Air-Blast upon the Spark Discharge of a Condenser. 689 



if the phenomenon o£ multiple discharge is taking place, 

 then the images are each seen to consist of 4, 5, 6, or more 

 small sparks rapidly succeeding each other. Also, if there 

 is any sensible arcing at the spark-gap it is at once seen in 

 the image in the revolving mirror in the form of a trail of 

 light of a different colour to that of the true oscillatory 

 discharge-spark. On examining the spark-discharge with 

 and without the air-blast for certain short spark distances, 

 the effect of the air-blast is easily detected, because it is 

 seen to blow away this trail which accompanies the image of 

 the spark-discharge. 



In order to avoid misinterpreting the results, it was neces- 

 sary to be sure that when using say one single spark of 

 0*5 mm. in length and comparing it with the effect of 5 

 spark-gaps of 01 mm. in series, any increase in the current 

 at the discharge circuit was not due to an alteration in the 

 number of discharges. 



As far as could be observed this was not the case when 

 sufficiently short spark-gaps were employed. 



The following Table VIII. (p. 690) embodies the results of 

 the measurements. Employing single spark-gaps ranging in 

 length from 0*1 mm. to 3 mm., observations were taken of 

 the mean-square value of the current iu the discharge circuit, 

 both with and without the air-blast, and these values are 

 recorded in the first two columns. It will be seen that up 

 to a certain length of gap (about 2 mm. in the case of these 

 •experiments) the air-blast had a very decided effect in in- 

 creasing the mean-square value of the discharge current, but 

 beyond that point it seems to have the effect of diminishing 

 it. On the other hand, if a single gap, say of 0*2 mm., is 

 broken up into two gaps in series each of O'l mm. and so on 

 for the other spark-lengths, it is found that dividing the 

 spark-gap into two parts also increases the discharge current 

 up to about 0*8 mm., and after that the current diminishes. 

 If the double air-gap is blown upon, the current is increased 

 as compared with the same two gaps not blown upon, but 

 is not increased as compared with a single gap blown upon. 

 The same is true when the spark-gap is broken up respectively 

 into 3, 4, 5, and 6 gaps. The effect of dividing the gap up 

 to a certain point is to increase the discharge current, but 

 the effect of the air-blast on the multiple gaps becomes less 

 and less marked in proportion as the number of gaps increases, 

 so that the effect of dividing up a gap, say 0*6 mm., into 

 five gaps of 0*12 mm. each, is to increase the discharge 



Phil Mag. S. 6. Vol. 17. No. 101. May 1909. 3 A 



