Page:Journal of the Asiatic Society of Bengal Vol 29.djvu/19

1860.] Cosmos, both apparently derived from the same source, must be rectified. The narrator states (in "Lyell") that the sea filled the circular valley round the cone.

Horsburgh states that in 1803 the volcano was observed to explode regularly every 10 minutes, projecting each time a column of black smoke, perpendicularly, to a great height, "and in the night a fire of considerable size continued to burn on the east side of the crater, which was then in view."

The oldest account on record is that of Captain Blair, already quoted, taken from his Report of the survey of the Andaman Islands. He must have visited the Island about 1790, as far as I am able to conclude from the publication in the researches and the date of his chart of the Andamans which is 1790. He approached nearly to the base of the cone, which he describes as the lowest part of the island, very little higher than the level of the sea, but he does not mention the black stream of lava. The acclivity of the cone he states to be 32° 17′, and its height 1,800 feet nearly, which, says he, is also the elevation of the other parts of the Island. On the other hand he remarks that the cone is visible in clear weather at a distance of twelve leagues, which would require a height of not more than from 900 to 1,000 feet. I think therefore that Captain Blair could have taken no accurate measurements, contenting himself with a rough estimate. If it could be proved otherwise, the Island would have subsided 820 feet since he visited it.

From the description in some of these accounts it would appear that the high vegetation which we found on the external slope of the Island, is of quite recent origin.

Mr. Adam's authority (1831) states as follows :—

"The summits to the N. E. were completely smooth and covered with ashes; those to the S. W., although partly covered with ashes, also have a good many small shrubs over them, with dry and parched grass growing on the surface."

He conjectures from this that the eruptions would take place only in the S. W. Monsoon or rainy season, at which time the S. W. wind will blow the dust and ashes on the hills in the opposite direction, or N. E.; such a conjecture is hardly admissible on the ground given, it being easier to account for the vegetation on the