74 Mr. W. Spottiswoode on Stratified Discharges. [^lay 18, 



Wiillner's observatious appear to have been dii-ected rather to the 

 nature of the coil-discharge than to that of the stratifications ; and some 

 of his principal conclusions are accordingly of the following kind, ^iz. 

 that at low pressures, i. e. down to 1 millim., when the discharge was 

 stratified, the striae showed an interniittence of intensity indicating 

 either a pulsation within the duration of the main discharge or a break- 

 ing up of the main into a series of partial discharges. At greater 

 pressures, e.g. at 26 millims., when almost all trace of stratification was 

 lost, this breaking up into partial discharges (especially at the com- 

 mencement) was distinctly perceptible. At yet greater pressures, i. e. 

 from 40 millims. to 75 miUims., a cloudy kind of stratification showed 

 itself; but, excepting a bright flash at the outset, no appearance of 

 partial discharge was visible. The observations, which were at first 

 directed to capillary tubes, were extended to tubes of various diameters, 

 and also incltided the effect of a magnet on the discharge. 



For some time prior to the pubhcation of the volmne in question I 

 had been engaged upon a series of experiments very similar in their 

 general disposition, but with a somewhat different object in view, viz. 

 the character and behaviour of the stride ; and of these, together with 

 some recent additions, I now propose to offer a short account to the 

 Society. 



My general instrumental arrangements appear to have been similar to 

 those of "W'iillner ; in fact they could hardly have been very different. 

 The tubes were attached to the coil in the usual way, and a contact- 

 breaker of the ordinary form A^ith its own electromagnet was in the first 

 instance used. By suitably adjusting the velocity of the mirror to the 

 rapidity of the contact-breaker the image could be kept tolerably steady 

 in the field of view. In order to obtain gi^eater steadiness a special 

 contact-breaker was next deAised. This was mechanically connected 

 with the spindle of the mirror, and so arranged as to break the current 

 when the image was in the centre of the field of ^iew. The only point 

 in this part of the apparattis which requires special notice is the fact 

 that this contact-breaker, like all others, should be placed in close 

 proximity to the condenser of the coil, otherwise a great loss of light 

 is sustained. For the last-mentioned form there was finally substittited 

 a mercurial break (successfully arranged by my assistant, Mr. '\\^ard), 

 the plunger of which works on a cam attached to the axle of the mirror ; 

 so that the action of the contact-breaker is regtilated by that of the 

 mirror, instead of the reverse as in the former arrangement. With the 

 broader tubes a slit was tised ; with the narrower this adjtmct was less 

 necessary; while vrith. capillary tubes, such as are used for spectrum- 

 analysis, it coidd be dispensed ^ith altogether. 



Li experiments for comparing the tmstratified statical discharge with 

 the stratified at the same pressure of gas within the tube, and for ob- 

 serving the transition from one to the other, a Leyden jar and a spark 



