312 METEOROLOGY AND ALLIED SUBJECTS. 



observations that were used by Eitter, who, in 1803, announced his 

 conclusion that the frequency and magnitude of aurora stands in an 

 intimate connection with the eighteen-andt wo- third year period of 

 nutation, and such that the maximum of auroras coincides with the 

 mean inclination of the ecliptic. The agreement of Bitter's results with 

 observations is frequently very remarkable, but on several occasions of 

 maximum frequency his conclusion differs wholly from observation, and 

 at present we look to the sun, rather than to the moon, as the origin of 

 these disturbances. The great aurora period of fifty-five years is equal 

 to three times the nutation period, 18.6 years, whereby an apparent 

 further connection is made out. 



(21) The optical phenomena of the aurora are described in Chapter 

 11. After speaking of the elementary matters known to all observers^ 

 Fritz concludes that the general result of previous studies has been to 

 show that the frequency of the colors exhibited by auroras diminishes 

 in proportion as the color is farther removed from the brightest part of 

 the spectrum; the order of frequency being white, yellow, red, green, 

 blue, and violet. In the auroral beams the most frequent arrangements 

 of colors are as follows : 



Upper color Green. Green. Orange. Purjile. 



Lower color Yellow. Red. Violet. Blue. 



When a beam moves horizontally, its advancing side is red. The 

 direction of an auroral beam seems to coincide nearly with that of the 

 freely suspended magnetic needle; but considerable departures from 

 this rule seems to be noticed; thus, in 1848, Kowalski, at Obdorsk, ob- 

 served the beams crossing each other and passing over the zenith with- 

 out forming a corona. The formation of a crown or corona is an optical 

 phenomenon due to perspective. The dark segments and dark beams 

 depend on the condition of the atmosphere, and are not entirely due to 

 contrast with the bright portions of the aurora. 



(22) The brightness of the aurora occasionally surpasses that of the full 

 moon, or of the atmosphere illuminated by the moon ; but of its intrinsic 

 brightness as compared with moonlight, or the electric discharge in 

 vacuo we can know nothing until we can locate the distance of the 

 aurora from the observer. More important than the determinations 

 of the intensity of the light are the investigations in reference to the 

 peculiarities of the light, since from such studies we must hope to make 

 further progress in our knowledge of the nature of the aurora. The first 

 publication in reference to the spectrum of the aurora dates from 1868^ 

 when Angstrom announced to the scientific association at Upsala his 

 discovery that the aurora light is monochromatic, consisting of one 

 bright line, whose place in the spectrum is near the calcium line, and 

 which has a wave length of 5567. Angstrom further announced that 

 spectroscopic observations of the zodiacal light had shown him that the 

 same line was prominent therein, and as this line is not known to be 

 produced by the combustion of any terrestrial substance, it seems legiti- 



