340 



A STUDY OK COKRKLATIO>fS AMO\(i TERRESTRIAL TEMPERATURES. 



studying this table it must be noted that the given epochs are not those deriv-ed indi- 

 vidually from the observations in each case, but are the results of the general formulae 

 which best represent all the observations. Consequently, the difference between the 

 sunspot epochs and the temperature epochs as derived are constant in each of the 

 respective phases of maxima and minima. 



Table IV. 



Comparison of Epochs of Temperature and Sim-sj)ots. 



It will be seen that, in the general mean of all the epochs, the temperature epoch 

 follows the spot epoch, the comparisons of each phase being 



Maximum temperature — minimum .sunspots — 0.33 



Minimum temperature — maximum .sunspots + 0.65 



Mean of all the comparisons +0.16 



The difference between the comparisons of the two phases arises from the fact 

 that, l)y the method adopted, the intervals between the maxima and minima of tem- 

 peratures necessarily come out equal, while those between the maxima and minima 

 of sunspots are unequal. 



The general conclusion is that the fluctuations of temperature follow very closely 

 those of the sunspots, according to the law first clearly brought out by Koppen. The 

 slight lagging of 0^.16, or two months, is too small to be regarded as the result of any- 

 thing but accidental deviations, being less than the probable error of its amount. 



Very remarkable is the fact that the actual fluctuation is less than half that 

 found by Koppen. In order to show whether, when treated by the more rigorous 

 method, the deviations of temperature used by him would give a different result from 

 mine, we have only to find the general result of his data taken separately. This we 

 do by deriving the mean values of x and y fi'om his data alone, the individual results 

 of which are found in the last nine lines of the table. These give 



.?■ = 0°.13 



,/ = - 0°.05 



Accidentally, the principal coefficient of the fluctuation is practically the same 

 whether derived from his observations or from the others. 



