8 Prof. L. T. More on Electrostriction, 



dielectric constant is greater, the denser the substance ; it 

 may well be that if a gas, or a solid to a less degree, were 

 compressed, and the molecules thus become closer together, 

 the dielectric constant might increase. It does not at all 

 follow that the increase is accompanied by a repulsion of 

 the molecules at right angles to the field, unless we make the 

 assumption that an electrical charge produces in some manner 

 a mechanical strain. If the dielectric constant is shown by 

 experiment to increase with a tractile force, not because the 

 molecules of matter are incidentally brought closer together, but 

 because of the stress itself, then the reciprocal relation that 

 a charge causes a strain perpendicularly to the field will hold, 

 or vice versa. But until one or the other is indisputably 

 proven, to introduce either one into the equation to prove the 

 other is tacitly assuming the whole question of the mechanical 

 relations between aether and matter. 



In a recent paper Dr. Sacerdote * discusses theoretically 

 the special case when the armatures are separated from the 

 solid dielectric by spaces filled with a non-conducting fluid. 

 He arrives at the conclusion that the modulus of elasticity 

 does not enter, the equation being 



The letters have here the same significance as before, and the 

 dielectric constant of the fluid is immaterial. 



This is, however, only apparently so, for this constant would 

 be involved in the calculation of the field-strength in the solid 

 dielectric. For if the dielectric constants are different, the 

 fall of potential between the armatures is no longer uniform. 

 The resultant field in the soKd would be equivalent to a field 

 produced by the potentials of the armatures acting through a 

 homogeneous dielectric plus a field produced by free charges 

 assumed to be on the faces of the solid dielectric. These latter 

 evidently require the treatment for a condenser with adherent 

 armatures, and the introduction of the coefficient of elasticity. 



In this paper he again asserts that my experiments gave 

 nugatory results because of lack of sensibility, and ErcoKni's j 

 show k^ to be positive. Granting this criticism to be true, 

 it should not be overlooked that I also experimented with 

 hard-rubber tubes and obtained no elongation, although the 

 effect is supposedly much greater than for glass. The results, 

 to be given later, using thinner glass, a less distance between 



* Sacerdote, Journal de Phys. (3) t. x. p. 196 

 t Ercolini, loc. cit. 



