ON SUNSFOTS AND TERRESTRIAL PHENOMENA. 



453 



fact would throw grave doubts on the reality of the connection, had it 

 not been pointed out that just at that time the sunspot curve was very 

 irregular. It has been shown by Koeppen that the maximum of tempera- 

 ture in 1779 and the minimum in 1785 occurred when it ought to have 

 done under the supposition of a regular course of sunspot activity. It is 

 useless to speculate how such a state of things could have been brought 

 about, we can only wait the farther development of events. 



Table I. 



Sunspots . 

 Temperature 



Table II. 



Hahn has paid special attention to the summers and winters which are 

 remarkable either by an exceptionally high or an exceptionally low tem- 

 perature, and he has shown that, on the whole, both hot summers and 

 warm winters occur near times of minimum sunspot frequency, while cold 

 winters and summers both occur most frequently directly after a time of 

 great solar activity. But the facts do not show themselves sufficiently 

 regular as to warrant at present any prediction on the general nature of 

 any particular summer or winter. 



The results obtained by Koeppen have been confirmed in many ways. 

 Stone l has shown that the mean temperature at the Cape of Good Hope 

 shows a decided tendency to imitate the curve of sunspot frequency, and in 

 such a way that to a great number of spots corresponds a minimum of 

 temperature. Piazzi Smyth 2 tells us in the same way that the thermo- 

 meters sunk into the rocks of Calton Hill show a decided period of eleven 

 years. The maxima of temperature occurred in 1846, 5 ; 1858, 6 ; and 

 1868. 7. These times fall shortly after the minima of spots, which took 

 place in 1843, 5 ; 1856, ; and 1867, 2. It is to be observed that thermo- 

 meters introduced into the ground are specially fitted for the examination 

 of long period oscillations. The rate at which the amplitude of an oscil- 

 lation decreases with the depth, increases quickly with the length of the 

 period, and the deeper thermometers are introduced into the ground, the 

 more will all short period inequalities be sifted out, and only the longer 

 ones will remain. The amplitude of long oscillations will of course be 

 reduced also, and there are limits, therefore, depending on the sensitive- 



Vrnc. Royal Soc. xix., 389. 



Astir. Obs. Ed. xiv. 



