190 J. Trowbridge — Electrical Discharges in Air. 



taper flame.) " About a quarter of the length of the tube, from 

 the lower or upper end, as the case may be, appears to be the 

 most favourable position." When the gauze is near the top a 

 slight modification is required, but the theory is essentially the 

 same. 



We have, then, three kinds of singing flames ; one depend- 

 ing on changes of pressure ; another on air currents ; a third 

 depending at once on both changes of pressure and on air 

 currents. 



In the above jDaper we have explained the general causes of 

 the phenomenon, but it is evident that in so complex a subject 

 it would be impossible to enter into more explicit details in 

 the limits of a simple article. It is hoped that these ideas will 

 attach a further interest to this old fashioned but interesting 

 experiment. 



Louvain. 



Art. XXL — Electrical Discharges in Air; by John 



Trowbridge. 



The flaming discharge from a large accumulator with its 

 nucleus, consisting of a dazzling white spark, is evidently a 

 form of voltaic arc ; and I was interested to discover if pos- 

 sible the mechanism, so to speak, of the voltaic arc. Does it 

 follow Ohm's law in respect to resistance, and is there an 

 oscillatory phenomenon ? It is well known that electric sparks 

 can be greatly increased in length by interposing a gas flame 

 between the terminals of a Ruhmkorf coil, or by moderately 

 rarifying the air between such terminals. The conditions in 

 the voltaic arc favor a greatly increased length of a disruptive 

 spark between the positive and negative carbons. This can be 

 seen in the photograph of the arc produced by a high tension 

 accumulator ; and doubtless the same phenomenon could be 

 observed in the ordinary voltaic arc if it were not so exceed- 

 ingly brilliant. 



I have lately studied the apparent resistance of the voltaic 

 arc in the following manner. In the circuit B, fig. 1, of forty 

 large storage cells, giving 80 volts, was placed a low resistance 

 choking coil, L, or coil of large self-induction. To the carbon 

 terminals, A, between which the voltaic arc was produced, were 

 led the terminals of a condenser, C. The latter was charged 

 by a step-up transformer, T. The oscillatory discharge of the 

 condenser was thus passed through the voltaic arc ; and a spark 

 in a gap in the circuit of the condenser was photographed by 

 the aid of a revolving mirror. The photographs gave the 



