1891.] the Discharge of Ley den Jars. 31 



arrangement with millimetre thread, and head divided into 400 parts, 

 made for reading Newton's rings, but supplied for present purposes 

 with a pair of knobs 1*940 and 1*965 cm. diameter, on insulating 

 glass pillars. These constitute the C spark gap at which the effect 

 of the recoil kick is to be observed. 



At the other end of the leads is arranged the condenser, usually a 

 pair of jars back to back in one line, so as to close the circuit in 

 a simple geometrical manner, and the Voss jars being at first used, 

 i.e., the small jars forming part of the Voss machine. 



The A knobs at which the exciting spark is taken are constituted 

 by the universal discharger, which, standing on a block under the 

 two parallel wires above it, and connecting them together through an 

 air gap, can easily.be moved along the table to and fro, always in con- 

 tact with the wires above it ; and its distance from the jars at one end 

 of the wires can be readily measured. This distance is called u, and 

 is indicated in fig. 16. 



The distance between the A knobs was supposed to remain con- 

 stant, but to avoid any uncertainty, and to eliminate the effect of the 

 different size of its knobs, the virtual length of the A spark was 

 measured with the spark-micrometer by bringing its knobs near enough 

 to just shunt out the discharger, so that half the sparks chose one and 

 half the other. The distance between the C knobs under these cir- 

 cumstances is entered in the table as the virtual length of the A spark. 

 They are then separated further, so that the spark occurs every time 

 at A, but it does not cease at C until they have been widened dis- 

 tinctly. The maximum distance to which they can be separated with- 

 out causing the C spark to altogether cease is then read, and the excess 

 of the one reading over the other constitutes the " recoil kick." This 

 procedure was then repeated for another position of the discharger A. 



The plan of operation usually was to begin with the discharger close 

 to the jars, and to move it away along the leads 5. cm. at a time, read- 

 ing in each position the minimum and maximum C sparks, as just 

 explained, the maximum reading being that at which C began to 

 fail. Tuning is more easily done by thus varying the inductance of 

 the discharge circuit than in any other way. 



A short range would have been sufficient to give the position at 

 which the maximum recoil kick occurred, but an extensive range was 

 often used with the idea of getting indications of harmonics. 



Measurement of Capacity of Jars used. 



33. Bough estimates of the jars employed in this and similar 

 experiments could be made in various ways, and no great accuracy 

 is worth aiming at with ordinary shapes of jar, because of the differ- 

 ence between the circumstance under which they are used from those 



