Prof. J. W. Judd — On Marekanite and its Allies. 247 



around the crystals in porphyritic glassy rocks, which have recently 

 been discussed by Mr. Eutley : and Mr. Waller, 2 there are other 

 conditions of tension which affect much larger masses of natural 

 glass, and which are comparable in the degree of their manifestation 

 to those which are exhibited by the well-known Eupert's drops. 



This condition of tension appears to be manifested in two different 

 ways. In the case of the Marekanite-balls, and the obsidian from 

 India described by Damour, the molecules are in a state of such 

 unstable equilibrium that a jar, or an incision, is sufficient to bring 

 about that partial re-arrangement which leads to their breaking 

 up with more or less explosive violence. But in the case of the 

 Ponza rock, the relief of newly-formed surfaces from strain leads 

 to the separation of a thin, more or less wrinkled, film. A con- 

 nection between these two cases is perhaps established by the 

 experiment of submitting the Marekanite-balls to moderate heat, when 

 thin white films like those formed spontaneously on the surfaces 

 of the Ponza rock are seen to make their appearance. 



The study of the mode of occurrence of these different rocks shows 

 that the condition of extreme tension belongs to the nuclei of per- 

 litic masses. It has long been recognized that the perlitic structure 

 is the result of contraction in a rock-mass, and in 1879 it was shown 

 by MM. Fouque and Levy, 3 and in 1880 independently by Mr. 

 Grenville Cole, 4 that this structure could be produced in colloid bodies 

 by artificial means. 



It has long been known that some of the natural glasses contain 

 a considerable amount of water and other volatile substances, up to 

 as much, indeed, in some cases, as 10 per cent, of their whole weight. 

 It has often been asserted that the presence of a large percentage of 

 water is characteristic of pitchstones, and of a much smaller amount 

 of the obsidians. But the perfectly vitreous Marekanite-balls show 

 by their behaviour when heated that they contain a very large pro- 

 portion of volatile matter, and other similar exceptions to the supposed 

 rule might be adduced. It is probably true that volatile matters are 

 more likely to be retained within masses which have cooled down 

 while subjected to the pressure of superincumbent material ; and 

 under the same conditions, that slower cooling would take place, 

 which would lead to the more abundant development of crystallites 

 and the acquirement of the mass of the resinous lustre characteristic 

 of pitchstone. 



Now, so far as my experience goes, the glasses containing a large 

 amount of volatile materials have an abnormally low specific gravity. 

 If the water be slowly separated from the mass, it is scarcely con- 

 ceivable but that contraction should take place and the density rise. 

 The materials in which the perlitic structure has been artificially 

 produced — namely, precipitated silica and Canada-balsam — both con- 

 tain volatile matters, and it is during the gradual removal of these in 

 drying that the perlitic structure is produced. The spirally-curved 



1 Quart. Journ. Geol. Soc. vol. xli. (1884), p. 340. 



2 Geol. Mag. Dec. III. Vol. II. (1885), p. 91. 



3 Mineralogie Micrographique (1879), pi. xix. fig. 2. 



4 Geol. Mag. Dec. II. Vol. VII. p. 115. 



