368 Hyndman and Bonney — Analysis of Spherulites. 



which are especially valuahle, as he compares results from acid 

 with those from intermediate and basic rocks, and finds a distinct 

 difference between them. The relations of these results will be 

 briefly considered at the end of the present paper. 



I. The Natural Spherulittc Eock. 



The specimen examined was obtained by Prof. Bonney at Boulay 

 Ba} r , Jersey, the noted locality for "pyromerides " ; it did not, 

 however, come from a rock in situ, but was a loose, somewhat 

 rounded block on the shore, measuring about eight inches across, 

 and the spherulites are smaller than in the ordinary pyromeride. 

 Still, he had no doubt that it had once formed part of the rhyolitic 

 mass, for this exhibits many variations. The matrix is of a pale 

 yellowish-green colour, something like sea-water, but looking rather 

 opaque. The spherulites are reddish, sometimes pale red, becoming 

 whitish towards the outside, sometimes with an outer zone of darkish 

 purple red. They generally exhibit some indication (occasionally 

 distinct) of radial structure. 



Under the microscope the matrix is cracked and sometimes dis- 

 tinctly perlitic. The green tinge is evidently caused by a mineral 

 which occurs in minute plumes, irregularly scattered, giving fairly 

 high colom-s with crossed nicols, and belonging probably to the 

 chlorite group. Microliths and granules, mostly ferrite, are not rare. 

 With crossed nicols the matrix is seen to be devitrified, a mosaic 

 of somewhat irregularly-shaped minerals, varying in size in different 

 parts and affording rather low tints ; quartz is certainly present, as 

 well as felspar, but as the one mineral is often rather dirty, and the 

 other is slightly decomposed and untwinned, it is not easy in all 

 cases to distinguish them. 



The spherulites, though the boundary is generally sharply 

 defined, are less regular in structure than one would have anticipated ; 

 they are not unfrequently cracked and traversed by veins of 

 chalcedonic quartz. Portions sometimes had been displaced before 

 the matrix became solid. The outer part often exhibits a kind of 

 lobate growth, which seemingly is developed from the surface 

 inwards, comparable with that of a nullipore like lithothamnion, as 

 seen in a slice of limestone. A radial structure is perceptible, 

 though not very distinct, as. in axiolites. Sometimes, however, this 

 structure is exhibited in the usual way, throughout the outer zone, 

 and is then replaced (for a space) by one less regular. The interior 

 often presents a more distinctly radial structure, but it is minute, and 

 crowded granules of clear quartz sometimes occur. 



When the rock was sampled for analysis, it was found possible in 

 many cases to break the spherulites out complete ; only those Avhich 

 were perfect in this way were used. To obtain a good sample of the 

 matrix was much more difficult. At first sight the rock left after 

 breaking away the spherulites appeared fairly homogeneous, but on 

 closer inspection it was difficult to find even a small portion which 

 did not contain part of an incipient spherulite. The end was 

 attained by carefully breaking away small pieces from the rock just 



