on a Fossil Fine-forest in Australia. 61 



The rock immediately round the roots is not of so harsh a texture as 

 it is in other parts ; in it, in the neighbourhood of the roots which 

 are in the water, there appear numerous white spots, which give the 

 stone a mottled appearance : this arises from a multitude of small 

 cavities which contain powdery silex, similar to what is often found 

 in the cavities of chalk-flints. On the shore, the surface of the rock 

 near the stems is worn into a number of little holes, which are owing 

 to the decay and removal of this powder. Mr. Clarke sees no other 

 explanation of these specks, than that they mark the situation of the 

 fibres which proceeded from the roots. The roots of the trees are 

 in some instances surrounded by an accumulation of sandy rock, 

 which forms a mound of a higher level than the rest of the stratum. 

 The roots do not descend, so far as has been ascertained, very far 

 into the substance of the rock, nor is there any appearance of a 

 dirt-bed. The stools stand from two to three feet above the surface 

 of the ground, and vary from two to four feet in diameter ; but one in 

 the lake is at least four feet above the level of the water, and five or six 

 feet in diameter. In several of the stumps from 60 to 120 concen- 

 tric rings of growth may be counted : a few of the stools are hollow 

 in the centre, but others are solid throughout : the wood appears to 

 be coniferous. Veins of chalcedony traverse the substance of the 

 trunks between the concentric rings, and also in the direction of the 

 radial lines. 



Many of the stems at Kurrur-kurran have the bark adhering firmly 

 to the trunk, and the bark in one instance was of the thickness of 

 three inches. Its appearance in one or two cases was such as to 

 show that it had been partly torn from the tree while yet standing, 

 as if it had been broken down and the bark had been rent by the fall. 

 The colour of the substance of the stems within varies from a 

 grayish white to a clouded gray, but their surfaces, when exposed to 

 the air, have become yellowish by weathering ; many are overgrown 

 by lichens, and have then exactly the appearance of the stumps of 

 recent trees. The upper extremities of the fossil stumps present 

 clean horizontal sections, which shows that they were not broken off 

 while recent, since no mode of fracturing recent pinewood could have 

 occasioned such neat, plain and parallel sections as the summits of 

 these stumps exhibit. 



In a fragment of the sandstone from the base of one of the fossil 

 stumps, the silicified impression of part of the leaf of a Glossopteris 

 was found. 



Immediately below the flinty stratum in which the trees are found 

 is a bed of lignite ; above the level at which the trees occur, 

 there are found, imbedded in the sandstones and conglomerates, 

 immense quantities of broken fragments of trees, apparently stripped 

 of their boughs and branches. These fragments are generally di- 

 vested of their bark, and appear to have been drifted. 



Fossil trees are found in this formation at other places, and nearly 

 at the same level above the sea as at Kurrur-kurran ; they occur in 

 sandstone similar to that of Kurrur-kurran, at the southern extre- 

 mity of the Tirabeenba mountain, immediately above and below a 



