﻿Yol. 6 1.] BETWEEN ST. DAVID'S HEAD AND STEUMBLE HEAD. 599 



frequently a banded structure, owing to the presence of lines of opaque 

 white spots. It has then the appearance of a banded spherulitic 

 felsite. Under the microscope, however, these white patches are 

 seen to be merely circular spots of incipient decomposition, the 

 banded structure being apparently due to shearing or fluxion - 

 structure. There is no trace of any spherulitic structure. Shearing 

 has largely confused the original character of the rock, but it 

 seems to belong to the same type of rock as that described above, 

 with the exception that phenocrysts of a ferromagnesian mineral 

 are represented to a limited degree. This is a pale pyroxene, very 

 fibrous, and looking very much like an altered rhombic form. 

 Some doubtful monoclinic pyroxene is seen, and the rock may, 

 perhaps, be designated an augite- or enstatite-porphyrite. 

 (See PI. XL, fig. 3.) It is mapped by the officers of the Geological 

 Survey as a contemporaneous lava. I think, however, that it is 

 intrusive, and is a modification of the lime-bostonite series described 

 above. 



For reasons already mentioneds (p. 593), I was not able to examine 

 all the exposures in the Mathry district, but among my specimens 

 from that area there is not a single indication of the Llaorian lava- 

 series. The appearance rather suggests a series of thin parallel sills 

 of lime-bostonite or porphyrite, which is the typical form that such 

 intrusions take in some other localities. I was equally unable to 

 trace the continuity of the enstatite-diabase sills of Trevine east- 

 wards to the Pwll-Strodyr Fault, all my specimens from this area 

 belonging either to the lime-bostonite type of Abercastle or to the 

 porphyrite-type. Much remains to be done, however, in accurately 

 mapping this area, and as the igneous exposures may belong to any 

 of the above-described varieties (which, in the field, are not always 

 easy to distinguish), considerable care will be necessary, not only 

 in separating the lime-bostonites and porphyrites from the contem- 

 poraneous flows, but also in ascertaining the exact limits of the 

 enstatite-diabase-intrusions. 



It is probable, also, that other exposures of this type occur beyond 

 the Mathry district. In the Geological-Survey collection of rock- 

 slices from this part of Pembrokeshire are two specimens, one from 

 the Tremynydd range, and the other from the Carn-Llidi area, 

 which closely resemble this type ; but their exact localities are not 

 quite certain. 



VI. Review of the Petrography of the Hock-types. 



Under this head attention is drawn to certain points of petro- 

 graphical or mineralogical interest, which have come under obser- 

 vation in the examination of a large series of rocks from the 

 areas dealt with in the foregoing pages. These I will now proceed 

 to describe. 



