258 PROCEEDINGS OF THE AMERICAN ACAUEMY 



the best results with Mr. Preece. For a transmitter we used a Hun- 

 uino- microphone, a form of transmitter in which granulated carbon is 

 employed, the powder being loosely confined between two metallic 

 plates. The battery used consisted of from one to twelve Grenet cells, 

 the maximum current used being sufficient to raise the temperature of 

 a platinum wire, .01 in. in diameter and about 10 inches long, to redness. 

 In order to avoid excessive heating of the transmitter it was placed in 

 derived circuit with the battery and receiver. In some experiments 

 the diaphragm was made of firm cardboard and in others of sheet-iron. 



One would naturally suppose that the longer the wire, the greater 

 would be the variations in tension, and hence the greater the loudness 

 of the resulting sound. If, however, the length of the wire is increased 

 beyond a certain limit, the quality of the articulation appears to become 

 very inferior, even though the loudness may be somewhat increased ; 

 and the wire tends to assume a free vibration of its own, producing a 

 confusing after-tone. 



The heaviest wire that we used was 0.0095 in. in diameter, as want 

 of the requisite current at the time of the experiments prevented our 

 increasing the diameter beyond this. With it a length of about six 

 inches gave in general the maximum loudness, as Mr. Preece had pre- 

 viously observed. 



Using a platinum wire .004. in. in diameter and from six to seven 

 inches long, with sufficient current to raise its temperature to dull 

 redness, very remarkable results were obtained. Not only could the 

 most unfamiliar conversation be understood, but words spoken into 

 the transmitter were heard and understood with the ear at a distance 

 of from two to ten inches from the diaphragm of the receiver, the 

 character and loudness of the sound being very similar to that ob- 

 tained when a Hunning transmitter is used with an ordinary magneto 

 receiver. Using a somewhat shorter wire at the same temperature 

 gave a decidedly better quality of sound, but the loudness was very 

 materially diminished. Lowering the temperature caused a rapid 

 diminution in the loudness of the sound. 



Almost as good results were obtained with German-silver wire, a 

 material not mentioned by Mr. Preece ; but with this the influence of 

 temperatui'e was far less marked than in the case of platinum. In fact, 

 quite as good results both as to loudness and distinctness of articula- 

 tion were obtained when the wire was simply warm to the touch as 

 when it was hotter. In this respect German silver differs from any 

 of the other materials that we used. With iron and copper but few 

 experiments were tried, and the results were quite unsatisflxctory. 



