St. John — Wave lengths of Electricity on Iron Wires. 317 



and also one pole of the condenser in the base of the coil; and 

 at Q was attached one pole of the coil and the other pole of 

 the condenser. The cups were filled with mercury to a height 

 of 8 mm and then filled with alcohol to within a few millimeters 

 of the top. They usually required cleaning only after several 

 hours' use, when the surface of the mercury consisted of very 

 fine globules, and sharp breaks were not made at each stroke of 

 the plunger as was indicated by the occasional failure of its 

 spark. The character of the spark depended much upon the 

 exact height of the mercury cups. The adjustment was best 

 made while the coil was in action by raising or lowering the 

 cups until the spark had a white body and a peculiar snap. 



The plates of the Hertz vibrator were 40 cm square and 

 fixed at 61 cm apart. The spark gap was supplied with 

 platinum-faced balls (3 cm in diameter) which worked with less 

 trouble than the usual brass ones. The side KM of the 

 secondary circuit was parallel to the conductor forming the 

 vibrator plates and fixed at 6 em distance with its center 

 opposite the spark gap. The long sides of the rectangular 

 secondary lay in a horizontal plane and ran through the center 

 of the room at a height of l;6 m above the floor. They were 

 held by their end supports at 30 cm apart. The induction coil 

 was 53 cm long, 19 cm in diameter, and was excited by the 

 current from five storage cells. A sparking distance of about 

 6 mm was most effective in producing oscillations in the second- 

 ary circuit. The following method was pursued in taking the 

 observations. The interruptor was set in action, the circuit 

 closed through the induction coil and an observation taken 

 of the first swing of the needle. The circuits were broken as 

 soon as the needle reached the end of its first swing, and the 

 extent of this excursion was the reading recorded. In accord- 

 ance with the experience of Paalzow and Rubens, it was found 

 that a steady deflection could not be obtained, but this first 

 swing was, under like conditions, satisfactorily constant, and 

 a preliminary calibration of the instrument by passing currents 

 of known strength through it showed that the square root of 

 the deflection taken in this manner was in a constant ratio to 

 the current. 



The same copper wire (diameter 0'1201 cm ) that had been 

 used in the secondary without end capacities was used for the 

 rectangle KLMN in fig. 2a, and with the exploring terminals 

 close to L and 1ST, the maximum point was found by lengthen- 

 ing and shortening the wire. Bolometer readings were taken 

 for each length used. To assure the constant activity of the 

 spark, a convenient length was taken as a point of reference, 

 and observations taken at this point before and after a series 

 of readings. If the spark had remained constant the readings 



