Vol. 54.] OF BOTJLAT BAT (jeESEY). Ill 



agent, and the very large and irregular cavities which characterize 

 his division of ' globules anormaux.' 



The description and illustrations which M. Michel-Levy ^ gives 

 of the rock from Gargalong, near Frejus, show that it bears a great 

 resemblance to that from .Boulay Bay, both in the characters of the 

 pyromerides and of the ' matrix.' 



Miss Raisin has referred to rings of quartz — often incomplete, 

 and possibly due to contraction — from Careg-y-defaid.^ In the same 

 paper also (pp. 264-267) flow-brecciation is recorded as one of the 

 causes to which formation of pyromerides is due ; and I have made 

 a similar suggestion in describing some acid lavas from Pem- 

 brokeshire. 



The resemblances existing in many respects between the spheroids 

 of the Jersey rocks and those of the Yellowstone district naturally 

 suggest a common origin, or at all events that similar factors may 

 have been operative in their production. The clotting in the magma 

 which, in some instances, seems to have been the cause of the for- 

 mation of nodules in Boulay Bay, forcibly suggests itself in the 

 case of some of the Yellowstone rocks. Through the kindness of 

 Prof. Bonney I have had the advantage of examining some specimens 

 from the last-named district, in which the resemblances to Boulay 

 Bay rocks were very strong. However, in the thin sections 

 examined there was no evidence, such as that brought forward 

 above, to suggest that the spherulites represent in any way an 

 element foreign to their more immediate surroundings. 



In Prof. Iddings's report on the rocks of Obsidian Cliff,^ chemical 

 analyses are given to show that the spherulite is ' nothing more 

 than a small portion of the magma which has crystallized with a 

 particular structure.' 



Hence, and the study of the literature of the Yellowstone brings 

 out the point strongly,* the difference which exists between these 

 growths and their ' matrix ' is due to a particular crystalline 

 structure; in many of the Boulay Bay, and, so far as I know them, 

 in many of the North Wales and Wrockwardine nodules, this 

 particular crystalline structure is of secondary import. Indeed, 

 in many cases, it is not only entirely absent, but superseded by a 

 kind of fluidal structure, which renders the formation of any radial 

 growth highly improbable.^ In the case of the Wrockwardine 

 specimens the uninterrupted continuance (in many instances) of 

 fluxion-lines from matrix to pyromeride, and a certain haziness at 



1 Bull. Soc. geol. France, ser. 3, vol. iii (1875) pp. 224-235, and Fouque & 

 Michel-Levy, ' Min. Micr.' pi. xv, fig. 2. 



^ ' Nodular Felstones of the Lleyn,' Quart. Journ. Geol. Soc. vol. xlv (1889) 

 p. 256. 



3 Seventh Ann. Report U.S. Geol. Surv. (1888) p. 283. 



* Ibid., and Iddings on ' Spherulitic Crystallization,' Bull, Phil. Soc. Wash- 

 ington, vol. xi (1891) p. 445. 



° Of such a type as that figured by Prof. Bonney from near the Conway Falls, 

 Quart. Journ. Geol. Soc. vol. XKxviii (1882) pi. x, fig. 4. Some nodules from 

 the Lledr Valley shovr a similar structure. One of these is interesting, from the 

 fact that it contains a few ill-shapen pink garnets altering to chlorite. 



