296 Dr. E. H. Cook on an Undulatory Movement 



a general thing ; for in other cases tried, the numbers did not 

 vary among themselves and agreed with those obtained with 

 the coils. 



The common centre of the circles, instead of being, as with 

 the coil, halfway between the terminals, seemed to be a little 

 nearer the positive electrode than the negative. Other expe- 

 riments are, however, necessary before this can be regarded 

 as established. 



In experimenting with this machine it is necessary to avoid 

 placing the plate too near the terminals, otherwise the powder 

 becomes electrified, and the plate also, and the resulting figures 

 are combinations of those showing the distribution of the 

 electricity upon the plate, and the undulatory effect discussed 

 in this paper. 



Most of the figures obtained with the machine were caused 

 by the spark passing between knobs. They are imperfect at 

 the sides, in consequence of the undulatory effect being pre- 

 vented from making itself felt in those directions by the solid 

 body of the knob coming between. Thus the portion of the 

 circles shown are at right angles to the direction of the spark. 

 The figure can of course be obtained from sparks taken 

 between points, but it is more difficult. It was particularly 

 noticed in the course of these and some other experiments 

 that, whilst substituting a knob for a point in a Wimshurst or 

 other machine increased the length of spark obtainable, it was 

 just the opposite with a coil ; for here the spark is consider- 

 ably reduced if taken between two knobs. The longest spark 

 with a coil is obtained either between two points, or between 

 one point and one knob or plate. 



It was also noticed that under the positive electrode the 

 powder was scattered so as to cause a space. This is some- 

 times, but not always, the case under the negative. 



11. Production of the Cleared Space, — It will be observed 

 that the more or less elliptical space in the middle of the 

 figures is not present in all. It is only produced when the 

 plate is held close up to the terminals and when the vigour of 

 the spark is considerable. At a first glance at a figure con- 

 taining the clear space, it appears that the lines are crowded 

 together at the ends of the major axis of the ellipse. Closer 

 examination and careful measurement will show, however, 

 that this is not the case, for the number of lines per unit of 

 length is the same whether measured at the ends of the major 

 or of the minor axis. In fact the appearance is just that of 

 concentric circles, formed of the powder, with the material 

 composing the smaller circles swept up so as to form the 

 irregular ellipse. 



