HORIZONTAL FORCE OF THE EARTH’S MAGNETISM. 
401 
action is inverted at sunrise*. We can then only suppose that the moon may attract (or 
repel) the electric medium through which the solar disturbances are propagated. 
If the recurrence of great terrestrial magnetic changes at given solar meridians holds 
for every rotation of the sun, it is obvious that they should happen equally for all the 
different positions of the moon ; this constant recurrence, however, does not exist, and 
it is possible that this may be due to the particular positions of the moon, or that the 
position of the moon may determine the action on the earth. In order to examine this, 
the same 28 cases were arranged relatively to the moon’s positions with reference to the 
sun and to the equator. 
It was found that on the whole more of the large disturbances occurred when the 
differences of the sun’s and moon’s longitudes were near -|- or — 90°; but the excess in 
the quadrants having these differences for the middle points is not sufficiently great 
alone to give any considerable weight to the conclusion that the disturbances happen 
preferably near these positions. It may be remarked, however, that the four greatest 
disturbances during the two years, namely those on March 28 and October 19, 1844, 
and April 12 and December 3, 1845 (Nos. 1, 9, 15, and 27), in which the mean daily 
horizontal force diminished from 0 0027 to O'OOSO of its whole value, occurred when 
the moon was nearly in her first quarter. 
On the other hand, when the cases are examined with regard to the moon’s position 
in declination, there can be no doubt that there was one position of the moon for which 
the disturbances were most frequent, as may be seen in the following Table (p. 402). 
The diminution of magnetic force began in eleven cases when the moon was within 
two days of the greatest northerly position ; there were three cases within three days of 
the greatest southerly position (see Table V.). When the cases are arranged in groups 
having nearly the same declination, we find 
15 cases, 
5 „ 
3 „ 
3 „ 
2 „ 
mean declination beginning 18‘2 N., ending 14‘0 N. 
12-4 N., 
15-3 S, 
12-0 S., 
2-5 S., 
16- 4 N. 
8-3 S. 
17- 0 S. 
7-0 N. 
There were therefore 20 cases for which the diminution of intensity began when the 
moon was on the average 16° - 7 N., and which ended when she was 14° - 4 N. There were 
only 6 cases for the moon wholly south of the equator. The majority of cases began 
immediately after the moon had attained her greatest north declination (21° nearly). 
Since the moon took nearly the same time to pass from the equator to 15° N. as from 
15° N. to the greatest northerly position, I have sought for which of these two intervals 
the cases were most frequent. It appears that of the 20 cases which occurred with 
the moon north of the equator, 13 happened when her mean position (from beginning 
to ending of the diminution of intensity) was 1 8°*7, and 7 when she was 10 o- 2, so that 
* Trevandrum Observations, vol. i. pp. 117, 121, 123. 
3 K 
MDCCCLXXVI. 
