ATMOSPHERIC MAGNETISM — CAPE OF GOOD HOPE. 
107 
minimum. This night effect of drawing the needle westward (3004.) proceeds slowly 
until 15^ or 16^ being assisted by the rising temperature on the east urging the end 
still more rapidly west until 20’’, when having reached its maximum in that direction, 
it at 21^ turns back and is driven to extreme east in the sun-swing, through an amount 
of variation more than twice as great as that produced in July or Cape winter. 
3039. I think the above is a true explanation of the reverse motion of the needle 
in the months of July and January, or Cape winter and summer. In winter the para- 
magnetic effect of cold air is on between 12'* and IQ'*, remaining longer on the east 
side of the magnetic meridian ; as it passes forward, both it and the sun region 
conspire at 19'* to carry the needle westward, for though they have opposite actions 
they are then also on opposite sides of the magnetic meridian (3005.). In the 
summer the cold region has much less power, occurs earlier*, soon passes over, for 
the summer sun is behind it, and then rather aids the sun in carrying the needle 
westward. 
3040. Some of the other months are still more striking in summer effect. February 
has a swing through 8' from west to east between 21'* and 1'* ; then from l'* to S’* it 
scarcely changes ; from S'* to O'* it follows the sun west ; from O'* to 10'* it varies but 
little, showing the merest trace of east effect about S'*; and after 10'* it passes west 
more and more rapidly, so that by 21'* it is at a maximum west, ready to swing back 
as the sun region passes over. The other and intermediate months are easily traced, 
and found to be beautifully consistent with the same principles of the hypothesis. As 
is evident, in almost every case each month partakes of the character of the preceding 
month in some degree, though not so much in this case of the Cape as in some 
others (3053.). The curves of December and January are more equal. 
3041. The time of the sun-swing illustrates exceedingly well the effect of the 
inclined magnetic meridian (3000.). In November, December and January, the swing 
is from 20'* to between l'* and 2'*. In these months the sun crosses the astronomical 
meridian about half an hour before he arrives at the magnetic meridian. In October, 
February and March, the swing is later, being from 21'* to 2'* or 3'*, for the sun then 
passes the magnetic meridian an hour or more later than the astronomical or time 
meridian. In September, April and May, the swing is still later, being from 22'* to 
2'* or 3’*, and the sun is still longer than before in reaching the magnetic meridian. 
In June, July and August, the swing is latest, being from 23’* to 3’*, and the sun is 
proportionately late in arriving at the magnetic meridian. What I describe as the 
passage of the sun is of course true of the warm region which precedes it ; but I prefer 
referring to the visible type rather than to the invisible reality, because it ties the 
considerations of time more simply together. 
3042. The inclination at the Cape varies singularly in the twenty-four hours, de- 
pending, I think, upon its mean degree. It is sueh that the warm and cold result- 
ants of action for the Cape will sometimes be above the line of the dip and sometimes 
* The minimum temperature below is three hours earlier, 
p 2 
