ON SUNSPOTS AND TERRESTRIAL PHENOMENA. 447 



connection by a strictly statistical inquiry. We are, further, not to discuss 

 whether the connection, if it exists, is sufficiently large to affect materially 

 the character of our summers or of our winters. The scientific interest 

 is the same whether the influence be large or small ; and though the dif- 

 ference in temperature between years of maximum and years of minimum 

 sunspots may only be the hundredth part of a degree, we must be satis- 

 fied with the result as soon as it has been sufficiently well established. 



Few people only are aware of the many careful and unprejudiced 

 investigations that have been made on the subject ; and there is a great 

 deal of misconception, even among those who take a special interest in 

 it, as to the uniform drift of these investigations. There can no longer 

 be a doubt that during about four sunspot periods (1810-1860) a most 

 remarkable similarity exists between the curves representing sunspot 

 frequency and the curves of nearly every meteorological phenomenon 

 which is related to temperature. This is not, in my opinion, a matter 

 open to discussion ; it is a fact. But it is equally certain that during the 

 thirty or forty years previously to that time no such relationship exists, 

 and that since 1860 the connection has again, in some cases, become less 

 distinct. The question which arises now is this : Does the absence of 

 any apparent connection at the beginning of this, and especially at the 

 end of last century, which, as far as we can judge, may be partially re- 

 peated at the present time, render it probable that the relationship which 

 held good for nearly, if not quite, half a century, is only accidental ; or is 

 it more reasonable to suppose that there is a true connection, but that 

 other causes are at work sufficiently strong altogether to hide the regu- 

 larity for a succession of years at a time ? This is a point on which 

 everybody will follow the dictates of his own common sense, and on 

 which, therefore, we must not expect any unanimity of opinion. In 

 giving an account of the work which has been done, and of the questions 

 which are pending, I shall not attempt to mention, much less to sum- 

 marise, every paper that has been written on the subject. There is hardly 

 one on which we possess more exhaustive summaries, and to them I must 

 refer the reader who wishes a more detailed account. 1 



Reduction of Sunspot Measurements. 



In order to compare terrestrial phenomena with sunspot activity, we 

 must first obtain a numerical measure of that activity. Messrs. De La Rue, 

 Stewart, and Loewy have in their researches measnred the spotted area 

 of the sun ; and recently the ' Solar Physics Committee ' has deduced, as far 

 as possible, this area from the measurements of Schwabe and Carrington. 

 The spotted area of the sun, no doubt, is at present the most scientific 

 measure of solar activity ; but we do not possess the necessary data for 

 its determination except for a very limited period. For the older obser- 

 vations, then, at any rate, we must be satisfied with a simpler measure, 

 and it has become customary to adopt Wolf's so-called sunspot numbers. 

 The sunspot number for a certain day, according to Wolf, is k (f + 10g), 



1 F. G. Hahn : TJeber die Bczielixvngen der Sonnenflecltervpcriode zu meteorologischen 



Erscheinungen. (Leipzig, 1877.) Hermann Fritz: Die Beziehiengen der Sonnen- 



Jleeken zu den magnctischen und meteorologischen, Erscheimmgen der Erde (Natitrli. 



Verh.d.lwll. Maatsoh. Wits. Haarlem, 1878.) Siegmund Giinther: Der Einjtuss 



der Himmebkorper auf Witterungsverhaltnisse, Niirnberg, 1884. 



