218 
THE FELSPARS. 
of symmetry parallel to the three pairs of faces. But supposing that 
in preparing our block one pair of sides had, while still at right angles 
to the second pair, been inclined at some other angle to the third pair, 
a little consideration will show that placing the block on the glass 
again we shall have the reflection as a continuation of the object in 
only two of the six possible positions—that, therefore, there is only one 
plane of symmetry, and with the block worked to a proper angle this 
would be a model of our Felspar crystal. 
If now we cut slices thin enough to see through, parallel to the two 
cleavage planes, we shall find some differences between them as to 
their relation to polarised light. In both cases we find double refrac¬ 
tion—that is, if the polarising and analysing prisms are so placed that 
the field of the microscope is dark (if the prisms are crossed, as it is 
called), the film of crystal will, in both cases, enable light to pass 
through the second prism. But, now, keeping the prisms crossed, 
rotate the specimens on the stage- Four positions will be found in 
which they no longer do this, but become dark like the rest of the field 
of view. If now these four “ extinction ” positions are accurately com¬ 
pared with the positions of the Nicols prisms, it will be found that the 
edge formed by the two cleavages is parallel to the principal planes, as 
they are called, of these (that is, to the shorter diagonal of the face of 
the prism) in one case, and inclined to them at an angle in the other. 
The first of these will be found to correspond with the most perfect 
cleavage, the other with that in the plane of symmetry. Some of the 
larger crystals, when examined on the best (or basal) cleavage, will be 
found to be divided into two parts, shown by the fact that the cleavage 
on the one side makes a considerable angle with that on the other; and 
where detached specimens can be observed it will be seen that the 
appearance is that of two thin individuals, one of them turned round 
half way with respect to the other, and partly penetrating each other. 
This is called twinning, and this particular form is the most usual in 
Orthoclase, and is termed the Carlsbad twin, from a locality where 
good examples are found. It will be seen that the rotation is not 
round an axis perpendicular to the faces in contact, but round one 
lying in it. The laws governing the twinning of crystals show that 
that plane, being the plane of symmetry, could not be the twinning 
plane, as it is called, in contradistinction to the plane of composition. 
It will be seen at once that merely turning these two rough models 
round on that plane produces no difference in shape, and neither can 
there be any in physical properties, seeing that these also are the 
same on both sides of the plane in question. 
In thin slices this twinning shows very strikingly by the different 
appearance of the two halves in polarised light, produced by the 
different distribution of the optical properties in them. If the section 
is accurately perpendicular to the plane of composition, although there 
will be differences of colour in most positions, still both halves will 
become dark at the same time. This, however, will happen but 
rarely, but it is important, in view of what we shall find afterwards. 
