314 MESSRS. G. A. J. COLE AND J. W. GREGORY 



the contrary, they are broken through, and ti\e " pseiidocry stall ite " 

 resolves itself into a minute rift filled with colourless secondarj^ 

 products. The dark granules that abound throughout the rock arc 

 frequentlj" grouped along these transparent lines, so as to suggest 

 that they also are developed as products of alteration. We feel 

 that until a similar structure is found in other rocks, so that ample 

 comparison may be made, the last word cannot be said on these 

 interesting " pseudocrystallites : " but in a large spherulite, 2 centim. 

 in diameter, which we have especially studied in reference to this 

 question, all stages between the coarser and undoubted cracks and 

 these little individualized and intersecting rifts can be determined 

 without any hesitation (PI. XIII. fig. 6). 



While dealing with these structures in the variolite, we would 

 note the great abundance of yellow epidote as a secondary consti- 

 tuent, whether in the occasional perlitic lines of separation, or in 

 the more ordinary fissures, or irregularlj^ developed in the matrix. 

 ]M. Levy's specimens appear to have been deficient in this mineral, 

 the importance of which was insisted on by Delesse in 1850, and 

 again in discussion in 1877. The epidote commonly forms a granu- 

 lar mosaic, but occasionally occurs in almost colourless prismatic 

 forms associated with the chloritic areas. It is unnecessary to call 

 attention to the little bunches and fibres, probably of actinolite, in 

 the matrix, and to the other mineral features that M. Levy has so 

 admirably deseribed. 



The more we examine the interesting structures of the variolite, 

 the more we are convinced that we are dealing with the altered and 

 devitrified selvages of an ancient basic andesite or basaltic lava. 

 The matrix of the rock, despite its epidote-veins and granules, and 

 its action on polarized light, shows so often a perlitic structure that 

 its former colloid condition seems placed almost beyond dispute*. 

 The alteration of basic glass to palagonite and to fibrous recrystal- 

 lized products, has been again and again described ; and the " vario- 

 lite of the Durance " received, indeed, a normal explanation if placed 

 in the category of the tachylytes. 



The varioles themselves have been already referred to as if they 

 Avere ordinary spherulites ; but this is a question that has received 

 considerable attention. Setting aside the old discussions as to 

 whether they were included pebbles or concretions subsequent to 

 consolidation, we note that even Morozzo f compared them to the 

 " taches variolitiques " of the lava of Yulcano, although he could see 

 no proof of volcanic action in the locality of his variolite near Susa. 

 In far later times, M. Michel Levj^J, wliile stating that the varioles 

 have only a superficial analogy to the colloid globules of acid rocks, 

 fully recognizes their spherulitic character, classing them as crystal- 

 lized varieties. Geinitz § also insists that they should be compared 



* Mr. Rutley has recently traced ia certain " epidosites " the extreme pliase 

 of alteration of perlitic lavas. Quart. Journ. Geol. See. vol. xliv. (1888), 

 p. 740. 



'■ Mem. Acad, royaledes Sciences, Turin, t. v. p. 171.' 



I Bull. Soc. geol. France, 3« ser. t. v. (1877), pp. 257 and 263. 



§ Tscherm. Min. \i. petr. Mitth. Bd. i. 1878, p. 143. 



