44 



Mr. Ellis on the Relation between Magnetic [May 8, 



necessary, in order to estimate progressive change, to treat them 

 in some way that will eliminate the actual annual inequality. This is 

 done, both for declination and horizontal force, by taking the mean of 

 each twelve consecutive monthly values, and again the mean of each 

 two consecutive values, giving a series of annual or smoothed means, 

 the mean epochs of which differ by one month. 



In the forty-third volume of the " Memoirs of the Royal Astro- 

 nomical Society," Dr. Rudolf Wolf, of Zurich, gives a table of relative 

 solar spot values, found by treating actual monthly numbers (for the 

 purpose of removing accidental irregularities), exactly in the same 

 way in which the monthly magnetic numbers were treated. 



The magnetic numbers and Dr. Wolf's sun-spot numbers are thus 

 strictly comparative, and from these numbers curves are drawn which 

 show a remarkable similarity. Not only do the epochs of minimum 

 and maximum accord, but smaller changes are also seen to appear in 

 all three curves. 



The epochs of minimum and maximum being tabulated, it is found 

 that, on the average, the mean magnetic epoch follows the sun-spot 

 epoch by - 27 of a year. By another method of tabulation the differ- 

 ence becomes reduced to 0*10 of a year. Some other interesting 

 details are here added. 



The general circumstance that the diurnal ranges of magnetic 

 elements are subject to an eleven-year .period, similar to that of 

 sun-spots, being considered to be satisfactorily established, it seemed 

 desirable to ascertain whether the more fitful changes of the pheno- 

 mena in any way also correspond. In order to make this comparison, 

 the magnetic numbers, instead of being smoothed, as before described, 

 were now cleared only of the average annual inequality, and compared 

 with the actual monthly sun-spot numbers (not the smoothed num- 

 bers as before). Curves are given, founded on these numbers, and 

 they show some very remarkable correspondences between the more 

 rapid sun-spot and magnetic variations, especially between the years 

 1869 and 1873. It is remarked that the magnetic results appear 

 to be sensibly improved since the instruments were placed (in the 

 year 1864) in an underground apartment in which the diurnal range 

 of temperature is less than 1°. 



Further inquiry seems to point to a variation in the annual in- 

 equalities of magnetic diurnal range. This question is discussed in 

 some detail, and the numbers and diagrams given appear to show that 

 such variation really exists, the inequalities being increased at the 

 time of a sun-spot maximum, and diminished at the time of a sun-spot 

 minimum. 



The following are the general conclusions supposed to be derived 

 from the whole inquiry : 



1. That the diurnal ranges of the magnetic elements of declination 



