158 



Mr. C. T. K. Wilson. 



is due to the continual production of ions throughout the air, this is 

 easily explained as indicating that the saturation current has been 

 attained ; the field being sufficiently strong to cause practically all the 

 ions which are produced to reach the electrodes ; the number destroyed 

 by recombination being negligible in comparison with those removed 

 by contact with the electrodes. Thus under the conditions of the 

 experiments the loss of electricity from the leaking system in a given 

 time is, if the charge be positive, equal to the total charge carried by 

 all the negative ions produced in the vessel in that time. 



The sum of the charges of all the negative ions (or of all the positive 

 ions) set free in the vessel is thus Tl x 4-1/300 E.U. per hour, or 

 4-3 x 10 -6 E.U. per second. If we divide by 471, the volume of the 

 vessel in c.c, we obtain for the charge on all the ions of each sign set 

 free in each c.c. per second, 9*1 x 10" 9 E.U. Finally, taking 

 6*5 x 10" 10 E.U., the value found by J. J. Thomson, as the charge on 

 one ion, we find that about 14 ions of each sign are produced in each 

 c.c. per second. 



There are, however, two defects in the older form of apparatus,, 

 with which the above results were obtained, tending to make this 

 number too small ; firstly, the field in the corners where the flat ends 

 meet the cylindrical wall must be very much weaker than elsewhere., 

 and some of the ions set free in these regions may have time to recom- 

 bine, although the strength of the field throughout most of the vessel is 

 more than sufficient for " saturation" ; secondly, since in this apparatus 

 both the rod supporting the leaking system and the contact-maker 

 projected for about a centimetre into the interior of the vessel, a. 

 certain proportion of the ions set free would be caught by them and 

 not by the leaking system. 



These defects are avoided in the other apparatus which has been 

 described (fig. 1). 



In this apparatus the capacity of the leaking system was 0*73 cm. 

 The constant potential of the supporting rod, and thus the initial 

 potential of the leaking system, was in all cases about 220 volts. 



At atmospheric pressure the fall of potential per hour was found to 

 be 2-9 volts. The loss of charge was therefore 0'73 x 2-9/300 = 7-1 



x 10~ 3 E.U. per hour = 2'0 x 10" 6 E.U. per second. This is the total 

 charge carried by all the positive ions, or by all the negative ions, set 

 free per second. The volume of the bulb being 163 c.c, the charge on 

 the positive or negative ions set free per second in each c.c. ^2-0 



x 10 -c /163 = 1*2 x 10~ 8 E.U., and the number of ions of either sign 

 set free per second in each c.c. = U2 x 10~ 8 /6*5 x 10~ 10 = 19. This 

 is somewhat greater than the number obtained before, but, as was 

 pointed out above, there were sources of error in the older apparatus 

 tending to give too low a result for the rate of production of ions 

 per c.c. 



