382 ME. J. V. ELSDEN ON THE AGE OF [Aug. I9O4, 



mentioned (p. 379), in certain of these crushed rocks secondary albite 

 has completely enclosed fragments of unaltered augite, causing a 

 total reversal of the original structure. 



Pyroxenes. 



It seems clear that there were two distinct generations of 

 pyroxene, causing an apparent deviation from the ordinary type of 

 the Bala diabases on the one hand, and from the post-Carboniferous 

 dykes on the other. It is true that Mr. Harker did find two 

 generations of pyroxene, in a rare instance near Llanrwst ; but in 

 the Llanberis dykes this occurrence is more frequent, although the 

 evidence is usually indirect, owing to the ease with which the 

 earlier form has yielded to processes of alteration, where it survived 

 the corrosive action of the magma. In a few instances, however, 

 comparatively-unaltered fragments of the earlier pyroxene are pre- 

 served as corroded remnants, included in the ophitic plates of the 

 second generation. More often these remnants are represented only 

 by rounded chloritic and serpentinous inclusions in the ophitic 

 augite. There is no sign of crystalline continuity of the two genera- 

 tions, and the circumstances seem to point to a complete change of 

 phase, the first-formed pyroxene being reabsorbed to a large extent 

 before the crystallization of the later variety, pointing to very slow 

 cooling, during which the conditions of equilibrium in the magma 

 underwent considerable change. The precise variety of the earlier 

 form is uncertain, but the fact that the included fragments are not 

 in crystalline orientation with the later variety would suggest the 

 possibility that the earlier forms were rhombic. In other similar 

 cases, such as the sahlite-diabase of Sweden, 1 the diabase of 

 Connecticut,- and in the Whin Sill, 3 the earlier pyroxene is of a 

 paler colour and more easily altered than the later form. In any 

 case, the rounded serpentinous and chloritic inclusions in the ophitic 

 augites of Llanberis are more probably to be referred to an earlier 

 pyroxene than to olivine, as has been suggested by some observers. 1 



Coming now to the ophitic augites, there is evidence that during 

 their crystallization the magmatic conditions were not stable. Their 

 pale colour when fresh and the comparatively-low cy extinction- 

 angle indicate a variety near malacolite. They very commonly 

 possess the peculiarity (noticed also b} r Mr. Harker in the Lleyn 

 diabases) that the crystals, although apparently homogeneous, are 

 seen between crossed nicols to be polysomatic. The separate areas 

 are crystographically continuous, but possess different extinction- 

 angles. This structure has been explained as a modification of the 

 hour-glass structure, for which the explanation of L. van Wervekc 5 



1 E. O. Hovey, Tschermak's Min. u. Petr. Mitth. n. s. vol. xiii (1893) p. 218. 



2 J. S. Diller, Bull. U.S. Geol. Surv. No. 150 (1898) p. 268. 



a J. J. H. Teall, Quart. Journ. Geol. Soc. vol. xl (1884) p. 653. 



4 A. Harker ' Bala Volcanic Series of Caernarvonshire ' 1889, p. 94: see also 

 J. M. Clements ' The Crystal-Falls Iron-bearing District of Michigan ' Monogr. 

 U.S. Geol. Surv. xxxvi (1899) p. 201. 



5 ' Beitrag zur Kenntuiss der Liinburgite ' Neues Jabrb. f. Min. 1879, p. 481. 



