418 



SCIENCE 



[N. S. Vol. XXXVIII. No. 978 



the then new Miehelson experiment with the 

 theory of astronomical aberration and with 

 other known facts, that he made his bril- 

 liant surmise : ' ' Perhaps the stone slab was 

 affected by the motion." I rejoined that 

 it was a 45° shear that was needed. To 

 which he replied, "Well, that's all right — 

 a simple distortion." And very soon he 

 said, "And I believe it occurs, and that 

 the Miehelson experiment demonstrates 

 it. " A shortening long- ways, or a length- 

 ening cross-ways would do what was 

 wanted. 



And is such a hypothesis gratuitous? 

 Not at all: in the light of the electrical 

 theory of matter such an effect ought to 

 occur. The amount required by the ex- 

 periment, and given by the theory, is 

 equivalent to a shrinkage of the earth's 

 diameter by rather less than three inches, 

 in the line of its orbital motion through the 

 ether of space. An oblate spheroid with 

 the proper excentricity has all the simple 

 geometrical properties of a stationary 

 sphere; the excentricity depends in a defi- 

 nite way on speed, and becomes consider- 

 able as the velocity of light is approached. 



All this Professors Lorentz and Larmor 

 very soon after, and quite independently, 

 perceived; though this is only one of the 

 minor achievements in the electrical theory 

 of matter which we owe to our distin- 

 guished visitor. Professor H. A. Lorentz. 



The key of the position, to my mind, is 

 the nature of cohesion. I regard cohesion 

 as residual chemical affinity, a balance of 

 electrical attraction over repulsion between 

 groups of alternately charged molecules. 

 Lateral electrical attraction is diminished 

 by motion; so is lateral electric repulsion. 

 In cohesion both are active, and they nearly 

 balance. At anything but molecular dis- 

 tance they quite balance, but at molecular 

 distance attraction predominates. It is 

 the diminution of the predominant partner 



that will be felt. Hence while longitudinal 

 cohesion, or cohesion in the direction of 

 motion, remains unchanged, lateral cohe- 

 sion is less; so there will be distortion, 

 and a unit cube xyz moving along x with 

 velocity w becomes a parallelopiped with 

 sides 1/k^, k, k; where 1/k^^l — u^/v^? 



The electrical theory of matter is a posi- 

 tive achievement, and has positive results. 

 By its aid we make experiments which 

 throw light upon the relation between mat- 

 ter and the ether of space. The principle 

 of relativity, which seeks to replace it, is 

 a principle of negation, a negative proposi- 

 tion, a statement that observation of cer- 

 tain facts can never be made, a denial of 

 any relation between matter and ether, a 

 virtual denial that the ether exists. 

 Whereas if we admit the real changes that 

 go on by reason of rapid motion, a whole 

 field is open for discovery; it is even pos- 

 sible to investigate the changes in shape of 

 an electron — appallingly minute though it 

 is — as it approaches the speed of light ; and 

 properties belonging to the ether of space, 

 evasive though it be, can not lag far behind. 



Speaking as a physicist I must claim the 

 ether as peculiarly our own domain. The 

 study of molecules we share with the chem- 

 ist, and matter in its various forms is in- 

 vestigated by all men of science, but a 

 study of the ether of space belongs to phys- 

 ics only. I am not alone in feeling the 

 fascination of this portentous entity. Its 

 curiously elusive and intangible character, 

 combined with its universal and unifying 



* Different modes of estimating the change give 

 slightly different results; some involve a compres- 

 sion as well as a distortion — in fact the strain 

 associated veith the name of Thomas Young; the 

 details are rather complicated and this is not the 

 place to discuss them. A pure shear, of magnitude 

 specified in the text, is simplest, it is in accord 

 with all the experimental facts — including some 

 careful measurements by Bucherer — and I rather 

 expect it to survive. 



