s(h\iE coi\rnMPOH.ii<y .idi'.ixces ix riiysics—ir 



475 



CltMr smooth siiu--liko ciirNfs with the [Hriodicity of the wave- 

 train are obtaintHl by usiii^ a rereiver of wliirh the natural freciuent^j^ 

 agrees with the fundamental freciiieney (or its octave) of the oscillator. 

 Such curves are s«x>n in h'ii;. 7; the two fiin<laineiiial fre(|uencies were 



Set of curves, N 4.2 to X 27. 



Fig. " — Curves Obtained with a Receiver in Tune with the Oscillator (Topmost 

 Curve with Receiver Tuned One Octave Below the Oscillator). (Physical Review) 



will strike the receiver at a time-inter%'al T; there will be nothing of the nature of 

 interference. But if T happens to be an even-Integer multiple of the half-period 

 of the receiver, the second pulse will reinforce the oscillations started by the first; 

 if it is an odd-integer multiple of the half-period, the second pulse will annul the 

 vibration started by the first. Thus as the Boltzmann mirror is niove<l along, the 

 receiver-reading will pass through maxima and minima with a spacing imixjsed 

 by the characteristics of the receiver. In actual experiment this might hap(K-n if 

 the frequency and the damping of the wave-train were much higher than those of 

 the natural vibration of the receiver. On the other hand it iloes not appear that a 

 frequency much higher than that of the wave-train could be simulateil by any 

 effect due to the receiver — an imponant point, in view of what follows. 



