Vol. 50.] PHYSIOGRAPHY OF THE LIBYAN DESERT. 545 



Whatever held up the river to the high level at Semna must have 

 been in the Second Cataract for the most part, as half the amount of 

 rise recorded at Semna would have flooded the temple site at Wadi 

 Haifa. 



One point I may suggest as worthy of consideration is that the 

 folds of Farafra and Baharia, and the Shebb and Murat wells, 

 intersect at a point up to which the old caravan-route S.W. from the 

 Dakhla Oasis directly leads. It appears possible, therefore, that in 

 the sandhills of this part there exists, or has existed, another oasis 

 of which we have at present no certain knowledge. 



VII. The Origin of the Silicified Wood. 



There is no doubt that the siliceous cementation of the sandstone, 

 and the molecular replacement of the woody structure of the fossil 

 trees by silica, are results of one and the same action, which has been 

 ascribed to geysers by Schweinfurth and by Sir J. W. Dawson, 1 while 

 Zittel has distinctly stated that he does not consider this a prac- 

 ticable theory, and points out the absence of siliceous sinter. 



"Whatever theory proposes to account for the sandstone and trees 

 near Cairo must of course also account for those west of the 

 pyramids of Giza, over the wide tract where I have shown the Jebel 

 Ahmar Sandstone to occur ; and, looking at the similarity of the 

 fossil wood from the Nubian Sandstone and of the silica-cemented 

 sandstone from various localities in Nubia, it is hard to avoid the 

 idea that similar agencies worked in each case to produce results 

 so identical. Over the whole of this area, at three points only, viz. 

 Jebel Burka near Wadi Haifa, Mandisha in the Baharia Oasis, and 

 Abu Zabel near Belbeis, have eruptive rocks been recorded, and 

 nowhere have I met with, nor does Zittel in his account of the 

 Rohlfs expedition mention, any siliceous deposit analogous to the 

 sinter deposited by the waters of geyser-springs. Considering the 

 large amount of decaying vegetable-matter there must have been in 

 the sands of an estuary into which such numbers of trees were 

 drifted, I would suggest the action of water holding natron (sodium 

 carbonate) in solution as a possible explanation. 



This would act upon the felspar-grains in the sands derived from 

 the crystalline rocks, and would form sodium silicate, while the 

 potash of the felspar would take up the carbon dioxide in place 

 of the silica. On this solution of sodium silicate coming into contact 

 with the decaying vegetation in the sands, the vegetable acids pro- 

 duced in the course of decomposition, and probably carbon dioxide 

 also, would replace the silica which would be deposited as a cement, 

 or as replacing molecule by molecule the woody structure of the 

 trees. And I think we may take this explanation as equally 

 applicable to the Jebel Ahmar or to the Nubian Sandstone, for in 

 the area occupied by the first we have the Natron Lakes with 

 their springs and natron-deposits, while round Bir Malha, south 



1 Geol. Mag. 1884, p. 386. 



