1886.] Sun-spot Areas and Diurnal Declination-ranges. 229 



closely together that we cannot disentangle them completely by onr 

 limited series of observations. Now it is probable that magnetic dis- 

 turbances would limit themselves in great measure to the especially 

 large solar Inequalities, so that if we could find some method of 

 treating not merely the larger but all the Inequalities, we might 

 probably rid ourselves to a considerable extent of the influence of 

 disturbance. But by our method we have the means of doing this. 

 We possess for each element, for each period altogether over 100 series, 

 representing Inequalities extending from —52 to +52 of our 

 notation. 



Furthermore, we have the same series of 24 years common to Toronto 

 declination, Kew temperature, and Prague declination, and it is with 

 this common series that we have made a comparison as follows. The 

 Kew temperature Inequalities have virtually only one maximum and 

 one minimum, and we have selected all those in which it is possible to 

 ascertain accurately the position of the maximum, that is to say, all 

 those which are according to type. Now let the Toronto and Prague 

 declination Inequalities be set in all cases so as to start from the 

 maximum of the corresponding Kew temperature Inequality, using of 

 course for this purpose not the whole 36 years of Prague observa- 

 tions, but only 24 of these. We are thus comparing 24 years of 

 simultaneous declination records at Toronto and at Prague, the setting 

 being in each case from the maximum of the corresponding Kew tem- 

 perature record for the same 24 years. 



In this comparison all the Inequalities, great and small, may be 

 imagined as made use of, and the influence of disturbance eliminated 

 at least to a great extent. 



10. The results of this process are exhibited in the following table, 

 and they may be at once compared with those given in Tables II and 

 IV. For the purpose of this comparison we have transferred the 

 starting points of the modified Inequalities to the solar minimum, so 

 as to make them comparable with, those of the previous tables. We 

 can easily make the change from the knowledge derived from our 

 previous paper that the Kew temperature maximum is about 2 days 

 before the solar maximum. 



The Toronto declination Inequality for 24 days is not greatly 

 altered by the modified process. 



In the Prague declination Inequality for 24 days the modification 

 produced causes the two maxima to be more clearly separated from 

 one another. 



In both of these Inequalities as modified, the great maximum is not 

 long after the solar maximum. 



If we turn next to the Inequalities around 26 days, we find that for 

 Toronto the subsidiary maximum of Table 1Y becomes when modified 

 the predominant one, and the prominent maximum of Table IV the 



VOL. XL. p 



