1896.] 



Generation of Longitudinal Waves in Ether. 



273 



complications, consider four metal spheres, A, B, c, d, with their 

 centres all in one straight line ; — their relative magnitudes and posi- 

 tions being such as shown in the accompanying diagram. Let each 

 of the four be initially electrified, A and c positively, B and d nega- 

 tively. Let the charges on c and d be so strong that a spark is only 

 just prevented from passing between them by the influence of B and 

 A. Let A be gradually brought nearer to B till a spark passes 

 between them. Will the consequent spark between c and d take 

 place at the same instant or a little later ? It is not easy to see how 

 this question could be answered experimentally ; but remembering 

 the wonderful ingenuity shown by Hertz in finding how to answer 

 questions related to it, we need not, perhaps, despair to see it also 

 answered by experiment. 



" The elastic solid theory restricted to the supposition of incom- 

 pressibility (which is expressed by Maxwell's formulas) makes the 

 difference of times between the two sparks infinitely small. The 



©© 



unrestricted elastic solid theory gives for the difference of times the 

 amount calculated according to the velocity of the condensational- 

 rarefaction al wave. 



" But I feel that it is an abuse of words to speak of the ' elastic 

 solid theory of electricity and magnetism ' when no one hitherto has 

 shown how to find in an elastic solid anything analogous to the 

 attraction between rubbed sealing-wax and a little fragment of paper ; 

 or between a loadstone or steel magnet and a piece of iron ; or between 

 two wires conveying electric currents. Elastic solid, however, we 

 must have, or a definite mechanical analogue of it, for the undulatory 

 theory of light and of magnetic waves and of electric waves. And 

 consideration of the definite knowledge we have of the properties of a 

 real elastic solid/which we have learned from observation and experi- 

 ment, aided by mathematics, is exceedingly valuable in suggesting 

 and guiding ideas towards a general theory which shall include light 

 (Old and New), old and new knowledge of electricity, and the whole 

 of electro-magnetism. 



" Kelvix." 



