416 The Law of tlie Condenser in the Induction- Coil. 



non-inductive resistance. The variable inductance was free 

 from iron. The curves in fig. 7 suggest that the optimum 

 capacity may be related to the inductance in the following- 

 manner : — 



-- + d, 



K = 



\//>-L + 



a, h, Cj and d being constants, 

 arising in an entirely different 

 paper on the Induction-coil, already referred to^. 



Curves of this form, but 

 way, are discussed in mv 



Table VII. 



Current = 



= 2-1 amperes. R^^lohm. E =4-2 volts. 



Inductance 

 in 



Millihenries. 



Optimum capacity in Microfarads. 



! 



! 





Efe^Oobm. lEi = -2ohm. 



E6 = -5 ohm. 'Rb='7 ohm. 



•00 



•182 



•226 



•263 i -280 



•03 





•226 



•300 1 •SOS 



•06 



•182 



•244 



•313 i -338 



•16 



•182 



•236 



•350 ! ^397 



•43 



•182 



•280 



•423 ^414 



1-60 



•192 



•293 



•420 I -506 



4-44 



•213 



•303 



•470 



6-04 



•189 



•318 



1 



7-64 



•283 







In all the experiments recorded above, unless otherwise 

 stated, the breaking-pole was negative. The solenoid used in 

 the experiments had an air-core and an inductance of '28 of 

 a millihenry. 



The optimum capacity was always taken as that capacity 

 at which the break-spark was reduced from a large spark to 

 a very small one. For small resistances and small currents 

 this point is usually very well marked, but as the current or 

 resistance increases it becomes more difficult to determine. 



A phenomenon which appears well worthy of notice and 

 which frequently attracted my attention in these experiments, 

 especially in the case of a large primary inductance, is that 

 as the capacity around the break is gradually increased the 

 volume and '' snappiness '' of the spark increases up to a 

 certain point, and then suddenly decreases, this value of 

 the capacity being that taken as the optimum capacity. As 

 the capacity is increased the potential appears to increase up 



* Physical Eeview, vol. xIt. pp. 300-304 (]902). 



