332 Lord Kelvin on the 



We shall consider only a perfect, and therefore an unstriated, 

 hexagonal prism. The sides of the prism may be, and 

 generally are, unequal in nature but the angles are all exactly 

 120°. For simplicity of reference to the natural crystalline 

 form, I shall suppose the prism to be equilateral, which 

 it may be in nature, as well as equiangular, which it must 

 be. Thus we have three planes of symmetry, which for 

 brevity I shall call the diagonal planes, being planes through 

 the opposite edges of the prism. We have also three other 

 planes of symmetry, which for brevity I shall call the normal 

 : planes, being planes perpendicular to the pairs of parallel 

 faces. 



§ 2. In the brothers J. and P. Curie's beautiful instrument 

 for showing their discovery of the piezo-electric property of 

 quartz, a thin plate of the crystal about half a millimetre 

 thick, I believe, is taken from a position with its sides parallel 

 to any of the three normal planes of symmetry ; its length 

 perpendicular to the faces of the prisms, and its breadth 

 parallel to the edges. The sides of this plate are, through 

 nearly all their length, silvered by the chemical process to 

 render them conductive *, and are metallically connected with 

 two pairs of quadrants of my quadrant electrometer. I find 

 that the effect is also well shown by my portable electrometer ; 

 the two sides of the quartz plate being connected respectively 

 to the outer case, and the insulated electrode, of the electro- 

 meter. In an instrument which has been made for me under 

 Mr. Curie's direction, the silvered part of the plate is 

 7 centimetres long and 1'8 broad. A weight of 1 kilogramme 

 hung upon the plate placed with its length vertical causes one 

 side to become positively electrified and the other negatively. 



§ 3. A plate parallel to any one of the three normal planes 

 of symmetry will give the same result, of transverse electro - 

 polarization ; but a plate cut parallel to any one of the three 

 diagonal planes of symmetry will give no result in the mode 

 of experimenting described in § 2. But with its sides 

 unsilvered it would, if properly tested, show positive electri- 

 fication at one end and negative at the other when stretched 

 longitudinally, as we see by the hypothesis and theoretical 

 considerations which I now proceed to explain ; and by § 11 

 below without any hypothesis. 



§ 4. Electric eolotropy of the molecule, and nothing but 

 electric eolotropy of the molecule, can produce the observed 

 phenomena. The simplest kind of electric eolotropy which 



* For a description and drawing of this part of their instrument, given 

 by the brothers Curie, see Appendix to the present paper. 



