PYROLECTRICAL CURRENTS. 489 



of the various regions which are not at the same temperature 

 (1081). 



Cubical crystals with a tetrahedric hemihedry have the same 

 characteristics. In this case the axes of electrical polarization due 

 to unequal temperatures are parallel to the longer diagonals of the 

 cube. Their resultant is still null, and all trace of regular electri- 

 fication disappears when a homogeneous variation of temperature 

 is produced. Pyroelectricity, on the contrary, is observed when a 

 plate perpendicular to one of the axes is touched with a heated 

 test body. This is the case with blende and sodic chlorate. 



Boracite is particularly remarkable. This body is powerfully 

 pyroelectrical, and has long been supposed to crystallise in cubes, 

 which would be contradictory of the preceding remarks ; but we 

 know from the researches of M. Mallard that a crystal of boracite 

 at the ordinary temperature is made up of twelve orthorhombic 

 hemihedral pyramids, and that it really becomes cubical at a tem- 

 perature of 265, retaining this condition to the fusing point. 



Now Friedel and Curie have shown that boracite has no pyro- 

 electrical properties at temperatures higher than 265 that is to 

 say, when it is cubical. But at the moment when, in cooling, the 

 crystal resumes its crystalline complex, there is suddenly a consider- 

 able disengagement of electricity with a well-marked polarization ; 

 for the effect changes its sign according as one or the other face 

 is observed of a plate parallel to one of the faces of the tetra- 

 hedron. 



1080. PYROELECTRICAL CURRENTS. When two poles of a 

 pyroelectrical crystal are joined while its temperature is changing, an 

 extremely feeble current is produced in the wire, which may, how- 

 ever, be shown by a galvanometer of very high resistance. The 

 intensity of this current is proportional to the quantity of electricity 

 disengaged on the crystal in unit time, and therefore is sensibly 

 proportional to the velocity of cooling. 



This method of observation enables us to recognise the existence 

 of electrical properties in certain crystals which are better conductors, 

 and with which electrometers would give no appreciable effect. 



The crystal is cut in the form of a plate with parallel sides, and 

 pressed between two metal plates connected with a galvanometer. 

 Friedel has thus found* that, for regular heating or cooling, fahlerz, 

 or grey copper ore, gives a very appreciable current changing its sign 

 with the variations of temperature, and that the axes of maximum 



* C. D. FRIEDEL. Ann. de Chim. et de Phys. [4], Vol. xvn., p. 92. 1869. 



