444 PEOr. G. A. J. COLE ON LITHOPHYSES AI^D 



with by Miss Eaisin.^ It is more difficult to decide as to the con- 

 centric coats which may be formed within the vesicle ; these can be 

 seen in an incipient stage in some examples from the Eocche Eosse, 

 and attain their finest and most onion-like development at Obsidian 

 Cliff, Yellowstone Park. The same structure is clearly traceable 

 in a mnch altered condition in the lithophyses of the Conway 

 Mountain.^ 



The lithophysal aggregates in the Lipari obsidians bear evidence 

 of decay, which is doubtless largely due to the attack of vapours 

 and liquids subsequent to the primary devitrification. Some of 

 Mr. G. W. Butler's specimens from the I'orgia Yecchia show a faint 

 red zone in the spherulites between the porcellanous outer zone and 

 the central fibrous area. Under the microscope this is seen to be 

 due to a reddening of the crystallites included in the otherwise 

 colourless second zone. Oxidation has taken place, resulting in the 

 developmeut of patches of powdery brown-red products. It is clear 

 that extended alteration of this colourless and markedly crystalline 

 zone is liable to leave a hollow shell in the lithophyse ; and this 

 accounts for the fact that the central fibrous mass appears again and 

 again on fractured surfaces of the rock as a ball, loosely set in a 

 socket formed of the outer and porcellanous zone. It will be seen, 

 from the considerations put forward in the preceding paper, that 

 this hollow shell does not mark the limits of the original vesicle, but 

 lies in reality outside it. 



In some specimens, as Mr. Butler has pointed out to me, two or 

 more shells have arisen in the lithophysal matter ; but the spaces 

 separating them are of merely local extent, the crystalline mass 

 being continuous in other parts of the lithophyse. These spaces 

 appear to result from the partial removal of certain of the concen- 

 tric layers. 



I am not, then, prepared to abandon the suggestion put forward 

 in the case of the Conway lithophyses, viz., that the interspaces 

 between the successive coats result from alteration of a formerly 

 solid mass. In the well-known spherulitic lavas of Esgair-felen 

 and other Welsh localities, as well as in those near the Wrekin, 

 where cracks full of chalcedony and quartz can be traced into the 

 spherulites, I have no doubt as to the production of ' hollow 

 spherulites ' by ordinary agents of decay. ^ I need not recapitulate 

 the arguments previously brought forward ; but we must now be 

 ready to recognize the fact that true spherulites and aggregates of 

 lithophysal origin may closely simulate one another. 



The typical Continental pyromerides are truly spherulitic ; so is 

 the great bulk of the lava of the Wrekin area, the lines of flow 

 running through the boldly-developed spherulites. But in the 



^ ' Nodular Felstones of the Lleyn,' Quart. Journ. Geol. Soc. vol. xlv. 

 (1889) p. 260, fig. 5. 



^ Gr. A. J Cole, ' Alteration of Coarsely Spherulitic Rocks,' Quart. Journ. 

 Geol. Soc. vol. xlii. (1886) p. 186, pi. ix. %. 6. 



3 3id. p. 184, &c., and ' On Hollow Spherulites,' Quart. Journ. Geol. Soc. 

 vol. xh. (1885) p. 162. 



