40 The Philippine Journal of Science ioib 



able, the wire was charged by a metal comb which was brought 

 near to another comb fastened to the negative terminal of a 

 Thordarsen induction coil capable of giving a 20-centimeter 

 spark. The silent leak across the air gap was usually sufficient 

 to keep the deflection of a Braun electrostatic voltmeter constant 

 at about 3,000 volts. After midnight, however, the humidity 

 was so great that the potential could not be maintained, so 

 a motor-driven static machine was substituted for the induction 

 coil, the aerial being connected directly to the negative terminal. 

 With this the potential was kept at 8,000 volts, the pressure 

 being regulated by varying the distance between two combs, 

 one of which was in contact with the wire, while the other 

 was grounded. After the wire had been exposed for thirty 

 minutes, it was wound on the reel and introduced into the 

 ionizing chamber. After ten minutes the leaf was charged to 

 240 volts by means of storage cells, and the time for it to move 

 40 divisions was recorded as the reciprocal of the amount of 

 active deposit on the wire. 



Amount of active deposit at Manila. — Elster and Geitel *- took 

 as a measure of the amount of active deposit present the leak per 

 hour in volts which would be caused by the active deposit collected 

 by a wire 1 meter long which had been exposed for two hours at 

 a negative voltage of 2,500, the testing system having a capacit>' 

 of 9.5. The capacity of my electroscope was about 18 electro- 

 static units, otherwise the experiments were carried on in the 

 same manner. A wire 30 meters long exposed at Manila gave 

 a mean discharge for 5 observations of 300 volts per minute. 

 If the capacity of my electroscope had been 9.5, as in the case 

 of the Elster and Geitel experiment, the discharge would have 

 been about 568 volts per minute. Therefore, a wire 1 meter 

 long would give a discharge of about 19. This is approximately 

 the same as the mean value found by Elster and Geitel for 

 Wolfenbuttel. 



Diurnal variation of the active deposit. — The diurnal variation 

 was determined as follows: The wire was exposed at 8,000 

 volts for a period of thirty minutes. It was then tested as 

 described above, and the time for the leaf to move 40 divi- 

 sions was recorded as the reciprocal measure of the amount of 

 active deposit collected. The actual determinations are given 

 in Table I. 



" Phys. ZeitBchr. (1903), 4, 526. 



