Nolan and Enright — Experiments on Large Ions in Air. 105 



conductor when tlie radius of the sphere is 10"° cms. With increasing radius the 

 effect of the charge diminishes. Tliis line of reasoning would indicate that 

 the radius of the large ion was distinctly greater than 10"' cms. 



Experimental Investigation. 



We have investigated this type of combination experimentally by 

 measuring the rate of disappearance of small ions in the presence of a 

 mixture of large ions and nuclei. A diagi'am of the apparatus is given in 

 fig. 4. A steady stream of air carrying large ions, which have been brought to 

 their stable condition, passes through the chamber B, where it is exposed to a 

 thin pencil of X-rays. The gas before entering the chamber passes through 

 the tube A, and after leaving passes through C and D in succession. A and D 

 are identical, being each fitted with a central insulated electrode 60 cms. long. 

 Voltages up to 800 volts can be applied to each tube, so that the gas passing 

 through can be completely freed from large ions, or if the central electrode is 



1^ \\^ Li_r^ 



Fig. 4. 



connected to the electrometer, measurements of the total charge can be made. 

 The tube C is fitted with four short insulated electrodes to serve for the 

 capture of the small ions at different times after they have been produced by 

 the X-rays. The voltage on this tube and the lengths of the terminals are 

 so adjusted that when each in turn is connected to the electrometer (the 

 others being connected to the outer tube), all the small ions arriving at that 

 .particular terminal are captured. The number of large ions drawn in at the 

 same time is not detectable. At H and K side tubes are provided by means 

 of which samples of the gas can be drawn off and examined for nuclei by 

 Aitken's counter. 



It is important in work of this kind that a very steady source of large 

 ions should be available. A diagram is given (fig. 5) of an arrangement 

 which proved very satisfactory in this respect. The flame burns inside a 

 large glass bell-jar sealed on to an iron plate. Through the plate passes a 

 tube fitted with a small platinum nozzle. Filtered coal-gas is supplied 

 through this tube and burns at the nozzle. Filtered air is passed in at a 

 steady rate from a gasometer. Tlie flame can be ignited without opening up 



