ATMOSPHERIC MAGNETISM — GREENWICH — HOBARTON — TORONTO, 
103 
warm region gradually into action with that of the nearer cold; that at first he stops 
the action of the latter, and then as he advances combines with and finally replaces 
it; causing the usual swing to come on, slowly at first and then quickly, from west 
to east by I*'. How this would happen is well seen, both as respects the place of the 
sun in the southern hemisphere, and in the two magnetic segments, by reference to 
the globe (2996.) and the diagram of the curves of variation, Plate 11. 
3025. Considering another and an intermediate month as March ; at P’ the upper 
end of the needle is at extreme east ; then until 9*^ it follows the sun as before (3023.). 
From to ll** it is stationary; then the paramagnetic action of cold from the east 
occurs, and the needle moves east until 13'^. It is then stopped, and two hours 
sooner than before ; for the sun now appears to Greenwich as early as 6 o’clock, and 
in a more favourable position for effect, both as regards the magnetic meridian and 
the segment in which it has for the time its place ; and so the needle is actually sent 
west for a couple of hours. It is then almost held steady until 19*^, after which the 
great sun-swing occurs. The holding west and yet the absence of more westening 
between 15^ and 19^, is not inconsistent with the southern place of the warm region, 
and it is probable that at that time the dip is increasing; an effect which would 
accord very harmoniously with the condition of matters at the time. 
3026. Other months are on this or that side of March in respect of their effects ; 
the corresponding month on the opposite side of the year (September) is the same as 
March, except in that portion of effect which is consequent upon a month following 
one that is warmer or colder than itself (3053.). Greenwich therefore satisfactorily 
illustrates the application of the hypothesis to the case of a difference in direction for 
the same hour in different months (3016. 3022.) ; and also the occurrence of the night 
effect, and its transition into the very marked eastening of the early morning, 
3027. The cases of Hoharton and Toronto are so similar, though in opposite hemi- 
spheres, that they may be considered together. A very important comparison of the 
phenomena at both places has been already made by Colonel Sabine in relation to 
the variations of declination, inclination, and total force*. When examined by the 
globe (2996.) the distribution of the quadrants is nearly alike, the sun being in two 
chief east and west quadrants, from about 18 to 6 o’clock, or during the day. The 
sun is in more influential parts of the quadrants in summer than in winter, and the 
effect is seen in the difference of the amount of declination variation. At Hoharton 
it is 12''05 in summer and only 3''6 in winter. At Toronto it is 14' in summer and 
5'‘2 in winter. The night action at both is alike in character, and has been suffi- 
ciently explained according to the hypothesis in the former cases (3010. 3024.). 
3028. Colonel Sabine has given the data by which the variations of the inclination 
and of the total force at Hobarton and Toronto may be compared with and applied 
to the hypothesis ; but I hesitate to enter upon them in this general view, inasmuch 
* Hobarton Observations, 1850, vol. i. p. Ixviii., &c. ; also Philosophical Transactions, 1847, p. 55, and 1850, 
pp. 201, 215, &c. See the Curves, Plate II., and Tables for Toronto, pp. 80, 81, 
