MICROSCOPICAL CHARACTERS OF THE WHIN SILL. 647 
shows the same striation, and we can identify the crystalline faces, 
it will furnish us with the means of determining this point. 
Looking at a fractured surface of this extremely coarse-grained 
rock, one sees long bright black cleavage-faces, measuring, say from 
2 to 3 centimetres in length, and from 1 to 3 millimetres in breadth. 
These are traversed by a fine striation, which runs at right angles 
to the length of the face, and which can be readily seen with a 
hand-lens. ‘These faces represent the orthopinacoid, according to 
the determinations of Phillips. If, now, we break the specimen so 
as to obtain an end view of the prism, we see that it is bounded by 
two small faces which have a bronzy lustre, similar to that of diallage, 
and which are inclined at approximately the same angle as the 
basal planes in a crystal of augite twinned on the normal plan. We 
see, further, that a number of similarly twinned crystals are placed 
close to each other, with corresponding faces parallel. Now these 
small bronzy-looking faces are the basal planes of the crystals ac- 
cording to the determinations of Phillips, and they are also the 
planes of easiest cleavage. Owing to the size of the crystals it is 
easy to prepare sections in definite directions for microscopical and 
optical examination. Such sections establish conclusively the cor- 
rectness of Mr. Phillips’s determinations. Those cut parallel to the 
long black lustrous faces (20 P &) give straight extinction, and show a 
marked striation at right angles to the vertical axis ; those cut at right 
angles to this axis (Pl. X XIX. fig. 1) show that the prisms are formed 
almost entirely of the ortho- and clino-pinacoids (the prismatic faces 
only appear as slight truncations), and that each individual is twinned 
in such a way that the face of composition is at right angles to the 
plane of the optic axes; that is according to the normal type. These 
sections also show that whereas the prismatic cleavages are well 
seen in the fresh crystals or portions of a crystal, they are obscure 
. wherever alteration has taken place. Other and somewhat irregular 
cleavages parallel to the two vertical pinacoids, especially the clino- 
pinacoid, may be observed. Sections cut parallel to the latter pina- 
coid (fig. 2) show an oblique striation, the strize making with the 
twinning-line an angle of about 75°, z.¢. the angle 6 (Naumann) 
of augite. In these sections the angle of extinction referred to the 
twinning-line is 42°. The facts mentioned above prove conclusively 
that the striation so frequently referred to is due to lamination 
parallel to oP and not to o Pa. The mineral therefore is not 
true diallage. That the striation observed in the pyroxene of the 
normal whin is of a similar character can be easily proved by the 
application of optical tests to the thin sections. We have now to 
deal with the chemical composition of this pyroxene. 
In a former part of this paper (p. 645) the method of separating the 
pyroxene and titaniferous iron-ore from the felspar and quartz was 
explained. The two former minerals were separated from each 
other by means of a small bar-magnet. It was found that the iron- 
ere could be entirely extracted inthis way. The pyroxene grains, 
after having been roughly sorted under the microscope, were analyzed, 
with the following result. They were not perfectly pure, for it was 
2x2 
