254 PROCEEDINGS OF THE AMERICAN ACADEMY 



the Blake transmitter. It was impossible to use the same loudness of 

 voice with the lips close to the mouthpiece of the latter transmitter, as 

 the scratching sound which always appears with a microphone too 

 strono-ly excited became veiy marked. Therefore, while keeping the 

 voice at the same intensity, we removed the mouth farther from the 

 transmitter, until the scratching ceased. In fact, the difference in 

 the size and shape of the mouthpieces of the different instruments and 

 the position of the mouth in speaking, whether pressed against the 

 mouthpiece or at a little distance from it, always affect both the abso- 

 lute and relative strength of the vibrations of the diaphragm with 

 different vowels, and hence of the resulting currents. A further inter- 

 estino- illustration of this will be seen by comparing the results with 

 the Hunning and Edison transmitters. 



In another series of experiments the sound of a C^ stopped organ- 

 pipe giving 512 vibrations per second was substituted for the voice. 

 Very great constancy in its intensity was secured by blowing it with a 

 blast and using a very large bottle as a pressure regulator. Three 

 tubes were inserted through the cork of the bottle. Through one of 

 these the air from the blast entered the bottle ; through a second it 

 passed to the organ pipe ; while the third tube was drawn to a small 

 aperture at its extremity, which allowed a gi'eater or less quantity of 

 air to pass out, according as the pressure from the blast tended to in- 

 crease or diminish. Observations were made at intervals for ten 

 minutes. Only slight oscillations of the coil occurred during this time. 

 The following are the mean readings with different transmitters : — 



Hunning ' 7.9 



Fitch 3.4 



Blake 0.7 



Edison 0.2 



Magneto 0.5 



The strength of the current in amperes in the different experiments 

 was as follows. 



Vowel Sounds. 



Hunning Transmitter (^from Series 3). 



Vowel. Mean Deflection. Current. 



a 10.9 0.000737 



12.8 0.000787 



u 6.8 0.000503 



^ 2.1 0.000213 



