest, very full of large gurjan trees (_Dipterocarpeae_), and a leaf-shedding forest, containing a large proportion of padouk (_Pterocarpus dalbergioides_), or into a mixture of these two types.

The great peculiarity of the Andaman flora is that, with the exception of the Cocos Islands, which are covered with them, and thence indeed derive their name, no coconut palms naturally propagated are found in the Archipelago.[86] This is the more strange when it is remembered that all the sh.o.r.es of the Bengal Sea are the home of this tree, and that it simply teems in all the islands of the Nicobar group to the south.

During the past thirty-five years advantage has been taken of trips in the Station steamer (or by other means) to plant coconuts at suitable localities[87] along the coasts of Great Andaman, and in recent years at Little Andaman. Great numbers of the nuts were consumed by the aborigines, or by wild pigs, but at several places fruit has been obtainable, for many years past, from trees which escaped destruction.

The climate of the Andamans is equatorial in its uniformity, and greatly resembles that of Tena.s.serim and Mergui. It is not generally healthy for Europeans, but improves from a hygienic point of view when the forest is cleared from any locality. During the first two months of the dry season strong winds blow from the north-east, causing sickness, and damage to vegetation.

As the islands are exposed to the full force of the south-west monsoon, only four months of fair weather (January to April) can be counted on.

December and January are the coolest months, with a mean temperature of 79 and a mean minimum of 75, while March and April are the warmest, with means of 82 and 83 respectively, and a mean daily maximum of 92 at Port Blair, where throughout the year the mean is 80, the highest 96, and the lowest 66; so that the absolute range is 30. The mean diurnal range is as much as 14 to 15 in February, March, and April, while between June and September it is only 8 or 9. The mean temperature throughout the year is about 80.

The south-west monsoon sets in, accompanied by heavy rain, in the early part of May, but occasionally in April, and lasts till October. March is the driest month, and January and February somewhat less so; but the rainy season lasts from the middle of May till the middle of October, and what is called the moderate season, from October to January. During the latter the average monthly rainfall is about 8 inches; in the rainy season it is 16 inches, with twenty-four rainy days per month; while, on the average, throughout the dry season rain falls on ten days to an extent of 5 inches. At Port Blair the mean humidity is 83 per cent, and the average annual rainfall is 117 inches; but elsewhere it varies from 100 to 155 inches, and there are about 180 wet days in the year.

At the change of the monsoons stormy weather is common, and the neighbourhood of the Andamans is considered to be the birthplace of many of the violent cyclones that occasionally visit the Indian and Burmese coasts of the Bay of Bengal.

The hurricanes generally both originate either between Ceylon or the N.W. portion of the bay, and take various courses, according to the season; but, although situated near the cradle of these storms, the islands are not often traversed by them. In 1864 one is recorded as having visited the locality, and on the night of November 1st, 1891, a violent cyclone pa.s.sed over Port Blair, which, after travelling north-westward across the bay, did much further damage at the mouth of the Hugli and on the Orissa coast. The maximum velocity of the wind registered at Port Blair on the latter occasion was 111 miles.

Concerning the geology of the islands, it is curious that in the valley of the Irrawadi hot springs and other evidence of volcanic action occur in the same relative position to the Arakan Hills as the two islands of Narkondam and Barren Island occupy in respect to the Andamans. There seems little doubt that these two islands--now respectively extinct and quiescent--belong to the great line of volcanic disturbance that appears in Lower Burma and extends right through Sumatra and Java and the further islands of the Malay Archipelago. Thus it would seem that the Andamans proper, possessed of no volcanoes themselves, lie just outside the line of activity, and, with the Nicobars, occupy the same position with regard to the volcanic track as do the chain of islands west of Sumatra.

Possibly the land now const.i.tuting the Andamans first appeared above the sea as an extension of Cape Negrais in the latter part of the Tertiary epoch, at which time occurred the elevation of the Arakan Yoma Hills, and later became isolated by subsidence due to neighbouring volcanic action. As Mr A. R. Wallace points out,[88] the presence of active volcanoes produces subsidence of the surrounding area by the weight of every fresh deposit of materials ejected either in the sea or on land.

"The subsidence such materials produce around them will, in time, make a sea if one does not already exist."

The Andamans are of Tertiary formation, and of similar geological structure to the Arakan Yoma Range, the line of elevation connecting them with which is represented by the Alguada Reef and Preparis Island.

(The Cocos are an integral part of the Andamans.)

Two sedimentary formations have been distinguished up to the present, the one older, the other newer than the serpentine which intrudes in various localities. They have been called the Port Blair and Archipelago series respectively.[89]

The former, occurring in the south, consists of fine grey sandstone--the characteristic rock, generally non-calcareous--and beds of conglomerate and limestone as subsidiary members; red and green jasper also occur; possibly, however, of an older series than the sandstone in which they crop out. This series seems to be of early Tertiary or later age.

The second series,--Miocene, or even newer--of which the whole of the islands of the Archipelago are formed, consist typically of soft limestone, of coral and sh.e.l.l-sand, soft calcareous sandstones, and soft white clay, with occasionally a band of conglomerate, the pebbles of which seem to have been coral.

The intrusive rocks of the Andamans--similar to those of Manipur and Burma in the north, and the Nicobars to the south, and of later date than the Port Blair series--are of serpentine, often pa.s.sing into crystalline diorite and gabbro.

