( 285 ) 



"lliat i\u' o\HH'\\s (if iiiaxiiiiiiiii itroiiiiiiciicc (lisdirhaiicc in llic liinlici' 

 latitudes are widelv (litlcreiil IVoiii lliosc ii(>ar llie e<|iial()i'. The latter 

 are elosely associated with tlic epochs of iiiaxinui spotted area, tlie 

 foriner occur Itoth N. and S. a( iuleix cuiuii' limes." 



Now it tolK)\\s IVoiu our llieorv, thai pi'oiuiueuces are seen in 

 [)laces where the line of sight touches whirhnii,- parls of surfaces of 

 discontinuity near the Sun's edge; so it is exident that at s[)ot niaxiiua 

 tins will happen most often in the spot zones, and that the most 

 favourable occasions for seeing them in olhei' laliludes fal! at other limes. 



Therefore, although in the curve of promiiuMu-e frequeiu-v the 11- 

 year spot period is easily recognized, yet in se\eral i)oints it deviates 

 from the spot curve. Smaller maxima and minima of frequency are 

 superposed upon the head curve and point to a three-year period. 

 We iind a raticmal explanation tor these minor Ihictuations too in 

 the successive jwsitions assumed l»y the Sun with respect to the Earth. 



l\. TUF, rKKlODlClTY IN THK VAKIATIONS OF MKTEOKOI.OOICAI^ AND 



f;akth-:\ia(tNhti( ■ klemknts. 



1. JJo thi'sc jiliriiomrihi iwiiiiiiw flic //i/iidf/ics/s tlxit tiw Sin/ 

 I'.vhihits (I rari/iiKj activity? 



In the preceding pages we ascribed the inconstancy of the solar 

 [)henomena priju'ipally to the continuous cliange of the point of sight 

 from wliich we look at the Sun. We su[)posed the modifications, 

 produced in the general con(Htion of the body of the Sun itself by 

 radiation and by tiie relative motion of the gaseous layers, to be 

 com[)aratively slight and regular. Our theory had no need of the 

 interference of violent eru|)tions, eonnnotions, periods of increased or 

 deci'eased solar "activity"'; it alloxved us to consider the (piantity of 

 energy emitted by tiie Sun in a unit of time, to l)e almost constant. 



For this reasoii it would now at first sigiit seem more difHicult 

 yet to account for the |»eriodical variations of sexeral terrestrial 

 phenomena which closely follow in their course the frec{ueiu*y (»f 

 s[»ots and prominences. Ihit e\e]i uith the hypothesis of a variable 

 solar output as a starting point, as far as 1 kiu)w, not a single theoi-y 

 has l)een advanced in explanation of the coiniexion between sun-spots 

 and terrestrial i)henomena, so convincing, that it would be a jiity 

 to abandon it. 



Let ns briefly consider what has been atlem|)te(l in that direction. 



Periods of maximum spot frcMjuency are marked by certain commo- 

 tions on Earth and by increased circulation ; the rainfall is greater, 



