168 SEroRT<-1840; 
of the year 1844, and indeed by the others. Compared with the entire year, 
the three months, December, January and February, present by far the 
greatest electrical tension. In shorter intervals than five years, the months 
of maximum vary, sometimes occurring in the one or the other of the three 
months; but it appears from the entire series of five years, that the greatest 
tension is confined to the three months above-named. 
The shortness of the period over which the observations at 9 A.M. and 
3 p.m. extend, combined with the irregularity appertaining to the movements 
of a single year, render it impracticable to deduce the relation existing be- 
tween the values at those fixed epochs. Nothing further than the general 
fact, confirmatory of the results deduced from the observations of 1845 to 
1847, viz. that the tension in the forenoon hours is higher than that, in the 
afternoon, is likely to be attained. This general result, which is very striking, 
is exhibited in Table LXXIX. 
During the entire period the electric tension increased from sunrise to 
9 a.m.; the mean value of this increase on the seventeen months is 454 div. : 
this, however, cannot be considered as of equal importance with the mean of 
the year, because the last five months of the year 1843 contribute to its de- 
termination. With only one exception, viz. December 1844, the tension de- 
clined from 9 A.M. to 3 P.M.—imean value as before, 30°6 div. It is not to be 
considered that the tension actually declines from 9 to 3, for we have already 
seen that 10 A.M. is the usual epoch of the forenoon maximum, but that the 
tension on an average is lower at 3 P.M. than at 9 a.m. The table shows an 
increase from 3 P.M. to sunset, with two exceptions: December 1843 and 
November 1844, mean value as before, 16:0 div., with the same limitation as _ 
to the character of the increase, 4 p.m. being the usual epoch of the after- 
noon minimum. These movements are further illustrated by the next table, 
which exhibits the excess or defect of the mean electrical tension above or 
below the mean of each month. 
There are two or three numbers in the above columns that require a 
passing notice; most of them proceed very regularly, exhibiting a higher 
tension than the mean at 9 a.m. and sunset, and a lower tension at sunrise 
and 3 p.m. The first exception that we have to this order is in December 
1843, the mean tension at 3 p.m. being in excess, while that at sunset is in 
defect. In this month the double progression disappears, the tension de- 
clining 18°6 div. from 9 A.M. to sunset. The second exception occurs in 
February 1844, when the tension at 3 p.M. was 40 div. higher than the 
mean; the usual order of progression was not interrupted ; but from Table 
LXXVII. it would appear that the increase of tension giving rise to the 
anomaly just noticed, occurred principally between sunrise and 9 A.M., and 
wes maintained afterwards. March and April 1844 present similar excep- 
tions to each other in the tension at sunset being below the mean; the usual 
course of progression was not, however, interrupted in either case, as appears 
from Table LXXVIIL. The next exception occurs in November 1844, the 
tension at sunset being 22°4 div. below the mean: an inspection of Table 
LX XVII. indicates that the increase of tension, as in the former instances, 
took place between sunrise and 9 A.M., but was not maintained afterwards— 
in fact a diminution instead of an increase occurred at sunset; the increase 
between sunrise and 9 A.M. augmented the value of the monthly mean ten- 
sion, and this, combined with the reversal of the usual movement at sunset, | 
occasioned the depression of the mean at sunset below the mean of the 
month. In December 1844 there are no traces of the double progression, 
the tension increasing from sunrise to sunset: the epoch of mean tension for 
the month occurs between 9 A.M. and 3 p.M.; the signs of these mean quan- 

