212 E. B. Tawney—Woodwardian Laboratory Notes. 
dichroic character, changing from brownish-yellow to clove-brown 
when the vertical cleavage is parallel to the vibration direction of 
the lower Nicol. Though no basal section was observed in the slice, 
the mineral has all the characters of hornblende; the divergence of 
the elasticity from ¢ axis was about 19°. The felspar is almost 
entirely decomposed to a white opaquish substance with occasional 
zeolitic formations; clearer lines in the granulated matter indicate, 
however, its triclinic nature; it incloses olivine and fills up spaces 
between other crystals. Two crystals of brown mica were found in 
the slice; it is abundant enough, however, to be visible to the eye 
in the hand-specimen. 
For a diabase it is rich ino olivine, and by this it differs from the 
next following rocks, to which it has some resemblances in other 
ways. Notwithstanding some little doubt about the nature of the 
pyroxene, I have ventured to class these rocks with the diabases : 
others from abundance of hornblende might prefer to call them 
proterobase. 
Hornblendic diabase, in appearance somewhat similar to the 
Penarfynydd rock; as noticed above, it forms part of the ridge 
near Careg llefain: the hand-specimen is blackish-brown to black, 
with cleavage planes of augite about 4-inch long, but no serpentine 
and scarcely any felspar; iron pyrites is seen scattered about. 
Slight effervescence with acid. 
The microscope shows that hornblende is now the main constituent; 
it forms large sheets limited by the adjacent minerals or inclosing 
them; the vertical cleavage is well developed, while abundant basal 
sections give a prism angle of about 124°; it is very strongly 
dichroic, precisely as in the last rock, but is here much more 
abundant; from pale yellow it becomes clove-brown when the 
e axis is parallel to the short diameter of the polarizing Nicol. 
The hornblende and pyroxene alike pass into alteration products ; 
hornblende first into a green dichroic substance preserving the 
original cleavage; then, as the structure becomes lost, iron separation 
sets in; a serpentinous substance is formed in part, while the bulk 
is a pale-green viridite of feeble doubly refractive power. 
The relation of the hornblende to the pyroxene is not quite the same 
as that described by Prof. Bonney (Q.J.G.S. vol. xxxiii. pp. 895, 912) 
in Balk and Coverack gabbro, Cornwall, neither has it the green 
colour except where undergoing change. 
Almost equal in amount to the hornblende is a pale-yellowish 
augite, not dichroic, the extinction angle about 385°, not a close 
lamellar cleavage, but rather distant usually (nearly rectangular 
in basal sections). The augite is intimately connected with the 
hornblende, sometimes its crystals are almost entirely surrounded by 
the latter, or scales of hornblende of shadowy outline are scattered 
in it, showing penetrating intergrowths; possibly the hornblende is 
partly a paramorphic change of augite, but I presume most of it to 
be original. r 
Plagioclase exists in the form of prisms, or larger masses filling 
interspaces, generally as prisms inclosed in or penetrating the 
