188 



MR. H. H. THOMAS ON THE 



[May 191 1, 



diameter, and are distributed more irregularly than those of smaller 

 size. They vary from perfect spheres to elliptical and disc-like 

 masses flattened parallel to the flow-handing. Their surfaces are 

 often ribbed or ridged parallel to the lines of flow, which may be 

 traced across any internal structures. In cross-section they have 

 a marked concentric (as well as radiate) structure ; but there is 

 evidence in many instances of hollow spaces, either concentric or 

 more often central, which have been filled up with dark siliceous 

 matter or secondary felspar. It has been argued that these hollow 

 spaces were original, and that the structures are lithophysal. The 

 proof of their secondary origin, however, lies in the fact that spheru- 

 lites are seen which are composed of two parts — an outer felspathic 



Fig. 5. 



A^Pseudoinorph after augite in secondary quartz and chlorite, in a purplish 

 soda-rhyolite. Base of the North Cliff opposite the Garland Stone 

 (Skomer). Slide E 7763. X 100 diameters, 



iB^Pseudomorph after augite in chlorite, in the soda-rhyolite (?) of the Mew- 

 stone summit (Skomer) ; from a slide kindly lent by Mr. F. T. Howard 

 X 85 diameters. 



layer, and an inner chalcedonic central portion. The outer layer is 

 radiate, but the needles, if produced, would meet at the centre of 

 the chalcedonic mass. The chalcedonic material radiates inwards 

 from a series of points on the surface which separates the fel- 

 spathic and chalcedonic portions of the spherulite. This clearly 

 points to the removal of the central part of the original felspathic 

 spherulite and its replacement by chalcedony. Some spherules 

 contain a nucleus of cryptocrystalline silica or flint ; and, in fact, 

 Mr. Harker's description of the nodules in the Carnarvonshire 



