Cole — The VarioUte of Annalong^ Co. Dotim. 517 



across, as if they held one another like so many freely suspended 

 magnets. The axes of these crystals are seldom curved. Among 

 their rays the almost colourless rods have developed ; and, while 

 many lie quite irregularly, others have grouped themselves into 

 fan-shaped bunches, probably sections of irregular cones, radiating 

 from local centres, and so forming the spherulitic structure trace- 

 able in the mass. Yery little matrix remains anywhere between 

 these close-set spherulites, the radial structure of which can only 

 be fully appreciated between crossed nicols. The pyroxene in this 

 central part of the rock is in the form of shorter needles, small 

 granules (like those of common basalts on a very minute scale), 

 and also of tiny patches micropegmatitically intergrown with 

 felspar. The two minerals seem here to have developed simul- 

 taneously ; while in the porphyritic products of a previous con- 

 solidation we find the pyroxene including the plagioclase, as in 

 normal doleritic rocks (PI. xxi. fig. 2). In the sections taken nearer 

 the edges of the dyke, the green rods seem to have originated before 

 the colourless ones, judging by their free and superior develop- 

 ment ; but the small diameters of the microlites of either sub- 

 stance (about '002 of a millimetre) would enable groups of the one 

 to penetrate among groups of the other, and both to pass in among 

 the magnetite skeletons, without sensible interference or displace- 

 ment. As exemplifying possible local differences in the order of 

 succession, owing to fluctuations of pressure and temperature in 

 the viscid oozing mass, I may record one solitary case where the 

 magnetite skeleton has formed around a colourless rod, presumably 

 of felspar. 



As already mentioned, translucent brown glass forms a thin 

 selvage to many of the olivine-basalts of the coast of Mourne. A 

 remarkably well-preserved edging to a fissile altered dyke, imme- 

 diately south of Annalong harbour, will serve as an example ; in 

 this the plagioclase crystals remain brilliantly fresh. A spherulitic 

 tachylyte has been described from the opposite side of the Mourne 

 Mountains, near Bryansford.^ A more interesting case in connexion 

 with the variolite is a dyke, 1*5 metres across, appearing somewhat 

 conspicuously just north of Arthur's Port, and shown even on the 

 one-inch Greological Survey Map. Both its selvages have given 



1 Quart. Journ. Geol. Soc, vol. xKv., p. 305. 



2S2 



