Geology.] 
natural history. 
601 
nearly to the water on their west sides ; but on the east, 
and more especially the south-east, they present steep cliffs; 
and the same conformation, he adds, seemed to prevail in 
the other islands *. If this structure occurred only in one or 
two instances, it might be considered as accidental ; but a« 
it obtains in so many cases, and is m harmony with the 
direction of the ranges, it is not improbably of still more 
extensive occurrence, and would intimate a general elevation 
of the strata towards the south-east. 
Now on examining the general map, it will be seen, that 
the lines of the coast on the main land, west of the Gulf of 
Carpentaria, between Limmen’s Bight and Cape Arnhem, 
— from the bottom of Castlereagh Bay to Point Dale,- — less 
distinctly from Point Pearce, lat. 14° 23', long. 129° 18', to 
the western extremity of Cobourg Peninsula, — and from Point 
Coulomb, lat. 17° 20', long. 123° IP, to Cape Londonderry, 
have nearly the same direction ; — the first line being about 
one hundred and eighty geographical miles, the second 
more than three hundred, and the last more than four 
hundred miles, in length t. And these lines, though broken 
* Flinders, Vol. ii, p. 235. 
+ It is deserving of notice, that the coast of Timor, the nearest 
land on the north-west, at the distance of about 300 miles, is also 
nearly straight, and parallel to the Coast of New Holland in this 
quarter : part of the mountainous range, of which that island con- 
sists, being probably more than 9000 feet high ; and its length, 
from the north-eastern extremity to the S.W. of the adjoining 
island of Rottee, about 300 miles. — But, unfortunately for the 
hypothesis, a chain of islands immediately on the north of Timor, 
is continued nearly in a right line for more than 1200 miles, (from 
Sermatta Island to the south-eastern extremity of Java), in a 
direction from eant to west. This chain, however, contains several 
volcanoes, including those of Sumbawa, the eruption of which,. 
