10 



LUNAR EFFECT 



The constant in Bessel's formula comes out zero, and hence it is inferred that 

 the moon has no specific action in deflecting the magnet by a constant quantity. 

 The coefficient of the first term of the formula is small, and it is from the second 

 term that the distinctive features, of the double-crested curve result. These results 

 are all represented by curves. 



Both the east and west deflections are well marked, those occurring when the 

 moon is east of the meridian being greater than those when west. 



It is not at all necessary to take in the third or higher terms. The progression 

 of the hourly values is systematic, and the agreement between the computed and 

 observed values is deemed satisfactory. The following diagram represents the 

 curve resulting from the above equation, the observed values being indicated by 

 dots. 



Lunar-Diurnal Variation. 



0b. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24>>- 

 U. C. L- C. U. C. 



From 21,644 observations at Philadelphia, from 1840 to 1845. 



The principal western maximum occurs 6 minutes after the lower culmination 

 of the moon, and amounts to 0'.23. The secondary maximum occurs 14 minutes 

 after the upper culmination, and amounts to 0'.18. The principal minimum occurs 

 at 6 h - 17 m after the lower culmination, the easterly deflection being 0'.22. The 

 secondary minimum at G d - 03 m - after the upper culmination, with a deflection of 

 0'.19. The greatest range is 27", and the secondary 22". The epochs of the 

 maxima and minima are found from the formula to be at a mean 10 minutes after 

 culmination. The probable error of a single computed value of the lunar declina- 

 tion is + 1".32. The Toronto observations gave + 1".37 from more than twice the 

 number of observations, so that the Philadelphia results are worthy of every 

 confidence. 



At Toronto, from the second investigation, embracing about 44,000 observations, 

 the western and eastern deflections balanced, giving for the range 38".3. The 



