458 PROF. W. S. BOTJLTON ON THE [Nov. ICJOzL, 



with an evident perlitic structure (see PI. XL, fig. 1). It is made 

 up of little bodies measuring 0*05 inch across, each consisting of 

 roughly-concentric arcs of glassy material, often of a bright-green 

 colour, alternating with crescentiform spaces, which are now filled 

 with clear quartz, or light-brown dusty material. Centrally there are 

 often irregular or roughly-circular spaces, usually filled with clear 

 quartz. Sometimes the peripheral, crescentiform, glassy portions 

 are very irregular in shape, though retaining their sharp edges and 

 curvilinear outline, strongly resembling in shape the cavities 

 and intervening glassy matter of the AVrockwardine lithophyses 

 figured by Mr. Parkinson. 1 The crescentic arcs of glass above 

 mentioned resemble closely the vitreous splinters, with sharp, 

 curvilinear edges, so characteristic of the rhyolite-breccias and tuffs 

 immediately to the north-west of the South-Eastern Ehyolite (see 

 p. 475 & PI. XLIII, fig. 4). In common with these, they often 

 show the characteristic longitudinal tension-lines in the glass, as if 

 formed by its distension ; but these might be explained by con- 

 traction during the development of the perlitic structure. 



It seems possible that some of these small bodies represent 

 vesicles (lithophyses), similar in structure to the much larger 

 vesicle in artificial slag shortly to be described (fig. 2, p. 461), 

 and now rilled with secondary quartz. At the same time, it is 

 clear that the structure of the matrix is largely perlitic, much of 

 the original glass having been replaced by silica, the remaining 

 devitrified portions (green and brown in colour) showing the charac- 

 teristic outlines of the perlitic structure. 2 



Prof. Bonney, in describing the nodular felsites of North Wales, 3 

 holds that the nodular structure has been produced 



1 by simple contraction and roughly-concentric cracking of the mass in cooling, 

 being thus intermediate between the perlitic structure common in glassy acid 

 lavas and the spheroidal structure common in basalt . . .' or 'by similar con- 

 traction in cooling, which is determined by the presence of a cavity.' 



It may be that the matrix of this Pontesford rock with perlitic 

 structure, and what appears in places as a microlithophysal structure 

 as well, owes its finely-nodular character to the causes referred 

 to by Prof. Bonney. 



The nodules proper in this specimen (15) are quite small (0*1 to 

 0'3 inch) and usually imperfect and irregular. Each consists of a 

 fibrous growth, in some cases, apparently, round one or more 

 vesicles ; but in the absence of flow-lines curving round them, or 

 other direct evidence of the gaseous origin of the cavities, it is 

 possible that these cavities may have been occupied originally by 

 spherulitic growths. Into these cavities, and around them, the 

 brown fibrous material has developed, forming tufted or mushroom- 

 shaped growths. The same fibrous material has finally surrounded 



1 Quart. Journ. Geol. Soc. vol. Ivii (1901) p. 221. 



2 T. G. Bonney & J. Parkinson ' On Primary & Secondary Devitrification 

 in Glassy Igneous Bocks' Quart. Journ. Geol. Soc. vol. lix (1903) p. 440. 



3 Quart. Journ. Geol. Soc. vol. xxxviii (1882) p. 295. 



