ON THE PERIODICITY OF THE AURORA BOREALIS. 



231 



1640, 1648, 1658, 1667, 1676, 1685, 1695, and 1704, are, by Hitter's supposition, years 

 of maximum for the aurora. Hansteen 1 appears to be satisfied with the result of Rit- 

 ter's prediction for the years 1806 and 1816 ; and he himself adduces the year 1843 

 as an additional confirmation of the theory. Fritz quotes from Hoslin 2 to the effect 

 that the latter had advanced a theory similar to Ritter's as early as 1784. Fritz him- 

 self calls attention to the fact that Ritter's maxima often concur with those of the solar 

 spots, and that, when they differ, the aurora mounts to at least a secondary maximum 

 in Ritter's years of mean obliquity, as was conspicuously true of the year 1797. The 

 year 1830, however, is an exception. But Fritz adds, that about this time the Ritter 

 and spot maxima were opposed to each other, whereas, in 1769, 1779, and 1788, the 

 periods of both classes of maxima coincided. 3 Bu'ckman also had intimated that the 

 great moon-period might determine the recurrence of auroras. 4 



table Lin. 



1 Xouveaux Mem. de l'Acad. Royale de Bru:c. XX. 118. 2 Meteorologische und Witterungsbcobachtungen, Tubingen, 1784. 

 3 Vierteljahrsschrift der Naturforsch. Gcsellsehaft in Zurich, X. 230, 258. 4 Gilbert, Annalen der Physik VII. 32. 



