260 PROCEEDINGS OF THE AMERICAN ACADEMY 



From these and other experiments, it appears that our best results 

 were obtained with the wire of .006 in. and .004 in. in diameter, there 

 being but little difference between them. The largest wire used (.0095 

 in. diameter) gave very poor results ; and with the battery power used 

 by us the finest wire employed (.002 in. diameter) was equally ineffi- 

 cient. Mr. Preece's best results were secured with platinum wire .001 

 in. in diameter. Also with the finer wire the most effective length was 

 less than with the larger sizes. The most interesting point, however, 

 developed in the course of our experiments is tlie effect of increasing 

 the temperature, notably with platinum wire, in increasing the loudness 

 of articulation of the thermal receiver. This will naturally be attrib- 

 uted to the greater expansibility of the wire at high temperatures. It 

 is possible that with German silver the effect, good at moderate tem- 

 peratures, is not bettered on heating the wire, because of a marked 

 influence of temperature in diminishing its elasticity. 



The question also suggested itself as to what effect, if any, would be 

 produced if the microphone were placed in the primary circuit of an 

 induction coil, and the wire receiver in the secondary circuit, so that 

 the receiver would only be acted upon by the induced currents from the 

 coil. Since the heating effect varies as the square of the current, with 

 such an arrangement difference of sign in the current could not be 

 perceived at the receiver, and therefore one would not expect to be 

 able to distinguish articulate speech. We thought that, if any sound 

 was produced, it would be interesting to study its quality as compared 

 with the quality of the sound actuating the transmitter. Our results, 

 however, were negative. A sound was produced on breaking the cir- 

 cuit, but no audible effect was observed on speaking or shouting into 

 the transmitter. In these experiments German-silver wire was used, 

 in the hope that even at ordinary temperatures it might prove sensi- 

 tive. Independent heating by a current of hot air was tried, but 

 without success. 



An attempt was also made to use the thermal telephone as a trans- 

 mitter. Sir William Thomson has shown, in his researches on the 

 electro-dynamic qualities of metals (Math, and Physical Papers, Vol. 

 II. p. 208), that the electrical resistance of wires is varied by their 

 tension, and indeed it was in following out this idea that Professor 

 Hughes was led to his invention of the microphone (Telegraphic Jour- 

 nal, May li, 1878). 



Slight as the variation in resistance must necessarily be with such a 

 transmitter, we yet hoped that the exceeding sensitiveness of the mag- 

 neto receiver would suffice to detect it. Our results, while not at all 



