Loiv Frequency and their Resonance. 519 



wire from outside was sufficient to upset it entirely. In the 

 course of these experiments it was observed that a gentle 

 handling of the iron wire produced much more sensitive 

 instability in the resonance. I suspected therefore, that 

 mechanical vibrations of the wire might have something to do 

 with the phenomena observed, and although I have been unable 

 as yet to determine exactly to what extent these vibrations do 

 influence the effect of iron on the resonance yet I feel that 

 this effect is to a considerable extent due to molecular action 

 of the iron. The following two experiments will, I venture to 

 suggest, throw some light upon this point. 



Experiment 6. — A sensitive resonance was obtained by plac- 

 ing a sufficiently large number of iron wires into the inertia 

 coil until the addition of another wire caused a collapse. 

 After the collapse the wire bundle was gently raised and then 

 lowered again, 19 microfarads being plugged in the con- 

 denser. Now the capacity was varied by removing carefully 

 one plug after another from the condenser. The capacity 

 could thus be diminished fully 6 per cent without anything 

 like a corresponding change in the resonant rise of potential. 

 But a critical point in the variation of the capacity is then 

 reached, after which the slightest change in the capacity will 

 cause the resonant rise of potential to collapse suddenly, after 

 which collapse the plugging in of the condenser plugs did not 

 restore resonance. To bring resonance back again it was nec- 

 essary to raise and lower again the wire bundle in the manner 

 described above. If however the capacity was varied by even 

 less than 1 per cent, but in such a way as to allow a bright, 

 snapping spark to take place when the condenser phig was 

 removed, then the spark had almost invariably the effect of 

 causing a collapse of resonance. 



Experiment 7. — All the preceding experiments were now 

 repeated with the iron wire tied very tightly together and 

 when the whole bundle was introduced into the inertia coil it 

 was pressed tightly against the table so as to prevent mechan- 

 ical vibrations as much as possible, but the phenomena de- 

 scribed above appeared again. I observed however that the 

 sensitiveness of the instability of the resonant flow was not 

 quite as great as when the iron wires in the inertia were stand- 

 ing loosely. 



All these phenomena are observable at high frequencies also 

 but they are not marked as strongly as at low frequencies. 



While working with weak alternating currents obtained by 

 my electrodynamic interruptor, this peculiar behavior of iron 

 was never observed by me, probably because in these circuits 

 the magnetizations are so xoeak that iron is capable to follow 

 every impulse of the magnetomotive force. Hence the persist- 



