430 MR. J. PARKINSON ON DEVITRIFICATION [Nov. 1903, 



The second product of spherulitic crystallization was the larger 

 variety, blue in a hand-specimen, but brown by transmitted light 

 in a thin section. Mention is made of the fact that the fibres 

 which form these spherulites ' are in sectors and do not radiate from 

 a single point,' and from the resemblance ' in structure and optical 

 behaviour ' to the ' fibrous granophyre-groups,' the inference is 

 drawn that the composition is essentially the same. 



Finally, a description is given of the characteristic ' porous 

 spherulites.' An account recently published l renders additional 

 notice unnecessary at this point. 



The foregoing order of crystallization afTords a simple basis for 

 classification. 



In the first place the minute colourless spheres, often, or 

 indeed usually surrounded by a crack which forms a boundary, 

 appear to have been produced as a result of strains set up in the 

 cooling rock, as Mr. Rutley has described. 2 Frequently a microlith 

 forms a central nucleus. When these spherulites lie within the 

 fibrous brown type no boundary-crack is visible, nor does the latter, 

 in any case, necessarily surround the spherulite completely. 



The fact is noteworthy that the formation of the fibrous brown 

 spherulites has not disturbed the orientation of the black microliths, 

 whereas in the small colourless spheres they lie tangentially. 



It is of interest to find that perlitic cracks may appear in the 

 neighbourhood of the small colourless spherulites, unconnected 

 with any sign of radial growth; hence we may infer that the 

 spherulites were antecedent rather than subsequent 

 to the formation of the cracks, and may indeed have 

 caused them as above suggested. Occasionally, an additional 

 crack appears concentric with, but entirely external to, the spherulite. 

 Here then the latter appears to be the earlier structure. On the 

 other hand, if this suggestion of strain is true, it is not easy to 

 understand why such isolated areas should occur in the midst of 

 the fibrous brown spherulites. 



The finely-fibrous structure which Prof. Iddings records suggests 

 mineral differentiation, rather than mere strain in a homogeneous 

 substance. 



Under a second heading may be placed the compact and 

 fibrous brown spherulites which occur, not merely at Obsidian 

 Cliff, but in the devitrified rocks of Pontesford, Wrockwardine, 

 Boulay Bay, and the Prescelley Hills. It is of interest to note that 

 in these examples — pre-Cambrian and Palaeozoic — the matrix in 

 which the spherulites lie affords clear evidence, by the abundance 

 of its perlitic cracks, that it solidified as a glass. This type of 

 spherulite would appear to form a criterion of secondary devitrifi- 

 cation in the adjoining matrix. 



The 'porous spherulites' constitute a third subdivision. 

 Little remains to add to what has been already written regarding 



1 Quart Journ. Geol. Soc. vol lvii (1901) p. 211. 



2 Ibid. vol. xxxvii (1881) p. 396. 





