part 1] PROCEEDINGS OE THE GEOLOGICAL SOCIETY. xiii 



The ash in which the specimen was found had suffered decom- 

 position, and this suggested some confirmation of the theory which 

 the speaker had advanced, that petrification was effected by colloidal 

 solutions. The colloid would be absorbed by the plant-fragments, 

 and the slightest decay of these would cause the colloid to 'gelate.' 

 The gel would no longer be able to escape through the cell- 

 membranes, and would ultimately be deposited on the cell-walls. 

 In this way petrification would be initiated, and the agate-structure 

 so frequently observed in each cell of a petrified plant could 

 be easily and adequately explained. In well-known localities for- 

 petrified plants, both in this country and abroad, decomposed volcanic- 

 ashes were associated with the fossils, the latter occurring in the 

 ash itself or in ashy sandstones. As localities for examples in, 

 this country, Rhynie, Pettycur, Gullane, Lennel Braes, Duns, etc. 

 might be cited. 



The structure of the coal-balls of Lancashire seems to be in con-, 

 sistent with this theory ; but there were probably other methods 

 of producing colloidal solutions, and there are some signs in the 

 coal-balls that such solutions were present, such as the fibrous 

 radiate structure of the matrix. In any case, the specimen from 

 South Georgia occurred in a decomposed ash, and gave some con- 

 firmation of the general theory that, in volcanic ashes which are 

 in process of decomposition, the conditions are suitable for the 

 petrification of vegetable fragments, because they favour the pro- 

 duction of colloidal solutions. 



Dr. J. W. Evans agreed that the rocks of South Georgia and 

 Elephant I. showed no affinities with the Andes, but neither did 

 the rocks of the centre and east of South America, and so it was. 

 still possible that they might represent part of that continent 

 which had been left behind in a westward movement. The facts, 

 disclosed by the Lecturer were consistent with either theory of the- 

 origin of the Atlantic (that which attributed it to the foundering 

 of former land, and that which supposed that the adjoining con- 

 tinents had drifted apart) or with a combination of both hypo- 

 theses, which the speaker favoured. He asked whether any of the- 

 islands contained erratics that might be attributed to ice-transport,, 

 when the climate was more severe and the sea stood at a higher level. 



Mr. G. M. Part referred to the very welcome addition which, 

 the Expedition had made to our knowledge of the Cape Verde - 

 Islands. Apart from Dr. A. Harker's description of the Beagle- 

 Collection, and certain work which was supposedly proceeding in. 

 Germany (but of which there were no results published as jet), 

 there had been no very recent descriptions of these rocks. The 

 island of St. Vincent consisted of the broken-down remains of a 

 strato- volcano with a central core of plutonic types. The Quest 

 Expedition had not collected any of these latter, owing to the 

 shortness of the time available, but had brought back a number of 

 interesting nepheline-bearing lavas (nepheline-basalts, basanites, 

 and analcite-basalts) as well as limestones associated with them, . 

 and specimens of dykes similar petrographically to the lavas. 



