﻿Effect of Clouds on Ionization. 355 



a steady state ; this interval, in fact, was very small in our 

 experiments, certainly under one second. Nevertheless, we 

 found it convenient (except where otherwise stated) to allow 

 three seconds to elapse between the expansion and the 

 unearthing of the quadrants. For the same reason it is also 

 desirable to measure the ionization current through the cloud 

 over as small an interval of time as possible. Here again we 

 found three seconds to be a convenient and reasonable 

 interval. 



It is clear that the keys could not be manipulated by hand 

 with the degree of exactness required when dealing with 

 such short intervals ; accordingly, all three operations of firing 

 off the expansion, unearthing the quadrants, and stopping 

 the charging up were performed by electromagnets con- 

 trolled by a single pendulum. The latter consisted of a 

 massive iron block at the end of a brass tube 2'5 metres long. 



.Referring to the dotted lines in the diagram which indicate 

 electrical connexions, it will be seen that when the key K x 

 is depressed the pendulum P on its next passage through the 

 mercury will actuate the electromagnet E 1 which fires off 

 the expansion. If the key K 2 be depressed immediately 

 before the next passage of the pendulum through the mercury 

 in the same direction, the electromagnet E 2 will be actuated 

 at the passage and the quadrants will be unearthed. In the 

 same way, by putting K ? over to the dotted position the 

 magnet F 3 will be actuated at the next passage and the 

 connexion between the ionization-chamber and the electro- 

 meter will be broken. Both keys on being pulled up are 

 retained in position by small spring catches XX. Thus the 

 time between the expansion and the unearthing, and that 

 between the unearthing and cessation of the charging up, is 

 one complete swing of the pendulum (about three seconds). 



The keys F, G require brief mention. They consist 

 essentially of pointed brass rods resting against brass plates 

 supported on sealing-wax pillars. This arrangement was 

 adopted in preference to the usual mercury-cup type, as we 

 found the latter to give trouble on account of electrostatic 

 and thermoelectric effects. The shaded part of key F is of 

 ebonite enclosed as shown in a metal " umbrella," in order 

 to overcome a difficulty which arose from the charging up of 

 the ebonite in its very rapid passage through the air when 

 pulled up by the electromagnet. Immediately after the 

 key F broke contact with the brass plate it made contact 

 with the earthed spring g, thus preventing inductive action 

 on the electrometer system which would otherwise have 

 resulted from the continuous charging up of the key F. 



2 B 2 



