280 CARNEGIE INSTITUTION OF WASHINGTON. 



by the radioactive material previously collected still contribute their effect, 

 so that the electrometer deflection falls, not to its true zero, but to a value 

 determined by the radioactive effect; and the difference between this "appar- 

 ent zero" and the deflection previously shown by the electrometer is the proper 

 quantity from which to calculate the true conductivity of the air. It will, of 

 course, be observed that any other variations of zero of the instrument become 

 automatically eliminated by the above procedure. 



In order to hasten the attainment of the appropriate zero-reading when 

 the potential of 250 volts is applied, an automatic arrangement is included 

 'oy which the two quadrants are first connected together for an instant through 

 a potential difference just sufficient to cause the spot to settle down near the 

 apparent zero-reading. The necessary potential-difference is obtained by 

 utilizing a sort of potentiometer constructed from a graphite line drawn upon 

 a sheet of ground glass, the ends of the graphite line being connected to the 

 poles of a single cell, and contact being made to any desired point by another 

 graphite line drawn to that point. The proportion of the e. m. f. of the cell 

 tapped off by this potentiometer can readily be adjusted by rubbing out or 

 thickening the graphite line on one side of the mid-contact point. The 

 potentiometer system would be unnecessary were it not that, even when the 

 electrometer is disconnected from the central cylinder, the reading obtained 

 on joining the quadrants by a wire is appreciably different from that obtained 

 when they are disconnected except through the radioactive cell. 



The main central cylinder is insulated in an amber plug protected by a 

 guard-ring which is kept at the potential to which the electrometer is raised. 



If C is the capacity of the portion of the main concentric-cylinder system 

 which is exposed to the air-current, V the potential to which the central cylin- 

 der is raised, X the appropriate unipolar conductivity, and dQ/dt the rate of 

 passage of electricity through the radioactive resistance, we have, 



Arrangements are made by which the apparatus may be cahbrated at any time 

 and the quantity dQ/dt rapidly obtained in terms of the electrometer deflec- 

 tion. The scheme is such that the calibration data are printed directly upon 

 the same trace as the conductivity record. The general method is as follows: 

 To the part of the apparatus in permanent conducting communication with 

 the main central cylinder is fastened a small brass cylinder which forms the 

 inner member of a small cylindrical condenser. The outer member is con- 

 nected to a sliding contact which moves on a long wire spiral wound on a 

 drum, one end of the wire being connected to the earthed case surrounding 

 the whole apparatus. Suppose now that ions are prevented from entering the 

 apparatus by means of the attachment already described. If a potential 

 difference is maintained between the ends of the spiral, and the latter is rotated 

 so that the contact slides along it, the electrometer will deflect, and, eventually, 

 a steady reading will be obtained which is determined entirely by the resistance 

 of the radioactive cell, the rate of alteration of potential of the sliding contact, 

 and a quantity k, representing the mutual capacity of the small cylinder con- 

 nected to the sliding contact and the cylinder which surrounds it. If dV/dt 

 is the rate of alteration of potential of the sliding contact, we have, for 

 this case, i,<E. = ^ 



dt dt 



By altering the potential difference between the ends of the wire spiral, 

 dQ/dt may be obtained for any value of the steady deflection, without vary- 

 ing the speed of rotation. 



