CHAP, i.] THE COMPASS. 529 



only do they not coincide with the geographical meridians, but they 

 must be distinguished also from the magnetic meridians. Among 

 them two are remarkable, namely, the lines in which the declination 

 is zero, which may be regarded as the continuation of each other ; one 

 crosses the American continent from Hudson's Bay to South Carolina, 

 from the mouth of the Amazon to Rio de Janeiro ; the other, less 

 regular, cuts Australia, curves to the west of India, and passes by the 

 Caspian Sea and the Aral Mountains to the White Sea. The two 

 lines of no declination divide the globe into two parts ; that which 

 contains Europe and Africa has all its declinations to the west, while 

 those of the other part are all to the east. There exists in Asia an 

 isolated elliptical portion of a line of no declination which surrounds 

 a space in which the declination is to the west. 



The magnetic equator is the line of points where the needle of the 

 dip circle remains horizontal, and for which consequently the inclina- 

 tion is zero. It does not coincide with the terrestrial equator, which 

 it cuts in two points and touches in a third. The first two points are 

 in the Gulf of Guinea, and in the Pacific Ocean about west longitude 

 175 or 180, and the point of contact is in Polynesia, about 135 

 west longitude from Paris. The isoclinic lines follow pretty nearly 

 the contour of the magnetic equator, thus differing very sensibly from 

 the geographical parallels. 



These systems of lines are not permanent, because the magnetic 

 state of the earth is subject to certain oscillations of which some are 

 periodic, and others variable. The declination, inclination and dy- 

 namic intensity vary continually in each place, and from place to 

 place on the surface of the globe. These variations are partly secular, 

 partly annual, and partly diurnal. For example, at Paris, the value of 

 the magnetic declination, which is now about 18, and is to the west, 

 was nothing in 1663, that is, a little more than two centuries ago; 

 before that it was to the east ; for example, in 1580 it was 11 30' E. 

 Since 1663 it increased continually to the west till 1814, when it 

 attained its maximum. Since then it has been going back. The dip 

 has varied in like manner since the earliest observations. It was 75 

 at Paris in 1671, it is now only 66; this, however, is a much less 

 marked variation than that of the declination. Independent of these 

 variations of long period, terrestrial magnetism is subject to annual 

 ones, which appear to depend on the position of the sun relatively to 



M M 



