522 



have boon considerable, and the number given can only be taken as giving a rough approximation to the 

 absolute potential gradient. The potential gradient appears, then, to have a mean value somewhat less 

 than that met with in Europe.* 



3. ABNORMALLY HIGH POTENTIALS ASSOCIATED WITH DRIFTING SNOW. 



It was noticed by Mr. BERNACCHI that " as soon as the slightest amount of drifting snow appears in the 

 air the difference of potential becomes large and fluctuating." 



An analysis of all the occasions on which the potential at the ordinary height of about five feet exceeded 

 200 volts is contained in the table which follows : 



In this table are recorded the number of cases in which potentials between the limits given in the first 

 column were observed (1) when drifting snow is recorded; (2) when snow was falling; (3) with ice 

 crystals in the air ; (4) with fog ; and (5) under other conditions. 



Of the 19 occasions on which the potential exceeded 400 volts drift is recorded on 15. The total 

 number of occasions on which drift is recorded in the journal of atmospheric electricity is 26. Of the 

 seven. occasions when the potential with drifting snow was below 200 volts, it lay between 150 and 200 

 on four, between 100 and 150 on two, and below 100 volts on only one occasion. The mean value of 

 potential when the abnormally high values associated with drifting snow are excluded amounts to 92 volts ; 

 the mean for the 26 occasions on which drift is recorded is 668 volts. 



Except when driven along by the wind, snow or ice crystals in the air do not give potentials which are 

 in any way abnormal. The table which follows contains the results of all observations where there is a 

 record of ice particles in the air (whether recorded as drift or snow, or ice crystals) classified according to 

 wind strength. 



The table gives for various wind strengths the number of cases on which there is a record of snow, drift 

 or ice particles in the air, and the mean value of the voltage on such occasions. The classification is 

 according to the maximum wind strength when that is entered as varying between two limits ; t'.r/., a wind 

 recorded as of strength 3 to 5 would be entered in the last column. 



The mean value of tho potential gradient at KCH-, from sehvlvd dy- during the seven Tear* Islts l.i l!)Mt, j. ir,<) V1 -,ltu 

 per metre. (C. CHUKK, ' 1'hil. Trans.,' A, vol. 200, p. 302.) 



