140 Mr. C. D. Child on the Production of 



It is first shown that the region to the left of c loses its 

 conductivity when a current is passing between b and c, 

 showing that the ions are approximately all withdrawn from 

 this field by the current. His further statement is as follows r 

 " Using the first one b as cathode, and the wire a as anode, 

 the current was 2*7 milliamperes. As already explained, the 

 removal of positive ions by b was practically complete. Thus 

 the current is a measure of the number of ions that reach 

 this point. Next, using c as cathode and b as anode, the 

 current was found to be diminished to 1*3 milli;imperes. It 

 follows, then, that in the absence of a field,' the number of 

 positive ions and consequently also the number of negative 

 ones reaching c is about half the number that reach b, the 

 remainder recombining between b and c. If, then, we make 

 c cathode aud b anode, we take out as many negative ions at 

 b as we take out positive at c. In other words, under the 

 particular conditions of this experiment, we take out half 

 the entire number of negative ions that reach b, and we 

 reduce by half the rate of recombination at this point. Yet 

 the luminous intensity is not at all affected. It does not 

 appear, therefore, that the luminosity can be attributed to 

 recombination/' 



This argument, however, is not conclusive, since the rate 

 of recombination is not necessarily reduced by half when 

 the number of negative ions passing a given point is reduced 1 

 by half. The rate of recombination depends on the number 

 in a region at a given time, and this depends not only on* 

 the number passing a given point but also on their velocity 

 after the point is passed. If, for example, the number of 

 negative ions passing 6 and the velocity with which they 

 move are both reduced one-half, then the number of ions 

 per unit volume is not changed and the rate of recombination 

 is the same as at first. 



In reality this is what occurs in the second case suggested 

 by Strutt. The field not only causes fewer negative ions to 

 pass b, but it also pushes those which do pass against the 

 flow of the gas, so that they move more slowly. Consequently 

 the number per unit volume is but slightly changed, as will 

 be shown in a following paragraph. To make this somewhat 

 clearer the following conditions may be assumed. If 100 

 negative ions pass a we may have the following cases : — 



(a) When there is no field, 50 may recombine between a 

 and b, 25 between b and c, and 25 pass through the wire net 

 at c ; 



(b) When a is positive, b is negative, and c is uncharged., 



