OF ARTS AND SCIENCES. 229 



To obtain a sound of constant intensity, we blew the pipe by- 

 means of an air blast driven by the engine in the Rogers Building. 

 The air was regulated by two pressure regulators, one allowing 

 part of the air to escape, the other balancing the air pressure by 

 a column of mercury. The height of the mercury could be changed 

 at will. 



The pressure in the Blake contact is regulated by the attachment 

 of the carbon electrode to a spring, whose tension is adjusted by a 

 screw. In addition to the spring, which we used for preliminary ad- 

 justment, we applied pressure by means of a lever arm carrying a 

 scale-pan at its centre, one end of which rested on the electrode ; the 

 other carrying a knife-edge resting on glass, acted as a fulcrum. The 

 scale-pan was covered by a piece of velvet, in order that the addition 

 of weights might cause no jar at the contact. In our experiments we 

 found that any attempt to take off weights had the effect of disturb- 

 ing the adjustment of the contact to such an extent as to break the 

 series. This same result was frequently brought about by the jarring 

 of the ground from the street traffic. 



"We used a more powerful induction coil than that in the Blake 

 transmitter. The resistance of its primary was 0.5 ohm, and of its 

 secondary, 899 ohms. 



We experimented on various forms of battery with varying ar- 

 rangements of the cells, to observe the effect of changes in electro- 

 motive force and in resistance. The currents to be measured were 

 very small, and consequently some extremely sensitive form of electro- 

 dynamometer was required. We used one of the Kohlrausch pattern 

 with movable coils, which we wound of No. 40 (B. & S. gauge) 

 double silk-covered wire. The two outer coils might be used either 

 in parallel or in series with each other, and in either way with the 

 inner (suspended) coil. 



This dynamometer, which differed in some of its details from the in- 

 strument as ordinarily made by Hartmann, was constructed especially 

 for experiments of this nature by Mr. Otto Scholl, the mechanician 

 of the Laboratory. 



The condition of maximum sensitiveness with coils of a given size 

 is obtained by arranging them so that the product of the ampere-turns 

 in the outer and the inner coils is a maximum. 



In designing our inner coil, we were limited by the size of the tube 

 in which it turned. This coil we wound with as many turns as pos- 

 sible, giving a resistance of 180 ohms, which was not too large for the 

 best conditions, the resistance of the induction coil being hicdi. We 



