236 PROCEEDINGS OF THE AMERICAN ACADEMY 



The similarity between the curves B and the curve C shows the 

 close agreement of theory with the results observed. 



The first part of the curve, corresponding to equation (2) of in- 

 complete contact, is the curve of imperfect transmission ; the last part 

 of the curve, corresponding to equation (1) of complete contact, is the 

 curve of good transmission. 



From equations (1) and (2) it will be noticed that the pressure 

 required for good transmission of sound is dependent on the intensity 

 of the sound, good sound being transmitted if 8 is less than P. The 

 sound in the receiver is loudest when S is nearly equal to P. 



In experimenting with electro-motive forces greater than 3 volts, 

 we met with unsatisfactory results. Good sound was not transmitted 

 except under heavy pressure, and all attempts to obtain satisfactory 

 measurements failed on account of the well-known disturbances set up 

 in the microphone by the current itself. 



In certain of our experiments the resistance of the primary circuit 

 was diminished by joining a number of cells in parallel. The uniform 

 result was, that the sound transmitted was louder. 



The results of our experiments may be summed up as follows : — 



The resistance of the primary circuit, and especially that of the bat- 

 tery, should be as low as possible ; the pressure at the contact should 

 be no greater than is required to transmit good sound, — that is, it 

 should be a little greater than that required to give the maximum 

 current ; with the present form of Blake contact, no electro-motive 

 force greater than 2 volts should be used ; and, finally, the contact 

 should be carefully guarded against jarring. 



Our work should be considered, not as a complete investigation, but 

 as part of the foundation for future work in the Rogers Laboratory ; 

 for our results have been more in the invention of methods than in 

 the use of them. We trust that the work in which we have had a 

 share may be successfully carried out in the future. 



Rogers Laboratory of Physics, 

 January, 1888. 



