EARTH'S MAGNETISM TO THE SOLAR AND LUNAR PERIODS. 105 



the same periods as in 1841 ; but the branch ascending to and descending from 

 the period of greatest N. declination is greatest, the periods of minima being 

 nearer the greatest S. declination, namely, about 5 days before it and after it. 

 The curve for 1845 is, however, more irregular after the S. declination maximum 

 than in any part of the other curve. Besides the non-elimination of the effect 

 connected with varying phase and disturbance, there is another possible cause of 

 difference, namely, the varying distance of the moon ; the period of perigee is 

 about two days before the greatest S. declination in 1844, and two days after it 

 in 1845. It should also be remembered that each point in these curves is a mean 

 of only 12 or 13 days; as for the minor irregularities in the positions of the 

 points, it is obvious that, as there are 27 days between the periods of the moon's 

 greatest N. declination, if the full moon occurs on the day of greatest N. declina- 

 tion in one month, it will occur on the second day after the greatest N. declina- 

 tion on the next month, the fourth day on the next, and so on. It will be seen 

 afterwards that this will cause a slight in-egularity. It is on this account that 

 I have projected the cm-ves among the points, giving a preference to the mean 

 positions of each two points. 



15. The similarity of the positions of maxima and minima in these curves, hav- 

 ing the moon's declination for abscissee to the annual cmwe, or that having the 

 sun's dechnation for abscissae, is at once evident ; by talcing the mean of the two 

 lunar curves, however, the cases will be identical, for then the moon's perigee 

 win occur at the time of its greatest S. declination, and its apogee at the time of 

 the gi-eatest N. declination ; this is the case with us for the sun. The resulting 

 means have been projected below the other curves. By comparing the mean 

 curves of No. 4 and No. G, it is at once obvious that the facts are the same for 

 both the sun and moon. I conceive, then, that I am justified in stating that the 

 same relation exists for the moon as for the sun between the variations of the 

 horizontal component of the earth's magnetic intensity, and the variations of de- 

 clination and parallax. 



16. We have, then, a law connected with two periods, namely, distance and 

 declination. To which does it belong, or does it belong to both ? It will take a few 

 years' obsei-vations to determine this for the moon : it may be determined for the 

 sun by observations for the annual period made in the Southern Hemisphere. Is 

 there a maximum at the greatest N. declination, and also at the greatest S. de- 

 clination ; or have changes of declination no effect ? and are the maxima due to 

 the moon's or sun's distance solely ? The supposition that at first sight seems 

 most probable is, that these variations are due to both ; that a maximum occurs 

 at the time of perigee, a minimum at the apogee, a maximum at the gi-eatest N. 

 declination, and a minimum at the greatest S. declination. It may easily be 

 shevra that two regular curves having these arguments, when superposed, would 



