8^ Proceedings of ihe Ro)/al Frish Academtj. 



the water takes up a positive charge, while the air becomes negatively 

 charged. If the negatively charged air can be removed by ventilation, the 

 charge on the water can be measured. Lenard was able to obtain in this 

 way 'l-S electrostatic units per c.c. of water. In our experiments, owing to the 

 position of the receiver and the improbability of any considerable circulation of 

 air, it is impossible that more than a very small fraction of this amount of 

 electrification could be developed by splashing, and most of the rain examined 

 had charges per c.c. much greater than -15 electrostatic unit. Test experi- 

 ments we tried by splashing uncharged water into the receiver convinced 

 us that this source of error was sufficiently guarded agaiiist. 



JRemdts. 



We shall now deal with the results of the observations made during the 

 period from 3rd March to 22nd June. Very little rain fell during this period ; 

 and as most of the small rainfall took place during the night, the total amount 

 examined was small. Accurate measurements were obtained of the charge 

 on 5795 c.c. of rain; and as the measurements were made on thirty-two 

 separate occasions, they deal with a variety of types of rain, as we shall 

 see later. These 5795 c.c. correspond to a rainfall of only 1"2 cm. Of this 

 total 5113 c.c, or 88-2 per cent., were positively electrified, and the remaining 

 682 c.c, or 11'8 per cent., had a negative charge. 



Considering the total charge brought down by the rain we find that the 

 5113 c.c. of positively charged rain had a charge of 4151*-J: electrostatic 

 units, and the 682 c.c of negatively charged rain had a charge of 289'7 electro- 

 static units. Thus, of the total charge of electricity brought down by the 

 rain we examined, 93-5 per cent, was positive in sign. 



At no time during these observations did thunderstorms actiially occur in 

 the immediate neighbom'hood, although, on some occasions, thunderstorm 

 conditions appeared to exist, and on one such occasion distant thunder could 

 be heard. As the above figures show when compared with Simpson's results, 

 the great excess of positive over negative charge on rain is equally marked 

 imder nonnal conditions as during thixnderstorms. 



Perhaps the most striking result of our observations is that rain consisting 

 of exceedingly small drops appears to be always negatively charged. This 

 class of ra,in is generally light ; but its characteristic feature is the smallness 

 of the di-ops rather than the rate of precipitation. Thus rain of this class was 

 observed so light as almost to be imperceptible, and the same class of rain was 

 dealt with falling at the rate of '002 cm. per minute. We examined this type 

 of rain on fifteeu separate occasiuus, and found it always negatively charged. 



