528 GENERAL REMARKS. 
stations in the second of these localities only, between Africa and 
New Zealand. At Hobart Town, Sydney, and King George Sound, 
there appears to have been little or no change in the dip since 
the commencement of the present century. 
The arrangement of the changes of dip in the southern hemis- 
phere in four divisions, characterised by an alternate increase and 
decrease of dip, is in correspondence with the double flexure of the 
lines of dip; and is a consequence of the western motion of the 
two southern magnetic poles. 
Careful observations made at St. Petersburgh, have shewn that 
the annual change of the dip in the northern hemisphere takes 
place altogether between the months of May and December; there 
being in fact a small movement in an opposite direction between 
December and May. This fact is of great interest in its bearing on 
the study of the causes of the magnetic phenomena. We have as 
yet no corresponding knowledge in regard to the southern hemis- 
phere. The magnitude of the annual change which Captain Fitz- 
Roy’s observations show is now taking place at the Cape of Good 
Hope, is deserving of attention in this respect. A large amount of 
annual change is obviously highly favourable for a determination 
of all the circumstances belonging to it; and its existence at the 
Cape, where there is already a fixed observatory, points to that 
station as most eligible for this investigation. 
The observations at Ascension shew that the epoch is fast 
approaching when the needle will pass from north to south dip at 
that island : it is extremely desirable that the period at which this 
change takes place should be determined with as much precision 
as possible. 
III. Intensity. 
I have discussed in the Seventh Report of the British Association, 
the very important inferences in regard to the general distribution 
of magnetism in the southern hemisphere, afforded by Captains 
King and Fitz-Roy’s most valuable series of intensity observa- 
tions ; but no inferences in regard to the changes which this phe- 
nomenon may be supposed to undergo can be drawn, as has been 
done in the cases of the variation and dip, because we possess no 
observations of the intensity made at a sufficiently early period to 
afford good materials for such a comparison. 
EDWARD SABINE. 
