Tol. 57.] OF THE YELLOWSTONE AND GEEAT BRITAIN. 225 



or stellate in outline, I conclude that at these spots were gas- 

 vesicles : such gases having been disengaged from the surrounding 

 rock at the time when solidification began. In the case of the 

 crescentic or annular amygdaloids I conclude that at these spots 

 the magma was in an extremely hydrous state, and that the area 

 now enclosed by the periphery of the nodule was practically composed 

 of two parts, that is, magma and water. Hence the excess of either 

 would crystallize out or separate out, as the case may be. As in the 

 Yellowstone, the water charged with sundry substances in solution 

 probably played some part in depositing encrusting minerals on the 

 surfaces of the cavities. 



Id conclusion, I wish to express my grateful thanks to Prof. Bonney 

 for much kind help in the microscopical work, and also for suggestions 

 concerning the arrangement of the various parts of the paper. 



EXPLAjS^ATION of plate VIII. 



Fig. 1. liithopliysal structure in a nodule from Boulaj Bay. X 4. 



2. Litbophjsa from near Wrockwardine. The left-hand side is slightly 

 dusty. On the right-hand side, the glass in which the lithophysa lies 

 is visible. X 4. 



Discussion. 



Prof. Bonnet congratulated the Author on the good use that he 

 had made of his visit to the Yellowstone region, and expressed 

 concurrence with the results at which he had arrived. The paper 

 would put an end to the idea (in which the speaker had never 

 believed) that hollow spherulites were formed by decomposition at 

 the centre, and it had shown them to have been formed about an 

 original cavity. Discontinuity of any kind, a.s the speaker had 

 pointed out in 1885, was peculiarly favourable to the formation of 

 spherulites. In the case of glass-bottles, softened by heat, they 

 developed abundantly from the inner and outer surfaces ; and 

 similarly in sheets of glass, which had become adherent through 

 heat. In the case of the formation of a spherulite round a cavity, 

 crystallization might be facilitated by the pressure of the imprisoned 

 vapour being outwards on the enclosing hardening mass ; but in 

 other cases, before that had time to produce an effect, the vapour, 

 contracting more rapidly than the rock, might cause strains in it : 

 sometimes, as the Author had suggested, magma and vapour might 

 be mixed in the interior, producing either alternating zones of rock 

 and excluded vapour, or loose spongy crystallization. As the 

 audience had not properly seen the slices, owing to the partial 

 failure of the lantern, he might say that he had examined the 

 supposed tridymite in the British specimens, as well as the other 

 structures described by the Author, and thought the former presence 

 of that mineral highly probable, for that it should be replaced by a 

 quartz-paramorph was not unlikely. He thought that the Author 

 had proved that what the Yellowstone obsidians now were, this the 

 felstones of Boulay Bay, Pontesford, and Wrockwardine had once been. 



The Author briefly replied. 



