347 



Variation of Frequency of the Aurora Borealis with the Moons 

 Age. — This investigation is evidently beset with considerable diffi- 

 culty, since the moonlight existing nearly extinguishes the appear- 

 ances of all the fainter class of aurorsc, and it renders the faintest 

 wholly invisible ; the careful watch, however, which was kept for 

 auroral appearances at Makerstoun, probably renders the table given 

 in the Transactions better fitted for such a question than any pre- 

 vious series of observations. 



It should be remarked, that the latitude of Makerstoun, or perhaps 

 even a lower latitude, is better fitted for this investigation, than much 

 higher latitudes ; at least this is the case as long as only frequency 

 of visibility can bo considered. The French Commission du Nord, 

 during their stay in Lapland, found aurorse existing, or probably ex- 

 isting, almost every night. In such places variation of frequency 

 there is none, and variation of intensity alone remains for investiga- 

 tion. It is obvious, that till some better mode of measuring this in- 

 tensity can be devised for these high latitudes, we are forced to per- 

 form this operation in a rude manner, by moving to lower latitudes, 

 where the fainter aurorse become invisible, and where, therefore, fre- 

 quency is a test of intensity beyond a certain limit. 



Combining the numbers of aurorse observed at each day of the 

 moon's age into six groups of 5 days (the first group, 4^ days), we 

 find the average number of auroras for one day of the moon's age in 

 each group as follows, from the 6^ years' observations : — 



Moon's age. SS""— i^ S*"— T"! 8<* — 12'' 13<»— IT^ 18>'— 22'' 2Z'^—2T^ 

 Number. 5-8 52 36 50 10-2 66 



Did aurorse occur indifferently at all ages of the moon, we should 

 expect to see the greatest number at conjunction, and the least num- 

 ber at opposition ; this, however, is not the case, the greatest num- 

 ber was seen about two days before the end of the third quarter, and 

 the least number about two days after the first quarter, or the visible 

 maximum and minimum occurred at times equidistant from the 

 epoch of opposition. The frequency of aurora;, therefore, is a func- 

 tion of the moon's age. In order to determine the actual law, we 

 may consider the probable effect of moonlight in obliterating the au- 

 roral appearances ; remarking, first, that 9'' p.m. is the epoch of 

 maximum frequency for the aurora, and that upwards of five-sixths 

 are seen before midnight. When the moon is about three days old, 

 in the months from September to March, it begins to set sufficiently 

 VOL. II. 2 G 



