69 
leaves of aquatic plants. The remains of some of these trees were 
still standing on their roots, while the trunks of the great part lay 
scattered on the ground, in every possible direction. The barks of 
trees and roots appeared, generally, as fresh as when they were 
growing; in that of the branches particularly, of which a great 
quantity was found, even the thin silver membranes of the outer 
skin were discernible. The timber of all kinds, on the contrary, 
was decomposed, and soft in the greatest part of the trees ; in some, 
however, it was firm, especially in the roots. The people of the 
country have often found among them very sound pieces of timber, 
fit to be employed for several economical purposes. 
The sorts of wood, which are still distinguishable, are birch, fir, 
and oak. Other woods evidently exist in these islets, of some of 
which the leaves were found in the soil. In general, the trunks, 
branches, and roots of the decayed trees were considerably flat- 
tened ; similar to the surturhrand or fossil wood of Iceland, as well 
as to that found in the lake of Thun, in Switzerland. 
The soil to which the trees are affixed, and in which they grew, 
is a soft greasy clay ; but for many inches above its surface, the soil 
is entirely composed of rotten leaves, scarcely distinguishable to 
the eye ; many of which may be separated by putting the soil in 
water, and dexterously and patiently using a blunt spatula or knife. 
By this method the Doctor, from whose account this extract is made, 
obtained some perfect leaves of ileoe aquafolium, which are now in 
the herbarium of the Right Honourable Sir Joseph Banks ; and some 
other leaves, which, though less perfect, seem to belong to some 
species of willow. In this stratum of rotten leaves, could also be 
distinguished some roots of arundo phragmites. 
By examining a well, dug at Sutton, by Joshua Searby, these 
gentlemen discovered, that a moor of the same nature is found under 
ground in that part of the country, at the depth of sixteen feet ; 
consequently very nearly on the same level with that which consti- 
