1873.] 



on a tendency observed in Sun-spots. 



401 



Hemispherical excess. 



Date. 



North. 



South. 



1 QCZfi T„1,t 1 



I A QAA 









AAA 



.Tnlv 30 



+2040 





A, n ~ • Q 





9400 



A, n „ Ol 



i a oo 









1 /too 



Sept 16 



. + 400 





Oct. 1 





-1180 



Oct. 9 



• + 800 





Oct. 19 





-2560 



Oct. 31 



(- 380) 





From these we derive, by taking the differences in dates of the posi- 

 tive extremes, 



29 days, 22 days, 26 days. 23 days, 22 days. Mean, 24-4 days ; 

 while doing the same with the negative extremes, we obtain 



18 days, 27 days, 26 days, 18 days. Mean, 22-25 days. 

 6. The third of these three periods extends from the beginning of 

 May to the end of August 1862. Treating this in the same manner, we 

 obtain : — 



Hemispherical excess. 



Date. 



North. 



South. 



1862, May 9 



+ 600 





May 22 





-1160 





+2960 









-2600 



June 29 



+1880 





July 16 





- 800 



July 26 



+2400 





Aug. 14 





- 200 



Aug. 23 



+460 





Taking, as before, the distances between the positive extremes, we 

 obtain 



25 days, 26 days, 27 days, 28 days. Mean, 26-5 days ; 

 while fr£m the negative extremes we obtain 



24 days, 31 days, 29 days. Mean, 28*0 days. 

 From the whole three periods we obtain, as the most probable mean 

 value, 25*2 days. 



7. We do not profess to have discovered the cause of these oscillations, 

 but we would nevertheless suggest that the observational facts here 

 brought to light may perhaps be connected with two other observational 

 facts, the one of which was first brought to light by Carrington, and the 

 other by ourselves. 



