116 Mr. G. V. Mac Lean on the 



E. Sarasin and de la Rive* concluded from their experi- 

 ments that the wave-lengths determined are independent of 

 the dimensions of the oscillator and vary with the size of the 

 resonator employed. They showed that the rate of propagation 

 of electric waves through air without wires is sensibly the 

 same as that along wires. Thus by using a resonator 26 cm. 

 in diameter the length of the internode along wires was 1*12 m. 

 and in air between 1*12 and 1'25 m., while a resonator 36 cm. 

 in diameter gave the lengths to be 1'47 m. and between 1 # 4 

 and 1*8 m. respectively. They claimed that the same oscil- 

 lator gives rise at the same time to waves of different lengths; 

 that the waves sent forth are not simple but are complexes of 

 an infinite number of different waves, in fact a continuous 

 electrical spectrum is formed, and that the resonator used 

 acts as an analyser, picking out from the spectrum those 

 waves whose period is peculiar to itself, and these alone it 

 resonates. 



Poincare f and Bjerknes J, independently of each other, 

 contended that the oscillator and resonator each set up their 

 own vibrations, which are not necessarily related, and that 

 the variations of the wave-length, when different resonators 

 are used, with the same oscillator is not due alone to multiple 

 resonance, as stated by Sarasin and de la Rive, but to the 

 unequal rate of damping of the waves by the resonator and 

 oscillator. If the rate of damping in the resonator is small 

 in comparison with that of the oscillator, then Sarasin and de 

 la Rive's theory holds ; if, however, the rate of damping of 

 the resonator is great in comparison with that of the oscil- 

 lator, Sarasin and de la Rive's theory fails. The length of 

 the internode does not alter when the same oscillator is used 

 with resonators of different sizes, but it varies with the 

 dimensions of the oscillator when the same resonator is em- 

 ployed; the wave-length being the same as is found when a 

 method is employed in which no resonator is needed. " If 

 the rates of damping of the oscillator and resonator are almost 

 the same, the lengths of the waves they produce exercise the 

 same influence upon the lengths of the internode measured." 

 Sarasin and de la Rive had also pointed out the neeess-iiy of 

 having the oscillator and resonator in tune with each other, 

 and more particularly was this needful when the waves in air 

 without wires were measured. 



* JBibliotheque Uniierselle; Archives des Sc. pht/s. et nat. 3e periode, 

 to in. xxiii. no. 2, 1890, p. 113, a so p. 557. 



t Poincarg, Llektricitat mid Optik, 1891 ; Archives des Sc. phys.et nat. 

 t. xxv. p. 609 (18yl). 



% Bjerknes, Wiedemann's Annalen, Band xliv. p. 75 (1891). 



