r 



Magnetic Declination. 



347 



TABLE XV. — Mean difference of a Single Observation from the Monthly Mean at the corresponding 

 hour, on each day of the Moon's Age, Position in Altitude, and Distance from the Earth. 



Variation of the Mean Differenees with respect to the Moon's Age. — The first portion of the previous tahle 

 contains the hourly means of the differences for each day of the moon's age. The result in this case, also, is 

 the same as that obtauied in the discussion of the diurnal ranges, namely, a maximum about the opposition, 

 and a minimum near the conjunction ; there is the appearance of a secondary maximum at the conjunction, 

 a minimum occurrmg a few days before it and a few days after it. The following are means of groups : — 



14 days to 16 days, Full Moon, 2'- 18 



17 ... 20 ... 2'-04 



21 ... 24 ... l'-45 



25 ... 28 ... l'-22 



29 days to 1 day, New Moon, l'-47 



2 ... 5 ... l'-35 



6 ... 9 ... l'-82 



10 ... 13 ... 2'-25 



So that the average difference of an observation from the monthly mean for the corresponding hour is 

 about twice as great at opposition as at conjunction. 



Variation of the Mean Differences with respect to the Moon's Declination. — From the two preceding results 

 (for the annual variation and moon's age) it follows that maxima of the differences should occur when the moon 

 is near the equator, and minima when it is farthest north and farthest south ; that this is the fact, may be seen 

 from the second portion of Table XV., and also from the following means of groups : — 



27 days to 

 2 ... 

 6 ... 

 9 ... 



1 day. Moon farthest North, 

 5 ... 

 8 ... 

 12 - 



l'-46 

 l'-97 

 l'-80 

 l'-64 



13 days to 15 days, Moon farthest South, l'-53 



16 ... 19 ... l'-58 



20 ... 22 ... l'-69 



23 ... 26 ... 2-08 



From these groups the principal minimum occurs when the moon is farthest north, and the two maxima 

 occur when the moon is north of the equator. 



Diurnal Variation of the Mean Differences. — Tables XVI. and XVII. contain the mean differences for 

 each hour in each month, each quarter, and the year. From the means for each hour for the year, it appears 

 that the average disturbance is a maximum about Si^ p.m. ; the minimum occurs perhaps about noon, but the value 

 of the average disturbance oscillates within small limits between 3 a.m. and 5 p.m. The two consecutive hours 

 that have the highest mean value are 8 p.m. and 9 p.m., and the two that have the lowest mean value are noon 

 and 1 P.M. Had observations been made at the even Giittingen hours only, the maximum would have been ex- 

 hibited at 11 p.m. Makerstoun mean time. 



