136 Prof. J. Zeleny on the Ratio of the Velocities of the 



If, however, the air-blast and the rays are in action at the 

 same time, the air entering at A becomes partly ionized at 

 B, and in passing through the spiral C more of the negative 

 ions strike against and give their charge to the sides of the 

 spiral because of their greater diffusion- velocity; so that, on 

 leaving C, the air has a positive electrification due to the 

 larger number of positive ions present. 



In passing through the glass-wool S the charges on all of 

 the remaining ions of both kinds are given up, and the 

 electrometer shows a positive deflexion. 



The arrangement has the advantages that all the air in 

 passing through the spiral is near to a boundary, and that no 

 charge is allowed to collect at the places of separation, since 

 the metal parts are connected to earth and the charged air is 

 blown away. 



As the action is allowed to continue the deflexion of the 

 electrometer increases with the time but with continually 

 decreasing increments, because of the leakage through the 

 ionized air on the inside between the tubes L and D caused 

 by the electric force due to the charge acquired by D. 



The deflexions that are obtained in a given time depend 

 upon the velocity of the blast ; for if this is too slow there 

 is time while the air is passing through the spiral for all of 

 the ions of both signs to recombine or to strike the walls of 

 the spiral, in which case no deflexion is obtained ; and, on 

 the other hand, if the blast is too rapid there is not time 

 enough in passing through for the best possible separation of 

 the ions. Some intermediate value, where all of the negative 

 ions just manage to reach the spiral, will give the largest 

 value of the positive deflexion obtained. In a sample case, 

 by letting the action continue for six minutes the quadrants 

 of the electrometer became charged positively to 2'5 volts. 

 The spiral C was then taken out from the apparatus and the 

 positive charge now obtained was but a minute portion of the 

 former, showing the great advantage of making the air pass 

 through the narrow openings. 



There is a searching test, however, for showing that the 

 effect under consideration is due to a difference of velocity 

 of the ions. It has been shown (§ 4) that in carbonic acid 

 gas the velocities of the two ions are very nearly the same. 

 With this gas therefore the effect should disappear. Some 

 of this gas was accordingly placed into the gas-bag and the 

 experiments repeated with the spiral in position, with the 

 result that a small positive deflexion was obtained, but it was 

 so very small compared to that with air that it is easily 

 attributed to some traces of admixed air in the C0 2 . Experi- 



