Hates of a Rocling Watch. 003 



illustration of Lord Raylergh's theory of anomalous 

 dispersion. In a system of two degrees of freedom the 

 reaction effect of. the forced vibration on the exciting vibra- 

 tion, is to shorten the period (increased rigidity) of the 

 latter, if its proper period rs the- shorter, and to lengthen it 

 if its proper period is longer than the proper period of the 

 forced vibration. The reaction virtually increases or decreases 

 the elastic forces of the medium respectively, in the two 

 contrasted cases. 



Now let the balance-wheel represent a vibrating aether 

 particle, and the watch (pendulum) the vibrating absorption 

 molecule, the proper period of the latter being greater than 

 that of the former in the above experiments. If the balance- 

 wheel starts the pendulum from rest, the compound period is 

 shorter than either component period. Forces of restitution 

 are increased for both. In other words the violet end, or 

 the spectrum beginning at the absorption-band, has moved to> 

 the red end beyond the absorption-band, becoming at the 

 same time more violet (periods generally decreased). Again, 

 if the watch represents the aether particle and the wheel the 

 vibrating absorption molecule, the watch with a wide swing 

 starting the balance-wheel from rest, the compound period 

 is greater than either component period. Elastic forces are 

 both virtually diminished. In other words the red end, or 

 the spectrum terminating in the absorption-band, has moved 

 toward the violet end beyond the absorption-band and 

 become more deeply red (periods generally increased). Had 

 the watch been less imperfect it would be interesting to 

 express this in detail, numerically. 



The question occurs, why are not these relations completely 

 reciprocal? To return to the first analogy, in which the 

 balance-wheel (smaller proper period) represents the aether 

 particle, and the watch the absorption molecule ; if the latter 

 in violent vibration were to excite the aether particle, the long 

 period should appear. In other words, the spectrum on both 

 sides of the absorption-band will move toward the red in 

 colour, and toward the violet in regard to refraction. Both 

 causes conspire to increase the index. Conversely, if bv 

 withdrawing ballast from the watch as above stated, its 

 proper period were reduced below that of the balance-wheel, 

 then, with the same terms of analogy, a compound period 

 shorter than either would appear. This implies greater 

 frequency and less refraction, or marked decrease of the 

 index of refraction of the medium. 



(7) The question may finally be asked with what accuracy 

 an apparatus like the above rocking watch is adapted to 



2.U2 



