488 MEASUREMENT OF CAPACITY. DIELECTRICS. 



quantity of electricity is proportional to the projection of this axis 

 on the perpendicular to the plate. 



Another method is still simpler. The plate being kept at the 

 ordinary temperature, and connected with the earth by its lower 

 face, a conducting test body is placed on the other face at a higher 

 temperature for instance, the base of a hemisphere ; the test body 

 is connected with the needle of the electrometer, or rather with 

 one pair of quadrants, the other pair of which is to earth while 

 the needle is kept at a constant potential. A deflection is at once 

 observed. The electricity changes its sign when the opposite face 

 is observed, or when the test body is at a lower temperature than 

 that of the plate. The results thus obtained with tourmaline are 

 in complete agreement with those which would be given by regular 

 heating.* No effect is observed with homohedral substances, or at 

 most a slight deflection in the same sense for two opposite faces. 



1079. The method of the hot or cold test body enables us to 

 demonstrate a special kind of pyroelectricity which seems to have 

 multiple axes, and which consequently would seem to contradict 

 what has been said above. With plates of quartz parallel to the 

 axis, or even simply with an uncut quartz prism, the existence of 

 three pyroelectrical axes is observed perpendicular to the axis of 

 the crystal and making angles of 120 with them. These axes 

 are in planes which pass by the edges of the hexagonal prism, and 

 directed from one edge on which are the rhombic facets, towards 

 the opposite edge which is not modified ; they are then alternately 

 in opposite directions. A plate perpendicular to the optical axis 

 gives no regular indication. 



In the same conditions tourmaline presents three axes of pyro- 

 electricity perpendicular to the axes of the crystal, and forming with 

 each other angles of 120. 



As the lateral axes of electrical polarization in the rhombohedric 

 system have from symmetry a resultant null, the crystal should not 

 manifest pyroelectrical phenomena by a homogeneous variation in 

 temperature ; and experiment then shows only traces of electricity 

 distributed in a wholly irregular manner. For the same reason, 

 if the surface of the test body is greater than that of the plate, 

 the heating is more regular and the electrification less. 



This apparent pyroelectricity appears, then, to be attributed 

 to the compression produced in the crystal by an unequal expansion 



* C. FRIEDEL and J. CURIE. Comptes rendus, Vol. xcvi., pp. 1262, 1389; 

 Vol. xcvn., p. 61. 1883. 



