276 
NATURE 
[Fuly 23, 1885 
nation-needle indicates the direction of the magnetic 
pole. At the present time in London the needle points 
18° to the west of true or astronomical north; hence, if 
auroral arcs were seen here to-night, their highest points 
would be nearly certain to be west of north. Next, the 
dipping- or inclination-needle (a very cheap and admirable 
form of which is now sold by Mr. Casella) points to the mag- 
netic zenith, which now in London lies 22” north of the 
true or astronomical zenith, in the magnetic meridian 
joining the north and south magnetic points of the 
horizon. Hence, if an auroral corona were seen here 
to-night, it would be nearly certain to lie in a point 22° 
north of the zenith. 
Let us limit ourselves for the present to the arc. In 
our latitudes, as has been said, it is seen to the west of 
north, generally low down near the horizon; but in the far 
north on the same magnetic meridian as ourselves it is 
seen east of south, while also in the far north, but in a 
widely different longitude—that of Behring’s Straits—it is 
seen north-north-east. 
Evidently, then, this are—this “common auroral arc,” 
as it has been called by Nordenskj6ld—is produced by a 
ring at some height between us and the north pole, but 
its centre does not lie at the north pole. Putting such 
observations as those referred to together, Nordenskjéld 
inferred the centre to be near the magnetic pole but not 
at it, in 81° N. lat. and 80° W. long., the thin ring of light 
haying a radius of 18° and a height of 200 kilometres. 
This, then, was Nordenskjéld’s main conception—an 
immovable common arc (a permanent stria, to speak in 
vacuum-tube language), though he acknowledged addi- 
tional ones sometimes, and shows by his observations 
that they are not always concentric. 
He also attempted to explain the frequency and posi- 
tions of arc aurore in different places by dividing the 
polar lands into five concentric regions (see NATURE, 
vol. xxv. p. 368). 
In Mr. Tromholt’s volume we find what may prove to 
be an immense advance on this view. He holds that ¢e 
auroral zone moves northwards and southwards daily, 
yearly, and eleven-yearly. 
Again, to speak in vacuum-tube Janguage, instead of 
one rigid stria, we may have many striz, and these 
moving towards or away from the auroral pole as ordinary 
striz move towards or away from the negative pole. 
Next, as to the proofs of this movement, some more 
quotations from Mr. Tromholt may be given :— 
“The daily period is apparent by a maximum of fre- 
quency and development which in most places in the 
globe occurs one to two, or three hours before midnight. 
This maximum seems, however, to occur /a¢ey the nearer 
we approach the magnetic pole. This will be clear from 
the following series, in which the figure in parenthesis 
denotes the geographical latitude and the other the hour 
when the aurora attains its maximum in the place 
named :— 
“ Prague (50), 83; Oxford (52), 93; Kendall (54), 92; 
Makerston (56), 91; Upsala (60), 94; Christiania (60), 
10; Bergen (60), 9%; Bossekop (70), 104; Pustosersk 
(70), 11-12; Quebec (47), 103; Fort Carlton (53), 124; 
Fort Simpson (62), 12; Point Barrow (71), 134. 
“For the Aurora Australis continuous series of ob- 
servations are almost entirely wanting. It seems, how- 
ever, from the fragmentary material which we possess, 
that the daily period for this does not differ from that of 
the Aurora Borealis. 
“ The individual types of the Aurora Borealis seem, like 
the phenomenon itself, to be confined to periods, and to 
attain their greatest frequency and highest development 
at certain periods. Thus, it appears from the observa- 
tion of the previously mentioned French expedition to 
Bossekop, that the arcs appear on an average at 7h, 25m. ; 
the streamers at 8h. 26m. ; the auroral clouds at 11h. 18m. ; 
the auroral waves between 13h. 12m. and 13h. 53m. ; 
the intensest colours at I1oh. 11m., and the greatest 
brilliancy between 1oh. and 11h.” 
Next as to the yearly change. 
Weyprecht was the first to advance the view that the 
auroral zone is furthest south at the equinoxes, and 
furthest north at the solstices. On this point Mr. Tromholt 
writes :— 
“My researches have led me to endorse Weyprecht’s 
theory. I feel satisfied that the dwrora Borealis moves 
towards the autumnal equinox southwards, and then 
northwards, reaching its furthest northern limit about 
solstice. After this it again moves southwards, being in 
zts most southern position at the vernal equinox, when the 
movement is again tn a northerly direction. 
vil IN 
Vil 
Fic. 3.—Curve of yearly auroral frequencies, Fritz. The Roman figures 
indicate the months. 
“ From this it follows that the two maxima occurring in 
the Temperate Zone at the equinoxes must approach each 
other more the further north the point of observation is 
situated. This is, in fact, the case. As some examples, 
I may mention that, whilst the two maxima occur [in 
March and September i in St, Petersburg, Abo, Stockholm, 
Christiania, Worcester (Mass.), and New Haven, they 
occur in February and October in Aalesund, Newberry, 
Quebec, and Newfoundland ; in December to January in 
Hammerfest and in January at Fort Reliance. Very 
instructive in this respect are also the observations 
from the three Greenland stations: Upernivik, Jacobs- 
havn, and Ivigtut. At Ivigtut, the southernmost of 
the stations, the yearly maximum must certainly be 
said to occur in January, but there is a second maxi- 
mum towards the autumnal equinox. At Jacobshavn, 
eight degrees further north, there is but one distinctly 
marked maximum, in January, and at Upernivik, the 
northernmost of the stations, the maximum falls at the 
winter solstice more marked and dominant than any- 
where else in the world.” 
(To be continued.) 
THE ECLIPSE OF CHUNG K’ANG 
I‘ China an eclipse of the sun is, and has in all ages 
been, considered as a bad omen. Indeed anything 
which disturbs the regularity of the movements or ap- 
pearances of the heavenly bodies is so considered. “On 
the first day of the last month of autumn the sun and 
moon did not meet harmoniously in Fang.” This pass- 
age occurs in the ancient classic, the “ Shu Ching,” in 
the “ Yin Cheng,” one of the books of the Hsia dynasty. 
Chinese commentators say that this passage refers to an 
eclipse of the sun in Fang, the fourth of the Chinese 
twenty-eight constellations. The last month of autumn, 
according to the Hsia Calendar, is the ninth month, the 
month after that month which contains the autumnal 
equinox, 
The constellation Fang extends from about a to o 
Scorpii, a distance measured along the ecliptic almost 
