SECT, xxx.] SECULAE VAKIATION OF THE SYSTEM. 345 



Eastern Siberia, and another in the Pacific Ocean, nearly in 

 the meridian of the Pitcairn Islands and the Marquesas. 



Complicated as the magnetic phenomena are, they are 

 rendered still more so by the secular, periodic, and transient 

 variations to which they are liable. The foci of magnetic 

 intensity, and the whole system of the magnetic curves, are 

 moving along the two hemispheres in different directions : 

 those in the northern hemisphere are going from west to 

 east, and those in the southern from east to west ; and as 

 the foci of greatest intensity move with different velocities, 

 the forms, as well as the position, of the curves are slowly 

 and ceaselessly changing ; so that, in the course of years, 

 the whole magnetic system is altered. Colonel Sabine, so 

 profoundly versed in this branch of science, and which has 

 assumed a new character from his observations and laborious 

 researches, has shown that the weaker magnetic focus has 

 moved through fifty degrees of longitude in 250 years. 



The declination is subject to periodic variations, de- 

 pending on the declination of the moon and her distance 

 from the earth, also on an annual disturbance arising from 

 the motion of the sun in the ecliptic ; but the horary varia- 

 tions are the most important, depending on the diurnal 

 revolution of the sun, in consequence of which the declina- 

 tion needle makes two deviations eastward, and two west- 

 ward, from its mean place, in the course of twenty-four 

 hours. Throughout the middle latitudes of the northern 

 hemisphere, the north end of the needle has a mean motion 

 from east to west from eight in the morning till half-past 

 one ; it then returns to the east till evening, when it makes 

 another excursion to the west, and returns to its original 

 position at eight in the morning. The angular extent of 

 the excursion is greater in the day than in the night, in 

 summer than in winter. It decreases from the middle lati- 

 tudes in Europe, where it is 13 or 14 minutes, to the 

 equator, where it is only 3 or 4 ; but the horary variations 

 there are so regular, that the hour of the day may be known 



