TERRESTRIAL MAGNETISM. 173 



of the Yiti Islands). At the commencement of the present 

 century,, I determined the point where the magnetic equator 

 crosses the chain of the Andes, in the interior of the new 

 continent, between Quito and Lima, at an elevation of 

 nearly 12000 English feet above the level of the sea, in 

 7 V S. lat., and 48 40' W. long, from Paris. To the 

 west of this point, throughout almost the whole breadth of 

 the Pacific, the line without dip, or magnetic equator, 

 though slowly approaching the geographical equator, conti- 

 nues in the southern hemisphere ; in the vicinity of the 

 Indian archipelago it passes into the northern hemisphere, 

 just touches the southern point of Asia, and enters the con- 

 tinent of Africa near the strait of Bab-el-Mandeb, which is 

 the point of its greatest distance from the geographic 

 equator. Thence, traversing the terra incognita of the 

 interior of Africa in a south-westerly direction, it re-enters 

 the southern hemisphere in the Gulf of Guinea, and, main- 

 taining a south-westerly course across the Atlantic, reaches 

 the Brazilian coast near Os Ilheos, north of Porto Seguro, 

 in 15 S. lat. From thence to the elevated plateau of the 

 Cordilleras, where I observed the inclination at a spot be- 

 ween the silver mines of Micuipampa and the ancient seat 

 of the Incas at Caxamarca, the line traverses a part of 

 South America as unknown to us as the interior of Africa. 



Recent observations, collected and discussed by Sabine( 156 ), 

 have taught us that, in the interval between 1825 and 1837, 

 the node jjear the Island of St. Thomas moved 4 from the 

 east towards the west. It would be extremely important to 

 learn whether the opposite node near the Gilbert Islands in 

 the Pacific, has undergone a corresponding westerly move- 

 ment, and advanced an equal amount towards the meridian 



