McClelland and KunnedY — The Large Ions in the Atmosphere. 79 



ionisation measured by observing the saturation current in the larger 

 apparatus consists almost entirely of large ions, and their number per c.c- 

 can be at once obtained if we assume that each carries the electronic charge. 



Over 400 sets of observations were made of this number ; and values were 

 found varying between 3,700 per c.c. and 60,000 per cc, the average number 

 being about 16,000. The observations extended with intervals for a period 

 of over twelve months from May, 1911, to June, 1912. The number present 

 is subject to large and rapid variations ; and the method employed in setting 

 forth the general results of the observations is to give merely the average for 

 any one day, unless when very large variations were observed or other 

 points of interest occurred. These are specially noted. 



In the earlier stages of the work we took some care to examine whether 

 there was any difference in the numbers of positive and negative large ions ; 

 but we found no evidence of such a difference of appreciable magnitude. In 

 the table N denotes the number of large positive ions per c.c, and n the 

 number of small positive ions per c.c. on the occasions on which they were 

 measured. The small values found for n will be discussed later. 



Discussion of Observations. ' 



The following tables show the very wide variations in the value of N. It 

 is difficult in seeking a cause for these variations to associate them 

 with definite accompanying conditions of the atmosphere, particularly 

 when it is considered that sudden changes of great magnitude often take 

 place without any apparent change in the meteorological conditions. Neglect- 

 ing abnormalities, there is no great difference in the mean value of N for 

 each month over which the observations extended. There is, perhaps, a 

 slight excess for the winter months. The changes seem to follow rather 

 the changes of weather. In the very dry and warm weather of May and June 

 the values of JV varied for the most part between 6,000 and 10,000, and 

 similar small values have been found in dry weather in November and during 

 a period of dry frost in January. Damp and wet weather seems, on the other 

 hand, to be accompanied by greater values of iV. 



Abnormally high values of JV were obtained on a few occasions during 

 an exceptionally heavy rainfall. A great number of observations have been 

 made during rain ; but the exceptionally high values were on each occasion 

 accompanied by an abnormally great downpour, the rain-drops being very 

 large and falling vertically. On one of these occasions N was 60,000. The 

 value 53,000 was obtained during a very dense " mist " fog. High values 

 were always obtained when observations were taken during fogs. On a few 



E.I.A. I-ROC, VOL. XXX., SECT. A. [13] 



