462 Messrs. V. Creinieu and H. Pender 



on 



was observed, as should be expected. However, when the 

 sectors S were charged so that the moving sectors passed 

 under Si carrymg their charge, the electrometer immediately 

 took a sudden deflexion, which gradually increased, the 

 more rapidly the faster the rotation. The sign of the charge 

 on the electrometer was the same as that of the charge 

 on the moving sectors, while the charge on the sectors Si 

 was of the opposite sign, as should be expected. However, 

 if the sectors S and M were then earthed, the deflexion of 

 the electrometer did not fall to zero, but indicated a residual 

 charge on the sectors Si, which charge was of the same sign 

 as that on the moving sectors. 



These facts seem to show that the rotation of the charged 

 sectors produced in the surrounding air electricity of the 

 same sign as the charge they carry. In accord with this 

 assumption is the above fact that the charge-current is in- 

 ferior to the discharge-current. In this connexion attention 

 ma}^ be called to the fact that the loss of charge from a 

 body in air is considerably less when the body is in rapid 

 movement than when the body is at rest. This fact was first 

 observed by Matteucci"^, under most excellent conditions. 



For the present, we merely state these facts without 

 attempting an interpretation. It is impossible to say whether 

 they should be attributed to the movement of the body itself^ 

 or rather to sudden variations in the field caused by the 

 violent shocks given to the air by the rapid motion of the 

 sectors. It should be added here that we tried to detect a 

 possible production of charge in the surrounding air- by 

 placing in the air between the charged disk and earthed 

 condensing-plate small metallic brushes connected to an 

 electrometer. The electrometer showed no deflexion when 

 the disk was set in rotation. 



2. Moving Sectors M hare, Fixed Sectors S covered with 

 Mica. — We found the same dissymmetry between the charge 

 and discharge current as in the preceding case, and in the same 

 direction. Further, for voltages above 3000 volts on the 

 fixed sectors S, the charging-current did not increase pro- 

 portionally to the voltage on the fixed sectors. When this 

 voltage reached about 4500 volts, the charging-current 

 appeared to attain a maximum, which could not be exceeded 

 even for a very great increase of voltage. 



Again_, w^hen the inducing sectors S were earthed after 

 havino- been submitted to a charge for several seconds, the 

 moving sectors continued to take up a charge as they passed 

 under S, as indicated by a current continuing to flow through 



* Ann. de Chimie et de Phys. 3rd series, vol. xxviii. p. 385. 



