G72 Undaniprd Vibrations in an Unhounded Dielectric. 



sepanitely, moving in the same way, but at an infinite distance 

 from the others. 



<). In the second paragraph of PockHngton's letter to 

 'Nature'' it is pointed out that a system of sources can be ar- 

 ranged on- the surface of a sphere so that there is no external 

 oscillating field. We have, in fact, to deal with such a system 

 in any of the free oscillations of the dielectric inside a per- 

 fectly coiiducting hollow surface. But this is quite different 

 from a system in which there is an actual oscillating field 

 external to the conductors and at the same time no radiation. 



7. Considering the matter without regard to mathematical 

 analysis, 1 submit that it is difficult to imagine how a system 

 of sources can maintain an infinite medium in vibration 

 without losing energy. To take what seems a parallel illus- 

 tration from another subject, can we conceive of a sonorous 

 .body emitting waves of sound into an infinite atmosphere, 



even if it be doubly-connected, which do not carry off energy? 

 And if Macdonald^s claim be tenable, what distinction is to 

 be drawn between divergeut and convergent waves, between 

 those which involve e'"^^^"''^ and those which involve ^"'(v^+'*>, 

 between a system of sources and a system of sinks ? 



8. A few errata in ' Electric Waves ■* may be here pointed 

 out. In the discussion of the vibrations in the space between 

 two concentric spherical surfaces (p. 49) the solutions involv- 

 ing harmonics of tesseral type * have been overlooked ; it also 

 seems doubtful whether or not it is intended to include that 

 corresponding to an harmonic of order zero, this being a 

 possible solution and the one cited by Pocklington. In the 

 investigation of the modes of oscillation of a dielectric sphere 

 surrounded by a different dielectric which extends to infinity 

 (p. 59), the solution given corresponds to absorption instead 

 of radiation, and at the surface of separation the tangential 

 electric displacement instead of the tangential electric force 

 is made continuous. In the expression for the rate of increase 

 of the energy inside a surface (1. 6, p. 70] there appears to be 

 an omission from the bracket of the terms 



__ ^_ _^_ ^. 



•^ dxdt ^ dydt * dzdt ' 



this, however, does not affect any result. 



Rnyal Collesfe of Science, Dublin. 

 October 14, 1903. 



\_Note hy Mr. Macdonald. — Prof. Orr's consideration in 



sending me a proof of his paper affords me an opportunity of 



withdrawing the hasty generalization referred to, which, as 



he remarks, had beeii already challenged by Prof. Larmor and 



* J. J. Thomson, Inc. cit. 



