DO H. I. JENSEN. 



W. J. S. and Sir Norman Lockyer, and others) may be sum- 

 marised briefly as follows : — (1) Violent outbursts of prom- 

 inences (which have their maximum and minimum at the 

 same time as sunspots) are immediately followed by mag- 

 netic storms, auroral displays, and often weather changes 

 in the earth's atmosphere. 



(2) Prominences and sunspots have a periodicity of 

 approximately eleven years, with a superimposed cycle of 

 35 years. This 35 year cycle corresponds exactly with the 

 Bruckner weather cycle of 35 years in accordance with 

 which long-period variations in climatic conditions take 

 place. 



II. Sunspots and Temperature. — (a) The curves of 

 Koppen and Nordmann 1 showing mean temperature for 

 tropical districts seem to indicate that the mean annual 

 temperature is highest at sunspot minimum and lowest at 

 maximum, the temperature curve corresponding to the 

 inverted sunspot curve. The main conclusion to be drawn 

 from M. Charles Nordmann's paper is that " The mean 

 terrestrial temperature follows a period sensibly equal to 

 that of solar spots ; the effect of spots is to diminish the 

 temperature, i.e. the curve which represents the variations 

 of temperature is parallel to the inverse curve of sunspot 

 frequencies." 



Mr. Alex. B. McDowall, f. r. Met. soc, has arrived at the 

 opposite conclusion by a careful study of the meteorological 

 statistics of last century. He finds that England receives 

 more heat when the spots are many than where they are 

 few, and he has dealt with his subject from many stand- 

 points. He finds that in England there are more frosty 

 days at, and a few years after, a sunspot minimum than is 

 normal, and in years of spot maximum the number of frost 



1 Nature, No. 1685, Vol. lxv., p. 353, and Comptes Rendus No. 18. 



