the lake, and other modifying circumstances; and I desire here to advert especially to the fact, that although it is constructed of successive horizontal additions beneath, when it decays in spring it consists of vertical prisms, penetrating its whole thickness, and standing side by side, like the columns of a basaltic cliff; which, in their mode of formation, have, I imagine, a close analogy. Dr Slagintweit informed me, that neither the ice nor the basalt forms exact prisms, the angles never having the precise measurements of true crystals. In this condition, the ice may be strong enough to support a considerable weight; and I have travelled over it with a large party on several occasions, when the prisms on which the foot rested were depressed at every step, and a pointed stick could be driven through the whole thickness into the water beneath, with as much ease as into a bank of snow.. The ice then, in fact, presents the physical characters of a semifluid mass, as pointed out by Professor Forbes, its parts being moveable on each other, not only vertically, but, as in the case of travelling glaciers, capable of gliding past one another horizontally.

In spring, when the action of the sun-light is very powerful, an incipient thaw takes place at mid-day on the surface of the snow, which, on freezing again, acquires a glassy crust. As the season advances, but while the temperature of the air is still even at noon far below the freezing point, the crust in clear weather becomes penetrated in the direction in which it is struck by the sun's rays at mid-day by innumerable canals, and finally crumbles into a granular mass like the firn of the high Switz glaciers, that crackles under the feet as soon as the sun sinks towards the horizon. This firn is not universal; it is more common within the Arctic circle, and in situations where there seems to have been originally a certain looseness in the texture of the snow, and where its surface is so much inclined that the sun's rays do not fall on it obliquely about noon. I did not notice it in any quantity on the level surface of a lake.

2. Rapid Evaporation of Snow and Ice.

The rapid evaporation of snow and ice in the winter and VOL. LII. NO. CIV.-APRIL 1852.

Y

spring, long before the action of the sun has produced the slightest thaw or appearance of moisture, is made evident to residents in the high latitudes by many facts of daily occurrence; and I may mention that the drying of linen furnishes a familiar one. When a shirt, after being washed, is exposed in the open air to a temperature of 40° or 50° below zero, it is instantly rigidly frozen, and may be broken if violently bent. If agitated when in this condition by a strong wind, it makes a rustling noise like theatrical thunder. In an hour or two, however, or nearly as quickly as it would do if exposed to the sun in the moist climate of England, it dries and becomes limber.

Mr Rae mentioned to me another example of the same fact, which bears on the transportation of boulders, and may interest geologists. During his memorable residence on the shores of Repulse Bay, he noticed several large boulders which were partially exposed at low water. When the sea froze they became engorged in the ice, and were lifted with it from the bottom by the flood-tides. The ice gaining at each tide in thickness beneath and losing above by superficial evaporation, the boulders in process of time came to rest in pits on its surface.

3. Dryness of Arctic Air.

In consequence of the extreme dryness of the atmosphere in winter, most articles of English manufacture made of wood, horn, or ivory, brought to Rupert's Land, are shrivelled, bent, and broken. The handles of razors and knives, combs, ivory scales, and various other things kept in the warm rooms, are damaged in this way. The human body also becomes visibly electric from the dryness of the skin. One cold night I rose from my bed, and, having lighted a lantern, I was going out to observe the thermometer, with no other clothing than my flannel night-dress, when, on approaching my hand to the iron latch of the door, a distinct spark was elicited. Friction of the skin at almost all times in winter produced the electric odour.—(Journal of a Boat Voyage through Rupert's Land and the Arctic Sea. By Sir John Richardson.)

י'ן

339

Volcanoes in the Bay of Bengal, &c. By Dr BUIST of Bombay. Communicated by the Author.

"One of the most terribly active groups of Volcanoes," says Mrs Somerville, begins with the Banda group of Islands, and extends through the Sunda group of Timor, Sumbaya, Balli, Java, and Sumatra, separated only by narrow channels, and altogether forming a gently-curved line 2000 miles long ; but as the volcanic zone is continued through Barren Island and Neercondam in the Bay of Bengal (lat. 12° 15′), and northward along the entire coast of Arracan, the entire length of the volcanic range is a great deal more.' The band is not, as will be seen presently, limited to Arracan, but extends northward to Chittagong, lat. 22°, or 600 miles beyond Barren Island. The first description we possess of the volcano in question, is that of Lieut. Colebrook,† who visited it in 1787, when it was in a state of violent activity; he does not seem to have landed on it, and he quotes entire the account of it given by Captain Blair in his survey of the Andaman Islands. The cone, which springs from near the level of the sea, rising at an average of 32°, 17', to 1800 feet nearly. Mr Lyell gives the following account of it; he does not quote his authority. “Barren Island, in the Bay of Bengal, is proposed as an illustration of the same phenomena (that of ancient craters of elevation, as contrasted with modern craters of eruption,) and here it is said we have the advantage of being able to contrast the ancient crater of elevation with the cone and crater of eruption, and its centre. When seen from the ocean, this island presents on almost all sides a surface of bare rocks, which rise up with a moderate declivity towards the interior: but at one point there is a narrow cleft, by which we can penetrate into the centre, and there discover that it is occupied by a great circular basin, filled by the waters of the sea, bounded all round by steep rocks, in the midst of which rises a volcanic cone, very frequently

