436 swan. 



difficulties was found by introducing into the 110 volt circuit several 

 60 watt Mazda lamps, L, connected in multiple. The lamps being 

 cold when the driving current of the fork is broken by the relay, they 

 permit a momentary large damping current to flow, but their rapid 

 increase of resistance as they heat causes exactly the amount and 

 nature of reduction of current-strength which is demanded. The 

 practical operation of this simple but effective device leaves nothing 

 to be desired, and is evidently applicable to many laboratory problems. 

 It was found necessary to use a current from a storage battery as the 

 small periodic fluctuations in the current from a generator were suffi- 

 cient to produce an audible vibration of the fork. To prevent arcing 

 at the relay on breaking the 110 volt circuit, the gap was immersed in 

 oil. 



Observations of the time of decay of the sound were made with four 

 pitches approximately one octave apart, this being the maximum range 

 over which the tuning fork method was found to be practicable. Each 

 series of observations was repeated on several successive nights and 

 checked by two different observers. In all of the experiments an 

 assistant was required to note the amplitude of the sounding fork and 

 to keep it constant. This required incessant watching and adjust- 

 ment by means of the rheostat on account of small fluctuations in the 

 driving current. With every pitch several different amplitudes of 

 vibration were used for determining the rate of decay in the room, 

 and the mean time of decay corresponding to these amplitudes was 

 obtained from fifty to one hundred chronometer observations in every 

 case. The results are shown in Table 2, where 



a = radius of diaphragm 

 a = amplitude of vibration 

 n = pitch 



/ = time of decay of residual sound 

 A — rate of decay 

 E' = energy in ergs per sq. cm. per second. 