The Archipelago series seem to cover a large area of the Andamans, while the Port Blair formation is restricted to the south. The greater part of Rutland Island is formed of serpentine, in which small layers of brown opal have been met with, and which throughout the group seems disseminated with minute crystals of chromium. The Cinques consist of intrusive rocks of serpentine, a.s.sociated with metamorphosed, indurated, and sedimentary series, mostly calcareous. The rocks of Little Andaman are chiefly lime and sandstone, with a good deal of actual coral rock on the east and south coasts; while occasionally occur outcrops of an igneous nature. At Entry Island and Port Meadows, beds of volcanic origin exist.

In view of the connection with them, the apparent barrenness of the Arakan Hills goes to show that little may be expected from the Andamans in the way of mineral products. Amongst others, however, discoveries have been made of lignite, and the ores of chromium, copper, iron, and _sulphur_,[90] although not in quant.i.ties that would pay commercial development.

It is indubitable, as Kurz has shown,[91] that the Andamans are now undergoing subsidence; but there is ample evidence, in the raised coral beaches that fringe the sh.o.r.es, to show that in the immediate past this has been exceeded by elevation.

The islands have been subjected to earthquakes from time to time. The first recorded took place in August 1868, the next in February 1880, and several slight shocks occurred until December 1881, when a severe earthquake visited the group, made itself felt over a large area of the Bengal Sea and surrounding countries, did much damage to masonry at Port Blair, and raised waves 3 feet high, following each other at fifteen-minute intervals for a period of twenty-one hours. Another slight shock was experienced in February 1882.

The origin of the name Andaman appears to be somewhat doubtful, and it is, of course, a word unknown to the natives. It is, however, very old, and may--as Sir Henry Yule suggests in his Commentary on Marco Polo--perhaps be traced to Ptolemy (who flourished at Alexandria soon after the commencement of the Christian era), and if so, we have by him the first-known reference to the Archipelago, for he mentions a group of islands under the name of Good Fortune, [Greek: Agdaimonos Nedos], or the like--"The Angdaman Islands"--whence have come the names Agdaman, Angdaman, Andaman. Even at this early date the inhabitants were said to be cannibals.

Doubtless the Chinese knew of the group in comparatively early times, for they have records of the neighbouring Nicobar Islands going back for more than a thousand years.

Skipping a long period, we next come to the ninth century, and there have the accounts of Arab travellers (A.D. 871), which, although of an alarming description, are tolerably correct in some details. "The people eat human flesh quite raw; their complexion is black, their hair frizzled, their countenance and eyes frightful, their feet very large, and almost a cubit in length, and they go quite naked." Those ships, the story goes on to say, which have been set back by contrary winds, and compelled to anchor for the sake of water, commonly lose some of their men on these barbarous coasts, and it is fortunate that the natives have no ships or other vessels, otherwise they would seize and devour all the pa.s.sengers.[92]

Reference to the Andamans in the thirteenth-century narrative of Marco Polo is very much of a traveller's tale. "Angamanain is a very large island. The people are without a king, and are idolaters, and no better than wild beasts. All the men of this island have heads like dogs, and teeth and eyes likewise; in fact, in face they are just like big mastiff dogs! They have a quant.i.ty of spices; but they are a most cruel generation, and eat everybody they can catch not of their own race. They live on flesh, and rice and milk, and have fruits different from ours."

Colonel Yule suggests that Angamanain is an Arabic (oblique) dual, indicating "The two Andamans," viz., "The Great and the Little."

In 1563, Master Caesar Frederike set out on his travels, and returning homeward three years later, pa.s.sed near the Nicobars on his way from Malacca to Goa. "From Nicobar to Pegu is, as it were, a row or chain of islands, an infinite number, of which many are inhabited with wild people; and they call those islands the Islands of Andemaon, and their people savage or wild, because they eat one another. Also, these islands have war one with another; for they have small barques, and with them they take one another, and eat one another: and if by evil chance any ship be lost on those islands, as many have been, there is not one man of those ships that escapeth uneaten or unslain. These people have not any acquaintance with any other people, neither have they trade with any, but live only on such fruits as the islands yield."[93]

An Italian doctor, John Francis Gemelli, who made a voyage round the world, touched at the Nicobars in 1695, and refers incidentally to the neighbouring group. "Friday the 3rd, we were in sight of the island of Nicobar. The island pays an annual tribute of a certain number of human bodies to the island of Andemaon, to be eaten by the natives of it.

These brutes rather than men, use, when they have wounded an enemy, to run greedily to suck the blood that runs. The Dutch are witness of this cruelty of theirs; for they, going with fire-ships to subdue them, and landing 800 men, though they were well entrenched to defend themselves against those wild people, yet they were most of them killed, very few having the good fortune to fly to their ships.... The chief motive of the Dutch to attempt the conquest of that island, was a report spread abroad that there was a well in that island whose water converted iron into gold, and was the true philosopher's stone.... No man in Europe or Asia can give any certain account of it, because those people have no commerce with any nation in the world." This vanished wonder was discovered by an English vessel that was driven to the islands; for a native who was carrying a sh.e.l.l of water accidentally spilt some of the contents on the anchor, and the part so wetted immediately turned into gold! The narrative goes on to say that the unhappy native, who thus by his clumsiness revealed the priceless secret, was immediately killed, not by his countrymen, as might perhaps be expected, but by the strangers! There is no mention of torture; but the well was neither discovered then nor since.[94]

The next historian is an Englishman, for Captain Alexander Hamilton, in his _Account of the East Indies_,[95] written about 1700, devotes some s.p.a.ce to these islands. "The Andamans are surrounded by many dangerous rocks and banks, and they are all inhabited with cannibals, who are so fearless that they will swim off to a boat if she approach near the sh.o.r.e, and attack her with their wooden weapons, notwithstanding the superiority of numbers in the boat and the advantages of missive and defensive arms of iron, steel, and fire."