Later authorities make 500 feet, and this is probably its true altitude.

in eruption. The summit of this cone is 1690 French feet in height, corresponding to that of the circular border which encircles the basin, so that it can only be seen from the sea through the ravines.” *

Barren Island was visited by Dr John Adam in 1831. The water close in-shore was then hot and steamy, while steam and smoke issued from the crater, but no lava or flame. He estimates the diameter of the base at about 800 or 1000 yards, and the orifice of the crater, which occupies the entire summit of the cone, at about 30. The latest description that has been published of Barren Island is that of Captain Miller, who visited it in 1843. His account of it is the same in its general features as that of Lyell, but he estimates the altitude of the cone at no higher than 500 feet; and, considering the limited distance to which it is visible at sea, this seems to be more correct than the other estimates. sets down the slope of the cone at 45, which would give an altitude of about 1000 feet, were Dr Adam correct as to the diamater of the base. Dr Adam states that it could only be ascended by climbing; and it is probable that Captain Blair's assumption of 32° 17', as the acclivity which was determined by measurement, may be near the truth, and this would bring us back to Captain Miller's view of the altitude. The volcano, like the others along the bay, is chiefly active during the south-west monsoon.

He

Next to Barren Island is the volcanic Island of Narcandum, lat. 13° 22'. The cone is about 800 feet high; no soundings are to be had within half a mile of the shore. § Crossing over to the other side of the bay, where perfect tranquillity seems for nearly a century to have reigned, we find a period when the Coromandel Coast was as much moved by volcanic agency as that of Arracan itself.

The earliest account we possess of any actual eruption in

* Principles of Geology, 1830, vol, i., p. 390. This seems to be taken from Captain Blair's estimate of 1800 feet. It is upset, as already stated, by Captain Miller, Dr Adams, and others, who gave it at 500.

† Bengal Asiatic Transactions, 1832, vol. i.

Calcutta Journal of Natural History, 1843, vol. iii.

§ Report of Calcutta Coal Committee, 1839.

the Bay of Bengal is that contained in the first volume of the Annual Register, 1776, reprinted in the Bengal Asiatic Transactions of 1847.* It was written by an officer on board a French East Indiaman, and addressed to his friend at the Hague there seems no reason to question its perfect accuracy. In July 1757, fires were seen from Pondicherry to break out on the surface of the sea three or four leagues from shore; these blazed out with the greatest fury, throwing up pumice stone and combustible matter. This was accompanied by a noise like thunder, or the discharge of heavy ordnance. An island, a league in length and about the same in breadth, with a cone and crater in the centre, then appeared. A vast quantity of dead fish were afterwards seen floating on the surface of the water, destroyed by the eruption. The sea was some days afterwards so covered with pumice stone that vessels found it difficult to make their way through it, while they ran the risk of being burnt from the showers of hot ashes with which the air was darkened. island seems speedily to have subsided again, as we hear no further mention made of it. A shoal called the Goris Bank, was seen by H.M.S. Melville, in a line joining Pondicherry† and Chittagong, and a shoal is noted on an old chart as having been met in with by an American ship in the line betwixt Pondicherry and Cheduba; both these have since then disappeared. Mr Piddington remarks that the middle of last century was the great epoch of earthquakes all over the world.

The

In 1750, Chili was visited by an earthquake, by which the town of Conception was destroyed; the sea rolled over it, and the entire port from thenceforth became useless. The whole shore seems to have sustained an upheaval of about 24 feet, and shells similar to those found in the adjoining

*Bengal Asiatic Transactions, 1847, vol. xvi., p. 499. Reports and Asiatic Researches, vol. i., p. 175. The papers of the Bengal Asiatic Society are indexed: the vast amount of most valuable information contained in the Reports can only be found by reading them through. This is the case with nearly the whole of that relating to volcanic phenomena.

† Abridged from the Remarks of Mr Piddington on the subject of the eruption. Bengal Asiatic Transactions, ut sup.