Cote—The Rhyolites of the County of Antrim ; with a Note on Bauxite. 99 
are often sharp and the sides steep. In the final grinding, a whole hemisphere or 
smaller portion of a sphere has become pulled away bodily from the perlitic 
globules, leaving the section thinner, and consequently lighter in colour, beneath 
the concavity. Of course, cases might be found in which the accident has 
occurred during the grinding of the lower surface, in which event the edges of the 
concavity must be found by focussing down on to the slide instead of away from it. 
The ordinary blue-black porphyritic obsidian of Sandy Braes is less strikingly 
perlitic than the two foregoimg examples; but the curving cracks can often 
be brought out, as Mr. Watts has mentioned, by the use of polarised light, aniso- 
tropic decomposition-products having separated along them. The purity of the 
glassy groundmass is due to the formation of minute curving crystallites, only a 
stage higher than ‘‘ margarites.” These have arisen even in the intrusions of 
glass which penetrate the porphyritic crystals. Besides the usual quartz-grains, 
and monoclinic and triclinic felspars, a green pyroxene is an important constituent. 
It occurs on a bolder scale than in the pitchstone of Carnearny, but its crystals 
are more injured. It is impossible to measure its angles of extinction in the 
sections studied, but its pleochroism corresponds to that of a soda-pyroxene 
approaching zgirine. One crystal has a thin outer zone of a slightly darker 
colour; the two portions of this zone that lie parallel to the long axis of the 
erystal become extinguished simultaneously, independently of the behaviour of 
the central mass; while the terminal portion, seen only at one end, becomes 
extinguished independently of any of the other areas. The central and principal 
area does not become dark in any position between crossed nicols, and the section, 
as shown by the figure in convergent light, is cut nearly perpendicularly to one of 
the optic axes. This in itself, owing to the phenomenon of conical refraction, 
would account for the absence of extinction; but the outer zone of another 
specimen in the same slide behaves similarly. Observations with sodium light, 
with red light having a wave-length somewhat longer than B in the solar spectrum, 
and with blue light obtained by using a solution of ammonio-cupric sulphate, 
show that there is a marked dispersion of the bisectrices ; the positions of extine- 
tion in the blue light, in the section examined, are 6° removed from those in the 
red light, and those in sodium-light are intermediate. Since the section is almost 
perpendicular to an optic axis, it can only be slightly oblique to the optic axial 
plane; consequently, if we are dealing with inclined dispersion only, this disper- 
sion would prove to be of a very marked character in a section parallel to the 
clinopinacoid. 
Inclined dispersion is, indeed, strong in egirine and in the soda-augites 
generally. In the artificial soda-augites, moreover, made by Backstrim,* complete 
extinction could not be obtained. 
* “Sur la reproduction artificielle de l’aegyrine,” Bull. Soc. francaise de Min., tome xvi. (1893), p. 182. 
