94 I. H. Bigelow—The Harth a Magnetic Shell. 
mum of frequency is about 20° from the poles, in the sche- 
matic diagram, corresponding to the maximum trace in the 
ovals. The auroras fall off towards the pole and towards the 
equator. The long and the short periods of auroral visibility 
are clearly due to the change of intensity of the external field ; 
an increase of magnetic force meanivg a corresponding spread- 
ing of the field towards the equator; the astronomical changes 
of position of the poles, in summer and winter, to the equa- 
torial field affecting the conditions of visibility ; the atmos- 
pheric constituents also modifying them. Like phosphores- 
cence and fluorescence we may regard the auroral light as the 
product of the transformation of vibrating energy into the 
required period, by means of the atomic and molecular ele- 
ments of the air, as a system of step-up transformers. 
It is well known to meteorologists that the locality of gener- 
ation of distinct cyclonic circulations is to be found in two 
belts, (1) sub-polar and (2) tropical.* One may readily per- 
ceive that the main track of the movement of the Lows is just 
along the edge of the auroral belt A, in an oval following it 
closely (Dunwoody’s International Chart, 53). Furthermore 
the High Pressure Belt of Lower Latitudes is underneath the 
second belt of concentration T, and receives nearly half of the 
energy of the external field. rom this belt a series of storms 
is known to proceed northward, omitting the eastward drift 
due to the general circulation. Thus the Colorado, Texas, 
Gulf of Mexico storms and the West Indian hurricanes are 
examples of this type of storm generation. The eastward 
march of these storm centers seems to be in part regulated by 
the mid-latitude low magnetic belt, where the exflected and 
inflected fields touch each other. This is the significance of 
the third curve in the group given above, and points distinctly 
to the source of energy mentioned in my earlier paper. 
The problems of the seat of the so-called permanent mag- 
netism of the earth, the secular variation of the same, the 
magnetic storms or perturbations, the earth’s electric currents, 
and the sources of free atmospheric electricity, have an almost 
obvious explanation in general terms. If the nucleus of the 
earth cannot sustain magnetism, as it seems necessary to infer, 
then the reason for the earlier attempts to account for the sur- 
face distribution by a system of small equivalent magnets in 
the interior is obvious. Hence Gauss’s solution of the poten- 
tial is the only available method for interpolating at any given 
epoch. The secular variation of the gwase-permanent mag- 
netism of a shell rotating in two variable external fields is the 
real problem to be considered. In my papert some considera- 
* See this Journal, December, 1894, for the United States. 
+ Amer. Meteorol. Journ., April, 1892. 
o~ ete 
a a cla aa a EE LAL AL LLL - 
—— 
