642 REPORT— 1892. 



across the sensitive surface of ordinary gelatino-bromide dry plates. Discharges 

 (1) from a Wimshurst influence machine, (2) from an induction coil, and (3) from 

 a Tesla high-frequency coil were among those shown. The entirely distinctive 

 character of figure produced by positive and bv negative discharges respectively 

 was very noticeable in all the photographs, the former being of an irregular 

 description somewhat resembling some forms of seaweed, the latter forming^ 

 straight lines or sweeping curves with feather-like extremities. In fact, while the 

 negative figures point to something of the nature of momentum in the discharge, 

 and give the idea of a stream issuing from the electrode and splashing over the 

 surface of the glass, the positive figures, on the other hand, seem more nearly 

 allied to streams flowing fi-om the glass to the electrode and sucking from all 

 portions of the plate. A remarkable similarity was noticeable between some of 

 the figures and organic forms, such as some descriptions of seaweed in the case of 

 the positive, and feathers or palm-leaves in the case of the negative. The figures 

 in fact present another curious instance of the remarkable manner in which under- 

 favourable conditions inorganic matter will assume the structural forms generally 

 associated with organic life. The figures appear to be accurate representations of 

 the actual discharges produced by the ordinary photographic action of the lumino- 

 sity of the latter, and the gelatine film seems to have no appreciable eflect in 

 altering the form of the discharges, which are the same as those obtained on 

 plain glass. 



6. Slow Oscillations produced on discharging Electric Condensers of great 

 Gapacitij. By James H. Gray, M.A., B.Sc. 



About the end of last winter Lord Kelvin asked the author to try to obtain 

 oscillations on the discharge of a condenser through inductive resistances which 

 would be slow enough to be observed by means of a mirror-galvanometer. For- 

 that purpose use was made of a set of eleven large condensers (total capacity about 

 140 microfarads) which are in the Physical Laboratory of Glasgow University. 

 The capacity being so large, the author hoped to be able to detect oscillations of 

 periods comparable with one second without making the ratio of damping in- 

 conveniently great. Applying the now well-known differential equation — 



d^ Hdq l^ Q 

 dt'' h dt GlJ ' 



where the symbols have their usual meaning, we see that, in order to get slow 

 oscillations, the resistance of the discharging circuit must be as small as possible 

 compared with its self-induction, and the capacity of the condenser as great as 

 can conveniently be made. The author endeavoured to obtain these conditions, 

 and succeeded in so far that, though the ratio of damping was considerably 

 removed from unity, it was found possible to record at least three semi-oscillations 

 on each discbarge. When the frequency of the oscillations is very great (for 

 example, about 10,000 -per second, as in the experiments of Feddersen, Paalzow, 

 Blaserna, and others), it is very easy to make the ratio of damping practically 

 unity ; but when the frequency is only two or three per second, this is very 

 diificult, and the only way to obtain results is to make the capacity and self- 

 induction as large as possible. In order to obtain a high self-induction, a large 

 electro-magnet which formerly belonged to a dynamo was used as part of the 

 discharging circuit. The diameter of the iron core was 13 centimetres, the 

 length of each coil of wire being 28^ centimetres. There were on each coil about 

 44 lbs. of No. 20 B.W.G. covered'copper wire, 7,300 turns, and 28 layers, the 

 outside diameter of the coils being 18| centimetres. The resistance of each was 

 about 95 (B.A.) ohms. It was found difficult to adjust the mirror galvanometer 

 to detect the oscillations, but they were observed by means of a telephone of low 

 resistance, and distinctly seen when, at the suggestion of Mr. C. T. Littlewood, B.Sc, 

 a thin strip of sheet iron was placed with one end about 5 millimetres from one 

 pole of the magnet, so as to vibrate freely. At each semi-oscillation the sheet 

 iron was attracted. To prove that this was not due merely to the free period 



