1912-13.] Records of Atmospheric Electric Potential. 197 
electrographs was not working on account of mechanical breakdowns. 
The curves reproduced show examples of the types of agreement referred 
to as A, B, and C. In order to facilitate the comparison, the time scales 
have been reduced to the same unit. 
It will thus be seen that for the six months, January to June 1912, 
42 per cent, of the 146 curves investigated came under group A, 37 per 
cent, under group B, and 21 per cent, under group C. 
Kahler,* in a series of papers in the Meteor ologische Zeitschrift, gives 
the result of a comparison of electrograph curves at three stations, two 
separated by 100 metres at Potsdam, and the third distant 14 km. He 
found a good agreement between the curves obtained from the various 
stations. 
As the Physical Laboratory is situated in the centre of Edinburgh and 
is surrounded on all sides by buildings and factories, there is always 
present in the air in the neighbourhood a considerable quantity of soot 
and dust particles. 
A comparison between the records obtained there and those at 
Blackford Hill Observatory should be of interest, as the Observatory is 
outside the town, and the air, if not free from soot, etc., is at least much 
clearer than that in the neighbourhood of the Laboratory. The two 
stations differ also in height above sea-level, the Observatory being about 
420 feet above sea-level, and the water-dropper 19 feet above the ground, 
while the Laboratory is about 245 feet above sea-level, the water-dropper 
34 feet above the ground. The distance between the Observatory and 
the Laboratory is 1*8 miles (2‘9 km. ), the Laboratory being north of the 
Observatory. 
While the curves obtained at the two stations differ from one another, 
the difference, generally speaking, is in the magnitude rather than in the 
direction of the variations. It is comparatively seldom that the potential 
is falling at the Observatory while rising at the Laboratory, or vice versa. 
In the magnitude of the variations, however, there is frequently a wide 
difference between the two stations. 
It is to be noted that the curves which show the best agreement are 
those which indicate a very disturbed state of the atmosphere. It is very 
seldom that a negative potential gradient is found at the one station 
accompanied by a positive gradient at the other. While the potential is 
negative, and on days when there is frequently a change from positive 
to negative values of the gradient, the curves are often almost identical. 
The occurrence of negative potential takes place most frequently during 
* Meteor. Zeit., 155-162, 289-299, xxv., 1908 ; 10-17, 347-355, xxvi., 1909. 
