East of Gelli Hill, Radnorshire. 501 



llwyn, notably forming a conspicuous elevation west of the Bog 

 Wood. The specific gravity of the fresh variety is 2-86. 



Examined microscopically, the rock is seen to be a typical 

 olivine basalt with remarkably fresh felspar and augite. The most 

 abundant constituent is the lath-shaped felspar, the crystals of 

 which reach a maximum of 1*4 mm. ; there is no indication of two 

 generations. In some cases there is an aggregation into groups, 

 producing a structure with some resemblance to giomero-porphyritic. 

 The refractive index is well above that of Canada balsam, and 

 some individuals are partially replaced by a green substance, 

 probably clinozoisite. Between crossed nicols the felspar exhibitf 

 the usual first order colours and lamellar twinning, and the angle ot 

 extinction is rather high. Some zoning is to be seen. The refractive 

 index and extinctions indicate a composition approximating to that 

 of labradorite. 



A very pale, almost colourless, granular augite occurs between the 

 felspars. It has the usual relief; a few fragments exhibit the 

 rectangular cleavages, and it is commonly quite fresh. The polariza- 

 tion colours are of the first and second orders. 



There are occasional pseudomorphs, maximum dimension 1-2 mm., 

 preserved in some cases in a pleochroic serpentine, elsewhere in 

 a mixture of serpentine and calcite or quartz. Their shape and 

 structure leave no doubt as to their derivation ; they were originally 

 olivine. 



Ilmenite occurs moulded on the felspar; it is sometimes fresh, but 

 the numerous grains of sphene present in the rock have probably 

 resulted from its alteration. 



A devitrified glass occurs as a groundmass. Some vesicles are 

 filled with similar material, which exhibits a system of black crosses 

 in polarized light. One vesicle is occupied by an almost isotropic 

 glass with a curious cellular structure, the cells in places near the 

 margin elongating to tubes with approximately radial disposition. 

 This structure appears to be a record of the infilling of the vesicle, 

 the various streams of viscous magma having failed to amalgamate 

 after entry. 



The observed occurrences of the basalt are all at or near a junction 

 where fossiliferous felspathic ashes rest on shales with tuning-fork 

 graptolites ; hence the obvious conclusion that this rock represents 

 the first product of the extrusive igneous activity responsible for the 

 thick overlying ashes. But in spite of the fact that no metamorphic 

 phenomena have yet been observed, its intrusive character is still 

 considered to be a possibility. An Ordovician flow of this character 

 would be absolutely unique in the igneous history of Wales,^ and 

 the rock is remarkably fresh. On this question of possible intrusion 

 the suggestion made by Prof essor Watts ' that the area of Tertiary 

 igneous activity outlined by Dr. Harker ^ may have to be extended 

 to the south has a possible application to this locality. 



1 W. G. Fearnsides, Geology in the Field, 1910, p. 801. 



^ Proc. Geol. Assoc, vol. xix, p. 179, 1905. 



' Tertiary Igtieous Bocks of Skye (Mem. Geol. Surv.), 1904, p. 3. 



