ATMOSPHERIC MAGNETISM — DIURNAL VARIATION — DIRECTION. 
57 
and night temperature is greater there than at other places, because the extent of the 
variation may be dependent in part upon that difference ; and next, whether the 
amount of effect to be expected is the same for the same difference in number of de- 
grees of temperature at every part of the scale (2886.). If the conducting power of 
oxygen (2800.) should be found by future experimental measurements (2960.) to in- 
crease in a greater proportion for a fall of a given number of degrees at lower tem- 
perature than at high ones (including the effect of contraction for that fall (2861.)), 
then it may be that parts more distant from the sun will be more affected than those 
under it ; or if the contrary be the case, less affected than otherwise would be ex- 
pected. 
2895. With regard to the daily variations, as respects the direction of the lines of 
terrestrial magnetic force, or the inclination and declination of the magnetic needle, 
the principles of the changes that may be expected to occur have been already referred 
to (2879.) ; and it remains for me to compare these expectations with a few simple 
cases of observation, in such a general manner as will tend to show whether the 
direction of action is, both in theory and fact, the same ; and whether there is any 
probability that the effect has been assigned to its true cause ; for this purpose I will 
confine myself entirely at present to a part of the daily variation, namely, the effect 
of the sun and air as the luminary arrives at and passes over the meridian. 
2896. Profiting by the last volume which has issued from the powerful mind and 
careful hands of Colonel Sabine*, I will take the case of Hobarton. The observatory 
there is in latitude 42° 52''5 south, and longitude 147° 27''5 east of Greenwich. 
The absolute declination is 9° 60''8 east, and the dip is 70° 39' south. In order to 
have the place of the sun and the time of maximum and minimum temperatures at 
hand, I have transferred the mean temperature for January (summer) for seven years, 
1841-48, and the mean temperature for June (winter) for the same period, cor- 
responding to every hour in the day and night, from pp. Ixxxiv. and cviii. to fig. 10, 
Plate I., where the middle series of numbers represents the hours, the line next below 
them a base line of temperature at 30°Fahr., and the two curves still lower down the 
mean hourly temperature for summer and winter. The short lines show generally 
the direction of the needle east or west of its mean position, the upper end being of 
course the north extremity. The positions about noon are distinguished by full lines, 
being those required for more immediate illustration. 
2897. The north end of the magnetic needle at Hobarton is most east at 2 o’clock, 
and most west about 21 o’clock. Being at the extreme west at the latter hour, it passes 
through the full range of variation, or to the extreme east in five hours, or by 2 o’clock, 
and then requires the remaining nineteen hours to return to the utmost west. The 
maximum east and west declination is at 2 and 21 o’clock for summer, and at 
3 and 22 o’clock for winter. The vertical positions show at what hours the declina- 
tion was 0, and correspond with Sabine’s zero. From 21 to 2 o’clock the needle 
* Magnetical and Meteorological Observations, Hobarton, vol. i. 1850. 
MDCCCLI. 
I 
