80 PROCEEDINGS OF THE AMERICAN ACADEMY 



the linear limit of my Rheostat, which I reckoned equal to nine hundred 

 miles of ordinary telegraph wire. Up to this time I had thought of the 

 invention solely in its scientific relations, and was not pursuing that as 

 I ought to, as I knew that Prof. Bell had pre-empted the ground. 

 Indeed, I was waiting for an opportunity to see him to tell him to use 

 the permanent magnet, as every account of his work said he was using 

 a battery ; but when the great value of the invention was shown to 

 me, and I learned that Prof. Bell patented his work as soon as pos- 

 sible, I concluded to try to profit by it myself. I therefore began at 

 once upon a pair of instruments to that especial end, but before I had 

 time to complete them I heard that the device had already been pat- 

 ented ; so I ceased my hurry ujjon them. Indeed, I didn't finish them 

 for some weeks. In these instruments the magnets were of half-inch 

 round steel, and the bobbins were taken from some old telegraph instru- 

 ments that I found at Hall's on Bromfield St., as they just suited my 

 purpose. I have these instruments now, and they are very fair tele- 

 phones, and will do good work. 



I next had a pair made in much better shape, and used them in 

 several public exhibitions during the spring and summer of 1877. 

 Upon these instruments I had a flaring horn mouth-piece. At one 

 time I took them to the railroad between Elm St. station and North 

 Avenue station, attached the terminals to the two lines of rails at either 

 end, and was able to converse over the rails in that way. This was on 

 the seventeenth of July, 1877. In the attempts to improve the telephone 

 a great many hundred experiments have been tried, and among them 

 not a few devices that have since been patented by other parties. 

 Among the first of these was the using of a magnet upon each side of 

 the plate ; this was done as early as March, 1877, and that device was 

 exhibited at a public exhibition that I gave of the invention early in 

 April of that year, at Tufts College. 



In April, I fixed several sets of magnets and bobbins to act upon one 

 plate. The results were no better than when only a single pair was 

 used. In my instruments made in February the magnets were com- 

 pound horse-shoe magnets, such as could be bought in the stores ; and 

 such an arrangement I have uniformly found to be the best for loud- 

 ness of sounds. 



My next improvement consisted of placing a cushion of felt under 

 the vibrating plate ; this has been adopted in the making of telephones, 

 as it prevents the too rapid diffusion of the vibrations from contact, and 

 also acts as a damper to absorb the sympathetic vibrations of the plate. 



Next it occurred to me that, if the armature of an ordinary sounder, 



