668 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1913. 



liculty was encountered in obtaining a powerful and, at the same time, 

 effective distribution of current. 



It may be of interest to note that the clue to the solution of this 

 difficulty came from an almost accidental observation. Working one 

 evening in the twilight, when the efficiency of the different points 

 could be roughly judged by the pale luminous discharge from them, 

 it was noticed that under the particular conditions employed at the 

 time this glow only became appreciable when the points had ap- 

 proached the plates almost to within the distance for disruptive 

 discharge, while at the same time a piece of cotton-covered magnet 

 wire, which carried the current from the transformer and commutator 

 to the discharge electrodes, although widely separated from any 

 conductor of opposite polaritj^, showed a beautiful uniform purple 

 glow along its whole length. The explanation lay in the fact that 

 every loose fiber of the cotton insulation, although a relatively poor 

 conductor compared to a metallic wire, was still sufficiently con- 

 ductive from its natural hygroscopic moisture to act as a discharge 

 point for this high potential current, and its fineness and sharpness, 

 of course, far exceeded that of the sharpest needle or thinnest metallic 

 wire. Acting on this suggestion, it was found that a piece of this 

 cotton-covered wire when used as a discharge electrode at ordinary 

 temperature proved far more effective in precipitating the sulphuric- 

 acid mist, which was then the object of study, than any system of 

 metallic points which it had been possible to construct. Perhaps 

 the greatest advantage thus gained lay in the less accurate spacing- 

 demanded between the electrodes of opposite polarity in order to 

 secure a reasonably uniform discharge. Much of the importance of 

 this discovery at the time lay in the limited potentials of a few 

 thousand volts then available to the experimenters in their laboratory 

 work. 



In practice, of course, a more durable material than cotton was 

 demanded for the hot acid gases to be treated, and this was found 

 in asbestos or mica, the fine filaments of the one and the scales of the 

 other supplying the discharge points or edges of the excessive fine- 

 ness required. These materials were twisted up with wires or other- 

 wise fastened to suitable metallic supports to form the discharge elec- 

 trodes in such wise that the current had to pass only a short distance 

 by surface leakage over them, the slight deposit of moisture or acid 

 fume naturally settling on them serving to effect the conduction. 

 With the further development of the electrical technique to provide 

 the far higher voltages now being used in commercial operation, the 

 choice and design of electrodes has become much more flexible, includ- 

 ing simple metallic wires, sharp metallic strips, and the like. In 

 fact, the very phenomena of so-called corona loss or direct leakage 

 from the wire into the air on high-tension transmission lines, which 



