Br Brewster m the Structure of Apopfiyllite^ 6 
ray, and have therefore one axis and a circular system of rings 
r 9 und tlie axis of the prism. The rectangles liV, V T, on the 
other hand, will be found to depolarise a bluish white of the 
first order, when the lines RS or TV are parallel or perpendi- 
cular to the plane of primitive polarisation, but all their tints 
vanish when the lines MN, OP come into that plane, in the 
same manner as in plates of mica, and all other crystals with 
two axes. 
If we incline the apophyllite in the direction MN or OP, 
when RS or TN are in the plane of primitive polarisation, the 
tint will gradually (Jiminish ; and at an angle of about 17'^ 4^ 
with the axis of the prism, it will entirely vanish, as the refrac- 
ted ray now passes through one of the poles of no~polarisation. 
By increasing the inclination still farther, the tints recommence, 
but with an opposite character, and the whole square MONP is 
covered with an uniform positive tint, the boundaries of the four 
large rectangles and the small central one being no longer seen. 
The cause of this singular result will be better understood from 
the following deductions, which the preceding experiments au- 
thorise. 
1. The apophy Hites from Fassa and TJto, have one axis of 
double refraction and polarisation, which is positive like that 
of quartz. 
2. The central rectangle ah cd of the tessellated apophyllite 
from Faroe, and the minute stripes of it formed by the lines 
m o np a r, cs,ht,d v, have likewise one axis of double refrac- 
tion. 
3. The four rectangles RT, RV, ST, SV, have two axes of 
double refraction, the principal one of which is positive, and the 
inclination of the resultant axes is about 34^. 
4. The plane of the resultant axes of the rectangles ST, RV, 
passes through their diagonals MN ; but the same plane in the 
rectangles RV, ST passes through OP. Henfe it follows, 
5. That the four rectangles, even if they were not separated 
by the portions with one axis, would not form an uniform plate, 
and that RV is the same as RT turned round 90® ; SV the 
same as RT turned round 180® ; and ST the same as RT 
turned round 270®. 
6. This singular construction has no resemblance to that of 
macled or hemitrope crystals, for the four rectangles are por- 
