Refraction of Quartz produced hy Electrical Force. 107 
direction. If, then, positive electricity be communicated to 
one end of an axis of no piezoelectricity and negative elec- 
tricity to the other, this electrification must produce a com- 
pression in the one direction and simultaneously an expansion 
in the other. If the electricities communicated are exchanged, 
then there will be also an exchange of compression for dilata- 
tion. But what has just been said is nothing else than was as- 
sumed above in explanation of the electro-optical experiments. 
The parallelepiped of quartz designated II. in the first paper 
was now examined in parallel as well as in convergent sodium- 
light. At first it was placed in the flask filled with benzol 
just as before, except that now the rays of light traversed it 
parallel to the principal axis. If, then, the analyzer was ad- 
justed so as to give a dark field, then upon electrification the 
place between the perforations became very bright; the distri- 
bution of the bright parts changed upon reversal. Both these 
results were to be expected, since the perforations were made 
in the direction of an axis of maximum piezoelectricity. 
In order to investigate the nature of the double refraction 
produced, the method previously used of an interposed glass 
plate which was compressed in a vertical or horizontal direc- 
tion could not be employed, since, in consequence of the rota- 
tion of the plane of polarization in the direction of the prin- 
cipal axis, the conditions were more complicated than before; 
and therefore I made the experiments in convergent light. 
For this purpose glass tubes bent upwards at right angles 
were cemented to the end surfaces of the parallelepiped, and 
these, together with the communicating perforations, were 
filled with mercury. The change in the system of rings lying 
in the middle between the perforations produced by electrifi- 
cation consisted again in the transformation of circles into 
ellipses having major axes longer and minor axes shorter 
than the diameters of the corresponding circles. The major 
axes lay in the direction of a line joining the perforations, and 
therefore parallel to the direction of the corresponding axis of 
maximum piezoelectricity, if the marked end of the crystal 
was positively, and the unmarked end negatively electrified. 
This axis, on the other hand, was at right angles to that 
direction, if the marked end were negatively electrified and 
the unmarked end positively. If we refer to the piezoelectric 
behaviour of the crystal used, described in the first paper, we 
easily find, by the aid of the piezoelectric experiments given 
under 3 and 4, that the optical phenomena observed may be 
completely explained by the simultaneous action of an elec- 
trical contraction in one direction, and an expansion in the 
direction at right angles to it. 
