ST. JOHN. — WAVE LENGTHS OP ELECTRICITY. 235 



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 

 readino- recorded. In accordance 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, quite 

 constant, and a preliminary calibration of the instrument by passing 

 currents of known strength through the bolometer branch W^ showed 

 that the square root of the deflection was in a constant ratio to the 

 current. The needle was quickly damped by making and breaking 

 the circuit through the induction coil, with the interruptor in action. 



The rooms at disposal were very suitable for such an investigation. 

 The main room was 18 ra. long, 6 m. wide, and 5.5 m. high, and it con- 

 tained a very small amount of metal, and as it was in the non-magnetic 

 part of the laboratory that small amount was of brass and bronze 

 except the temporary addition of a small steam radiator in the corner 

 back of the oscillator. The oscillator was placed at 4.5 m. from one 

 end, and the parallel wires ran through the middle of the room at a 

 distance of 1.6 m. from the floor. The leading wires carrying the 

 currents from the batteries were of twisted cable and placed high up 

 against the walls. The bolometer and galvanometer were in an 

 adjoining room, where the observation table was equipped with the keys 

 necessary for complete control of the interruptor and induction coil. 

 By this means it was possible for one person to carry on the in- 

 vestigation, though it was very trying. Not only were the observer's 

 eyps in use, but it was necessary to listen intently to the sparking of 

 the oscillator, as after some experience very slight changes could be 

 detected and a close judgment formed of the steadiness of the spark ; 

 besides, it was necessary to note the sound made by the interruptor, 

 as small variations in its sjieed were easily noticed. The only time the 

 interruptor was likely to sliovv much change in speed was when the 

 battery was beginning to fail, or the brushes had become worn. 



The theory of the investigation was based upon the principle of 

 electrical resonance. 



Bjerknes has shown, in the paper previously noted, that, if damped 

 electromotive impulses obeying a sine law be assumed to act upon a 

 secondary circuit, there will be produced in the secondary circuit oscil- 

 lations of the period belonging to the primary impulses, and at the 

 same time oscillations proper to the secondary circuit, and that these 

 induced oscillations will reach their maximum amplitude when the two 

 circuits have the same period. His investigations also show that the 

 oscillations of the Hertz vibrator damp out much more rapidly than 



