1891.] 



the Discharge of Ley den Jars. 



25 



the long wires close to the jars (Ci), and bridging them at their far 

 ends (C 3 ), fig. 14. The position Ci manifestly does not essentially 

 differ from A ; the position C 3 is the interesting one. Of course if C 

 were too short, the main spark occurred there instead of at A, but if 

 the main spark occurred at A, a much longer supplementary or recoil 

 kick spark often occurred at C 3 , especially 'when the capacity of the 

 jars and the length of the wires were suited to each other. As the 

 following table shows : 



Length of sparks. 



Without any jars 



With small Voss jars 



With pint jars (two in series) . . , 



Lengthen A till sparks just choose C instead 



Shorten A till recoil sparks just fail at C 



Get maximum C spark 



Without any jars again (size of knobs accounts 



for this slight difference) 

 With the two pint jars, in parallel, shifted to the 



far end of wire near position 3, with overflow 



knobs to represent C 3 

 Same arrangement of jars shifted back to near 



position 1 



Get the sparks at C 3 instead of as A 



Arrangement as at first 



Pair of gallon jars in series 



Large condenser (0'02 mfd.) 



Thus the large condenser is as much too big as the Yoss jars were 

 too small. The gallon jars seem to show the effect best. They were 

 therefore replaced, but this time insulated from the earth by standing 

 them both on the same insulating stool with tinfoil top. Very long 

 recoil sparks could now be got. 



