390 



Results of Makerstoun Observations, 1844. 



The probable error of an observation of the vertical component from the monthly mean for the 



,. , ■„ ii i. , ■ r December and February about 0-00008 "1 ,. ^, , , 



coiTesponding hour will theretore be in < , H T Iv 000010 1 wiiole component. 



These values are estimated from a consideration of the results for the declination, p. 352. 



The mean diftcrence of an observation is least at lO*" a.ji., being 11-6 micrometer divisions, when the 

 mean for the whole year is considered. The probable error of an observation fi-om the monthly mean for the 

 corresponding hour, without refei-ence to which month, will be less than 0-00008 of the whole vertical 

 component. 



Magnetic Dip. 



The following results are deduced from the variations of tlie horizontal and vertical components of mag- 

 netic force by means of the formula 



A sin 2 ^ / A Y A X \ 

 0-0002909 V~Y X~j 



A Y A X 



where -^^ and -^^ are the variations respectively of the vertical and liorizontal components, as given in the 

 Y X 



previous Tables, 6 is the magnetic dip, assumed to be 71° 18', A ^ is the variation of dip, as given in the follow- 

 ing Tables, and 0-0002909 is the value of 1' in parts of radius. 



Sccukir Change of Magnetic Dip. — This deduced from the mean secular changes for the two components, 

 pages 356 and 374, = — 6'-l. It is very probable that this change is considerably too gi'eat, and therefore 

 tliat the secular change for one or both components is also in error, or is due to instrumental causes. 



Annual Period of Magnetic Dip. — The following monthly means are obtained from the means for the two 

 components, pages 356 and 374, where the secular change is eliminated : — 



A 6 



From these quantities the magnetic dip appears to be a maximum about the beginning of April and the 

 beginning of November ; it is a principal minimum about the beginning of July, and a secondary minimum 

 about January. This result differs somewhat from the result obtained from the observations for 1843 (p. 248) ; 

 this is obviously due to the diiference in the results for the vertical component in the two years. 



TABLE LI^'. — Variations of Magnetic Dip with reference to the Moon's Age, Declination, and Dis- 

 tance, as deduced from Tables XXIII. and XXXIX. 



