G8 
FIFTH REPORT— 1335, 
on lines of equal dip the intensity is greatest in America 5 that 
it diminishes in going eastward towards Europe and Africa, and 
again increases, as the line of equal dip is traced, still eastwardly, 
towards the centre of Asia and the Indian Ocean, Also, that in 
equal spaces on a meridian, greater changes of intensity take 
place in the longitudes of America and New Holland than in 
those of Europe. It is unnecessary to dwell on the accordance 
which these facts present with the inferences drawn from the 
configuration of the lines of dip and variation, of the action of 
four points of directing influence. 
To obtain the approximate situations of these four points, 
M. Hansteen constructed tw T o maps of the polar regions of the 
globe on a polar projection. (Plate 3.) On the south-polar map, 
the variations observed by Cook and Furneaux in their voyages 
of 1773 and 1774 are represented by arrows, indicating the an¬ 
gles made by the compass needle with the geographical meridian. 
On the north-polar map are represented, in like manner, the va¬ 
riations observed in the northern hemisphere by Cook, Phipps, 
Lowenhorn, Schubert, Billings, and others, from 1769 to 1805. 
The head of the arrow marks the place of observation, and its 
direction is that of the compass-needle, so that the angle which 
it makes with the meridian is the observed variation. In the 
map of the south-polar region, all the arrows between the me¬ 
ridians of 60° E. and 140° E. are perceived to have nearly an 
uniform convergency : their directions prolonged w T ould all in¬ 
tersect somewhere about 135° E., and 69° or 70° S. latitude : and 
all the arrows comprised between the meridians of 240° E. and 
320° E. are convergent to a second point of intersection, situ¬ 
ated about 240° E. and 78° S.; whilst in the spaces intermediate 
between the meridians where the arrows are thus respectively 
convergent, the magnetic directions have no point of common 
convergence; the arrows prolonged do not intersect; ail point 
intermediately between the turn positions of directing influence ; 
but the particular direction of each arrow appears to be deter¬ 
mined by its relative proximity to one or other of the directing 
points, the influence of which predominates accordingly in the di¬ 
rection assumed by the compass-needle. 
M. Hansteen computes by spherical trigonometry the lati¬ 
tude and longitude of the mean points of intersection of the two 
converging portions of the observations represented by the ar¬ 
rows: the one to the south of New Holland is in 136° 15' E. and 
69° 27' S.; the other, to the south of Terra del Fuego, is in 236° 
43 ; E. and 77° 17^ S.; the observations employed in these de¬ 
ductions were made in the years 1773 and 1774. 
Looking next to the map of the north-polar region, we find 
