in different Latitudes. 13 



tKe dip to be 3" 1 i' 42", which j^ives X' or the magnetic latitude 

 of Toil) pendii equal to 1° .'35' 36". By calculating it froin its 

 geographic position, we obtain 1'^ 28' 55". 



Another example: At Quito in Peru, the magnetic latilude com- 

 puted according to the geographic position is A'= G°33' 10'' 

 Doubling it we get the dip . . . . i = 13 6 20 



Huml>oldt found it bv observation .. «=13 21 54 



The observations of La Peroiise and Lacaille near t!ie mag- 

 netic .efiuator, in the Atlantic Ocean and the Indian ISea, being 

 rcduccil in the same manner, present us with results wiiich agree 

 in a similar way with the formula. Unfortunately these simple 

 laws do not extend to the opposite side of the globe, which is 

 aflfected by the inflexions of the magnetic equator. If we at- 

 tempt to apply the exact formula to some one of the southern 

 islands of the South Sea, — to Otaheite, for example, where Cap- 

 tain Cook so often observed, — we find the southern dips far toe 

 great; and on the contrary, in countries situated in North Ame- 

 rica, nearly in the same longitude, the computed dips are much 

 too small. These deviations result necessarily from the inflexion, 

 which in this part of the globe brings the magnetic ec[uator 

 towards the south pole ; and they give a striking confirmation 

 of it. 



To agree with these phcenomena, it is therefore necessary to 

 suppose some perturbating local cause to exist near the archipe- 

 latai of the South Sea, sucli as a particular cciitre of magnetic 

 forces, which influences ihem in tiiis hemis])here and modifies 

 the central action. Indeed, by adopting this supposition it 

 makes all tlie results agree : and they only retjuire a very weak 

 force in the secondary centre, which derives almost the whole of 

 its energy from its proximity. But before we try to define and 

 measure ir, we must consider attentively the alterations that the 

 variation of the needle and the Intensity of the magnetic forces 

 experience in diti'erent latitudes. For these phsenomena, being 

 also the results of the magnetic action of the earth, must be taken, 

 into consideration, if we woidd account for them completely. 



To understand the dips, we began by seeking the series of 

 places where they are nothiiig. In the same way, to discuss the 

 jdiieuouuMia of tlie variations, we must hegin by finding the points 

 on the glolje where they also are nothing. The series of these 

 points form what are called linai of no variation*. These 

 lines do not follow the direction of the geographic meridians on 

 the globe, but are very oblique to them ; and they present us 



* 'I he luxcssary elements for this discussion were rurnislicil hy M. de 

 Iluiiihiildt. 



with 



