On the Faraday-Maxwell Mechanical Stress. 641 



itself a sufficient cause for the strong electrical fields of 

 thunderstorms. 



2. If, however, we assume the occurrence of supersaturation 

 in the atmosphere, condensation on the negative ions is likely 

 to be of importance in connexion with the production of 

 precipitation, and as tending to cause a preponderance of 

 negatively charged rain to be carried down to the earth. 



3. The question of the occurrence of supersaturation in the 

 atmosphere must still, I think, be regarded as an open one. 



LVII. On the Faraday-Maxwell Mechanical Stress ; and on 

 JEtherial Stress and Momentum in general. By 0. Y. 

 Burton, D.Sc* 



1. A LTHOUGH the more recent development of elec- 

 J\. trical theory has shown the difficulty of accepting 

 as anything more than an analogy the Faraday-Maxwell 

 " tension along the lines of force and equal pressure across 

 them/' this specification of setherial stress must always be 

 regarded as one of the most profound and important con- 

 tributions ever made to the progress of physical science. 

 Maxwell's analysis from which this mechanical ?etherial 

 stress was deduced f, and which gave a precise form to 

 Faraday's theory, was probably as cogent in directing atten- 

 tion to the possibilities of an " intervening medium " as was 

 his great and enduring system of electromagnetic equations. 

 As the merely analogical significance of Maxwell's " mecha- 

 nical stress " does not seem to be quite generally appreciated 

 — that specification of stress being still spoken of by physicists 

 of distinction as if it necessarily represented the actual stress 

 in the medium — the brief exposition which follows is perhaps 

 not wholly uncalled for. 



2. For an electrostatic field in free aether, the Faraday- 

 Maxwell mechanical stress at any point is a tension l\ 2 /87r 

 aloncr the lines of force, with a pressure of like magnitude 

 in all directions perpendicular to the lines of force ; R being, 

 in electrostatic measure, the electromotive intensity at the 

 point in question. To the student of twenty odd years ago 

 this stress was certainly a stumbling-block ; for in a medium 

 whose electromagnetic properties are expressible by a system 

 of linear equations, and in which electromagnetic distur- 

 bances are propagated with a velocity independent of ampli- 

 tude and of wave-length, it is not easy to realize why any 



* Communicated by the Author, 

 t '' Treatise/ vol. i. chap. v. 



Phil Mag. S. 6. Vol. 17. No. 100. April 1909. 2 X 



