Discharge of Electricity. 397 



When p is small compared with 203, we have 



V= Ax/203 \/p; 



which agrees with what the conclusion of Masson* leads to 

 when corrected by means of the law regulating the discharge 

 between two balls (v. infra, p. 402). 



Messrs. De La Rue and Miiller f have investigated, in the 

 case of hydrogen, the nature of the curve for pressures lower 

 than that of minimum electric strength (*6 millim.). The 

 difference of potential is inversely proportional to the cube 

 root of the distance. Assuming that the discharge through 

 their long tube (33 inches long and 2 inches diameter) and 

 between a ring and a point electrodes is similar to that between 

 our disks at *5 centim. apart, I obtain for the complement of 

 the equation V = *024 v / {p 2 + 600 p} the following, 



V='67^, 

 Vp 



as the equation which is true for pressures less than *6 millim. 

 Let N denote the number of molecules in unit of volume ; 

 then when p is less than *6 millim., V is inversely propor- 

 tional to v^N; whenp is greater than *6 millim. but less than 

 600, V is directly proportional to \/N; and w T hen p is greater 

 than 600 millims., V is directly proportional to N. Thus, on 

 the one side of *6 millim. the greater the number of molecules 

 the greater is the facility for the discharge; while on the 

 other side the greater the number of molecules the greater is 

 the resistance to the discharge. This appears to be a strong 

 argument in favour of Mr. Crookes's contention for a fourth 

 state of matter; and the pressure *6 millim. seems, in the case 

 of hydrogen, to separate the two states. It is probable that 

 the mode of discharge is different in the two regions. 



Effect of Changing the Capacity of the Conductor. 

 The first two series of observations for varying pressure 

 indicate that a change of the capacity of the charged con- 

 ductor has little effect on the difference of potential required 

 to produce the discharge between the disks. But to test this 

 ;question further, I took readings for a *5-centim. spark through 

 air at the atmospheric pressure, with the capacity of the 

 charged conductor, 1st, increased by means of a large Leyden 

 jar, 2nd, increased by means of the couple of small Leyden 

 jars of the Holtz, and, 3rd, without being increased. I append 



* Ann. de Chim. et de Phys. 3rd ser. t. xxx. p. 41. 

 t Phil. Trans, vol. clxxi. p. Q5. 



