March, 1918] A Lecher System — Experimental 153 



the inner edges of the plates would touch. We then reduced 

 all the summarizing sets of Figure 3, together with all the other 

 sets not shown, to correspond to the standard for the fundamental 

 for 7/ = 18.2, viz., 142 (see Figure 3). In this way the curves 

 of Figure 6 were obtained. They certainly cannot be much in 

 error. 



The curves for the ninth and eleventh harmonics represent an 

 average set of values since not every peak was investigated for 

 those tones for every change of oscillator. The other curves of 

 Figure 6 give the fundamental and the thirds, fifths and sevenths 

 on both sides of the fundamental. By taking the average of the 

 two thirds, the two fifths and the two sevenths, the damping is 

 eliminated and the results can be compared directly with the 

 fundamental. 



V. The Question of Damping. 



Before making the comparison it is well to ask two questions. 

 How is the intensity of a peak affected by the fact that it is 

 split in two? Is the damping sufficiently small that by taking 

 the arithmetical mean of any two corresponding tones equally 

 distant from the fundamental it can be eliminated? We 

 proceed to the answers to these questions. 



For the results shown in this paper the coupling, x , was kept 

 constant at 11 cm. But we have shown before that for x = S 

 cm. and less and for small values of y the fundamental is 

 split up. Evidently, then, an increase of y acts like a decrease 

 of .r; it causes two periods for a given tone. In our arrange- 

 ment we detect these two possible periods (or wave lengths) by 

 means of a sliding bridge. Manifestly, when one period is 

 present, the other is absent, so far as the receiver is concerned. 

 In other words, all of the energy is in each of the two peaks of 

 the fundamental. 



The second question, it would seem, can be answered in this 

 way. The fact that the peak on the oscillator side of the 

 fundamental is higher than the corresponding peak on the 

 receiver side of the fundamental makes it plausible to say that 

 if 7o is the intensity at the oscillator spark-gap say and / the 

 intensity at some point along the system the equation 

 /=7„e~" would hold where a is the logarithmic decrement. 



