THE AURORxl BOREALIS 721 



instant. In spite of persevering and most eager endeavour, I 

 have never succeeded in describing the formation of the Aurora ; 

 the phenomenon is there, but how or whence it comes it is im- 

 possible to saj. 



" Speaking quite generally, three forms are distinguishable : 

 quiet regular bows, which rise from the southern horizon, and 

 moving slowly over the zenith pale at the northern horizon ; then 

 bands of light, which, twisted in themselves, change place and 

 form continuously, and either consist of decided rays or j^else of 

 light matter ; lastly, the phenomenon of the corona, i.e.^ the 

 direction of the rays from or towards the magnetic pole. Gene- 

 rally the colour is a powerful white with a greenish tinge ; when 

 the motion and intensity are greater, particularly when the flash- 

 ing motion of the rays takes place, the prismatic colours often 

 appear in considerable intensity. 



" I spent much time and trouble on the spectral observations of 

 the Aurora, but the spectroscope we took with us was much too 

 small. I was never able to observe more than the well-known 

 green line ; in comparison with the spectral observations of the 

 Swedish Expedition with more perfect instruments, ours are 

 worthless. 



*^ Most of us made the observation that the Aurora was con- 

 nected with the state of the weather ; storms mostly follow 

 strong, and particularly the flaming Aurora. However the com- 

 parison of the meteorological data alone, can decide whether this 

 view is a right one. I myself have come to the conviction, with- 

 out being able to give any positive weighty reasons, but from 

 the observation of hundreds of Auroree, that the Aurora is an 

 atmospheric phenomenon, combined with the meteorological 

 conditions." 



11. Comparison of Auroras, Magnetic Disturbances, and. 

 Sun Spots. — (Prof. Loomis.) (From the American Journal 

 of Science, vol. v., April 1873.) 



A comparison between the mean daily range of the magnetic 

 declination and the number of Auroras observed in each year, and 

 also with the extent of the black spots on the surface of the sun, 

 establishes a connexion between these phenomena, and indicates 

 that auroral displays, at least in the middle latitudes of Europe 

 and America, are subject to a law of periodicity ; that their 

 grandest displays are repeated at intervals of about 60 years ; and 

 that there are also other fluctuations less distinctly marked which 

 succeed each other at an average interval of about 10 or 11 years, 

 the times of maxima corresponding quite remarkably with the 

 m.nxima of solar spots. 



Professor Loomis gives reasons for thinking that in high 

 northern latitudes the inequality of auroral displays in different 

 years depends more on their unequal brilliancy than on their 

 frequency, and on this account, and because the observations are 

 only made in northern latitudes for single years occasionally, they 

 have been omitted in making the comparisons. 



36122. Z Z 



