406 Prof. V. Ball— Volcanoes in Bay of Bengal. 
the later ones indicate that the rate itself has diminished as the cooling 
has progressed, in general accordance with the well-known law. 
It is proved by the historical records quoted in my paper, and also 
in Mr. Mallet’s, that the volcano was in a state of violent eruption 
in 1789 (Blair) and 1803 (Horsburgh), and that columns of smoke, 
or rather as it should be, steam, were observed rising from it in 1787 
(Colebrooke) and 1791 (Horsburgh).' I have now to add a hitherto 
unnoticed record of its appearance on the 29th January, 1804, as 
seen by an officer of His Majesty’s ship ‘Caroline,’ who writes: 
“‘On the same evening we got sight of Barren or Volcano Island, 
which at the time was burning very fiercely, the eruptions taking 
place every eight or ten minutes, with a hollow rumbling noise. 
“This is a small circular island lying almost in sight of the Hast 
Andaman, between that and the Malay coast; it appears to be a 
perfect cinder, or at least covered in every part with lava; it is of 
considerable height, and the volcanic opening or crater is in the 
centre of the island. 
«We passed within a mile of it, and as the winds were trifling 
we observed the eruptions for three days and nights successively.” * 
As we have no records of the condition of the volcano between 
1804 and 1832, we cannot fix an exact or even approximate date for 
the more recent lava flows, nor can we fix the total duration of the 
violent stage, which, probably, with quiescent intermissions, lasted 
for at least 15 years, namely, from 1789 to 1804. 
Blair does not refer to the lava flows, from which it has been 
concluded that they did not exist when he visited the island, but 
this negative evidence is perhaps hardly sufficient for a definite con- 
clusion. Indeed, the break in the side through which one of the flows 
reaches the sea has very much the appearance of having been formed 
by the flow itself, as in the case of Campo Bianco in the island of 
Lipari (Volcanoes, by Professor Judd, p. 125). If such was the 
case, as the breach existed in Blair’s time, the lava flow must also 
have been in existence then. Mr. Mallet has some remarks upon 
the order of sequence of the three flows, which probably exhaust 
the possible conjectures in regard to them. ‘The earliest mention of 
them is by a visitor who landed in 1882, and they are referred to by 
Captain Miller in 1848. 
Since my paper was published a sketch of the Island by an officer 
of the Kwang-tung steamer, with a brief descriptive account, was 
published in the ‘‘ Graphic ” for April 16, 1881; but as the drawing 
gives an incorrect representation of the form of the cone, and the 
1 There is a very circumstantial account of a violent volcanic eruption in the Bay 
of Bengal, by which an island a league in length is said to have been formed in the 
sea three leagues from Pondicherry, in the year 1757. As in the case of Graham’s 
Island, in the Mediterranean, the Island subsequently disappeared, having been 
eroded by the sea. The original is to be found in the Annual Register, vol. 1, 1708. 
It is quoted in the Journal of the Asiatic Society of Bengal, 1847, vol. xvi. p. 499. 
2 Account of a Voyage to India and China, etc., in His Majesty’s Ship Caroline, 
performed in the years 1803-4-5, interspersed with Descriptive Sketches and 
Cue Remarks by an Officer of the Caroline. See Phillips’ Voyages and Travels, 
vol. v. 
