88 The Rev. II. Lloyd on the Results of Observations made at the 



The mean results for the two triennial periods, together with the means of 

 all, are graphically represented in Plate III. figs. 3, 4, 5 ; the scale being 0-2 inch 

 to one minute of arc. To facilitate comparisons hereafter instituted, t\ie positive 

 ordinates correspond to westerly deviations. The correspondence of the course 

 of the variation during the two periods is as close as could be expected, the 

 dilFerence consisting chiefly in the epoch of the westerly maximum. 



16. The following are the laws of the changes, as deduced from the mean 

 results : 



I. From the beginning of April to the beginning of July, the north end of 

 the magnet moves to the west ; the maximum of westerly declination takes place 

 about the 8th of July, but the epoch varies considerably in diiferent years. 



II. Diuring the remainder of the year (i. e. from the beginning of July to the 

 beginning of April), the north end of the magnet moves to the east; the move- 

 ment, however, is very slow diuing the first three months of the year. 



ni. The range of the westerly movement is 2-7 minutes ; and that of the 

 easterly 8-7 minutes. Thus, at the end of twelve months, the north end of the 

 magnet has advanced to the east, by about 6-0 minutes, as has been already 

 shown. 



17. The annual variation of the declination was discovered by Cassini in 

 1786.* It appeared from the observations of Cassini, that the north pole of the 

 magnet moved to the east from the vernal equinox to the summer solstice ; 

 and that, during the remaining nine months of the year, it moved to the west. 

 The westerly movement, during the nine months, preponderated over the 

 easterly, which took place during three ; and thus the westerly declination was 



* The determination of the annual variation is much more difficult than that of the diurnal, 

 both on account of the much smaller frequency of the period, and the difficulty of preserving the 

 instrument in the same \inchanged condition during the much longer time, or of determining and 

 allowing for its changes when they do occur. Accordingly, although the annual period may be 

 traced in the observations of Gilpin, and is decidedly displayed in those of Bowditch, it has evaded 

 the researches of recent observers. There is but a faint indication of its existence in the Gottingen 

 observations, which were made at the hours of 8 A. M. and 1 P. M. ; and Professor Gadss and 

 Dr. GoLDscHMiDT find, in their analysis of these observations, no " important fluctviation dependent 

 on season." A similar negative result is deduced by Dr. Lamont from the Munich observations, 

 wliich were made twelve times in the day. 



