404 Appendix. 
down to the level of the sea on the west-south-west side for the space 
of 10 or 12 yards. This edge he supposed, from these appearances, 
to be formed of cinders and mud: a supposition which is contradict- 
ed by the consideration that an embankment of such materials, and © 
of such slight thickness, could never retain the mass so violently agi- 
tated in the interior; and is farther opposed to the statements of 
Captain Senhouse and Mr. Osborne. 
The isolation of the volcanic action is also demonstrated by the 
fact, that the temperature of the sea, within ten or twelve yards of 
the crater, was only 1° higher than the average ; and to the leeward, 
in the direction of the current, it was not at all affected, though a 
mirage played on the island. (Captain Swinburne ; Report to Ad- 
miral Sir H. Hotham.) 
There was at a subsequent period of the eruption, on the south- 
west side of the island, adjoining the principal crater, a terrific ebul- 
lition and agitation of the sea, apparently seated in another canal of 
communication; attended by the emission of a dense white steam, 
and a temperature increased to 190° Fahr. (Letter of Mr. Osborne); 
and the information has, I believe, since reached the Admiralty, that 
this crater is now elevated above the level of the waters. De Buch 
has already pointed out, that the internal action which manifests itself 
at the surface of the soil or sea by a crater of elevation, may consti- 
tute at the same time a permanent volcanic crater beneath ; the erup- 
tions of which may take place sometimes by the centre of the crater 
of elevation, sometimes by neighboring points. 
Though it is difficult, from the meagre details hitherto obtained, to 
form any correct opinion of the mineralogical character of the up 
raised mass, yet there is nothing in those details to warrant the sup- 
position that there is no stability or permanence in the composition of 
the island. The pulverised remains of coal deprived of its bitumeD, 
the hard scorie, dense and sonorous (phonolites?), the amorphous rocks 
with metallic lustre and ironstone clay, would appear to associate the 
eruption with the rocks of the carboniferous series ; a opinion whieh 
receives additional probability from the ejection of unchanged pines 
of limestone : as we see, between Pettyeur and Bruntisland in Fife- 
shire, beds of limestone and of non-bituminous coal elevated by rocks 
of plutonic origin, and argillaceous and argillo-calcareous rocks chang- 
ed into leucostines and spilites. If this is the case, 4 farther ar 
more accurate investigation of the mineralogy © Graham Island ¥ 
be of as much utility to the study of changes produced by volcani 
action, as the forms and characters of the upraised formations have 
been in pointing out their geological age and associations. 
a 
