relating to Atmofpheric EleElricity. 21 1 
A monthly account of eledtrical fparks, and of pofitive and 
negative eledtricity, as indicated by the pith-ball ele&ro- 
meter, and fometimes by only flaxen threads without balls 
to them. 
Number of days in each 
month in which fparks 
were perceived. 
Times. 
Times. 
Days. 
23 days of May, 1789, 
8 days of May, 1790, 
| Pofitive 
1 7 
Negative 
18 
9 
June 
Pofitive 
32 
Negative 
36 
12 
J u) y 
Pofitive 
13 
Negative 
22 
12 
Auguft 
Pofitive 
19 
Negative 
19 
9 
September 
Pofitive 
9 
Negative 
23 
7 
October 
Pofitive 
*7 
Negative 
7 
7 
November 
Pofitive 
12 
Negative 
8 
8 
December 
Pofitive 
12 
Negative 
6 
7 
January 
Pofitive 
26 
Negative 
4 
*3 
February 
Pofitive 
26 
Negative 
0 
3 
March 
Pofitive 
30 
Negative 
1 
3 
April 
Pofitive 
28 
241 
Negative 
12 
IS 6 
8 
98 
It appears from this journal, that there were only /even days 
throughout the year in which no figns of eledtricity were per- 
ceived ; viz. the 15th and 23d of November, and the 6th, 
15th, 17th, 2 1 ft, and 2 2d of December. 
Remarks on the phenomena exhibited by the rod on the 
31ft of Auguft. 
I was for a long time extremely puzzled to account for the 
rapid changes which the pith balls on fome days lo frequently 
F f 2 exhi- 
