86 
FIFTH REPORT — 1835 . 
the influence of a single axis alone : although derived from ob¬ 
servations made very near to the respective points, still the ob- 
Fig. 4. 
servations were not entirely free, particularly in the vicinity of 
the points of the weaker axis, from the influence of the more 
distant axis. To obtain the points of convergence which should 
be due to each axis separately, M. Hansteen proceeded as fol¬ 
lows : he selected certain of the most suitable observations of 
variation from which the situation of the points of convergence 
in the neighbourhood of the poles of the stronger axis were de¬ 
termined, and computed the disturbing influence on them of the 
weaker axis, employing for that purpose the approximate ele¬ 
ments of the weaker axis as already stated (except that a was 
taken at one third its former value), and assuming the proportion 
of the absolute forces of the axes to be as 2 to 1. The direction 
of the needle that would be due to the influence of the stronger 
axis alone was thus obtained, and the intersections anew of the 
variations thus corrected gave fresh points of convergence, 
which if the elements of the weaker axis employed in the cal- 
