100 DR. FARADAY’S EXPERIMENTAL RESEARCHES IN ELECTRICITY. (SERIES XXVII.) 
of power. All are important, and all should be considered at once. I presume that 
the theory of the variations cannot advance very far without their joint considera- 
tion. 
3018. Greenwich presents a fine case of the night episode, and the different direc- 
tions of the magnetic variation for the same hours in different months. In these respects 
it is very much like St. Petersburgh, but has great additional interest, because of the 
large western declination*, and the effect produced by it on the places of the active 
quadrants (2979. 3000.), and the times of the variation phenomena. On setting up its 
position on the globe (2996.), it will be seen that the equatorial plane is not likely to be 
much concerned in the midday action, and that the sun or warm region passes nearly 
across the middle of the two chief quadrants in summer; which with its nearness at 
the same time, ought to make the midday swing to east very great. In winter it is 
further off and in much weaker parts of the quadrant, so that the swing ought to be 
far less, and such is the case. The greatest summer variation is ll''30, and the least 
winter variation only 5''88. In April, May, June, July and August, the great west 
declination of the south or upper end of the needle is at 19'* 20', and the chief east 
position at 1^ 20'. The latter position remains the same all the year round, but the 
extreme westening is in the other seven (cold) months at 9*^ 20' and 11*^ 20' -f-, or 
verging towards midnight ; it then surpassing the morning west deflection. Thus the 
sun’s effect in summer, in weakening the cold night effect (3005.), is very evident; 
and so also is the manner in which the night action grows up, until very prominent, 
in the winter months, through the strengthening of the cold action (3006.), when the 
sun is towards the southern tropic and in the weaker parts of the segments. The 
assumed principles of this action have been already given in the case of St. Peters- 
burgh (3010. &c.). 
3019. The magnetic meridian is much to the east of the astronomical meridian, 
where the warm region passes it, especially in winter, for then the sun crosses it about 
10 o’clock, and in summer about 11 o’clock. Hence the swing ought to be earlier 
in winter than in summer, though, because of the slower angular motion of the warm 
region in relation to Greenwich (3015.), it ought then to occupy a longer time ; and 
yet, as above said (3018.), be, by reason of distance, of smaller amount. All this ap- 
pears to accord remarkably with the fact. The swing begins at 17^ in the winter but 
not until 19^ in the summer, and ending at the same hour at both seasons, namely, 
1 o’clock, is much longer in- its occurrence in winter than in summer. It begins earlier, 
because the magnetic meridian is sooner passed than in summer ; and the reason also 
appears why the extension in time is at the beginning rather than at the termination 
of the swing ; for, because of the declination, the warm region is at the same hours 
much less east of the magnetic meridian in the morning and much further west of it 
* Mean declination 22° 51' W. Mean inclination 69° N. 
t See the Curves, Plate II. The observations are only for every tw'o hours, so that no degree of nicety 
can be expected in assigning the time of any given change. 
