117 
backwards and forwards, so that when the sun is in the 
north, a station near the equator, but north of it, exhibits a 
more distinctively northern character of oscillation; while, 
when the sun is in the south, it will exhibit a more or less 
southern character in its oscillation. 
If we now venture to ascribe the actions represented in 
(1), (2), (3), (4), to currents in the upper atmospheric regions, 
we shall have— 
(1) When the sun is north caused by a positive current 
going from south to north. 
(2) Caused by a positive current going from west to east. 
(3) When the sun is south caused by a positive current 
going from north to south. 
(4) Caused by a positive current going from west to east. 
The resultant of (1) and (2) would be a horizontal positive 
current going in a direction not far from south-west ; and 
the resultant of (3) and (4) a similar current going in a 
direction not far from north-west. The analogy in direction 
as well as oscillation to the two systems of anti-trades is at 
once apparent, and it will be strengthened if we reflect that, 
in the magnetical as well as the meteorological system, we 
must have a vertical current at the equator. This current 
might probably be represented by one carrying positive 
electricity down or negative electricity up ; whereas, that at 
the north magnetic pole might be one carrying positive 
electricity up or negative electricity down. We say most 
probably, because it is exceedingly difficult to imagine that 
either of these vertical currents goes through the lower 
regions of the atmosphere into the earth, and it is likewise 
very difficult to imagine that the system of currents is an 
open one. They must, therefore, somehow close themselves 
in the upper atmospheric regions, and we may thus, perhaps, 
imagine that while we have an ascending current at the 
north magnetic pole, we have a series of descending positive 
currents at the equator. 
