110 DR. FARADAY’S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XXVII.) 
that effect harmonizes with the idea that the previous holding of the needle in an 
eastern position, from 22^ to S*’, is the snn effect ; then the slow eastening from S'* 
to midnight and beyond, is the cold effect coming on. 
3049. Colonel Sabine has shown that the months of May, June, July and August, 
may be classed together, so that I will not speak of eaeh. Whilst they show the 
analogies they have between themselves, they also indicate the transitions to and 
from the other months. Let us consider September. From 7^ through midnight on 
to lO'*, the needle stands nearly at the mean. From lO** to 18**, the upper or north 
end eastens through the effect of the early morning cold. That the eastening should 
be fully effected an hour sooner than before (3048.) is quite consistent with the prin- 
ciples, for the path of the sun and its diamagnetic region is far nearer to the station 
than before, being now about the equator. From 18'' to 22'' it sends the end west- 
ward, in conformity with all former observations, and then comes on the sun-swing 
from west to east, between 22'* and 24’', and a hold at extreme east an hour longer. 
The shortness in time of this transit is, I think, a beautiful point. The sun is still 
north of I8t. Helena, but is now so much nearer that he passes through the same angle 
east and west, in respect to the place of observation, in less than half the time of the 
former sun effect in June (3041). After this the needle end travels west from C to O’*, 
following the sun as on other occasions ; and then from 6’' to O’* it moves a little east 
by the evening cold in the east, and remains near the mean position until the greater 
cold before sunrise (3005. 3011.) takes it more east between 16'' and 18'' of the 
coming day. 
3050. In looking at the curves of variation (Plate II.), it will be seen that the curve 
for the next month, October, is remarkable for being like in general character, and 
yet far removed in place from that of September ; and this effect appears due to the 
circumstance that the sun has now arrived at the latitude of St. Helena, or nearly so. 
According to my supposition, there has been a feeble night effect (3010.); and at 
midnight the needle is at the mean position and moving slowly westward, when the 
greater cold which precedes the sunrise coming into action on the east, counterbalances 
and arrests the western progress, and even draws the needle, as before, east a little for 
a couple of hours, till 18\ Even the sun region is at 16'' in that quadrant (the south- 
east one), that if it could act on the needle it would combine with the cold in the 
next or north-east quadrant in setting it eastward. By 18'' both the preceding cold 
region and the following sun region are so far advanced in their respective qua- 
drants that they combine to carry the needle end west as before, until 20'', and then 
comes on the swing from west to east until 24'“. Why this begins sooner, lasts longer, 
and is above four times the extent of the September swing, appears to be that the 
sun region comes up upon the latitude of St. Helena, and so acts in respect of the 
magnetic meridian more powerfully, and also sooner and longer : also that because of 
the mean declination west it arrives at equidistant points from and passes over the 
magnetic meridian sooner; and also from the effect of an accumulative action added 
to it from former months (3053.). 
