10 On the Laws of Terrestrial Magnetism 



The agreement of these results is certainly very remarkable. 

 We appear to he well authorized in concluding that the mag- 

 netic equator is really a great circle of the terrestrial sphere, in- 

 clined to the equator in an angle of about 12°, and having its 

 western node situated at about 1 15" 34' of west longitude from 

 Paris* ; that is, in the South Sea a few degrees from the island 

 of Gallego, at nine hundred leagues from the coast of Peru. This 

 places the opposite node in 295'' 34' of west longitude. Such 

 indeed has hitherto been the general opinion. But what is ex- 

 traordinary, these elements are wholly defective, in all that part 

 of the South Sea, situated below the western node between 115° 

 and 270^ of longitude, which comprehends nearly a whole hemi- 

 sphere of sea. Indeed, in discussing some observations made 

 with the utmost care by Captain Cook and Mr. Bayly, in two 

 different vessels which sailed together in the South Sea, in 1777, 

 I find that they have both of them met with the magnetic equa- 

 tor in 158= 50' 9" of west longitude, and in 3° 13' 40" of south 

 latitude ; whereas, by prolonging the great circle, that our first 

 observations have given us, this equator ought then to have been 

 found in north latitude 8° 36' 30". This shows us, therefore, 

 that the magnetic equator, after having crossed the terrestrial 

 equator, at about 115° of west longitude, descends again into 

 the southern part of the globe : and as the observations of Mr. 

 Bayly which are confirmed by those of Mr. Dalrymple, show 

 likewise that this line is without dip in about 7° of north lati- 

 tude in the Chinese sea, at 256° of west longitude, we must con- 

 clude, that between this longitude and that of 158° 50' deter-^ 

 mined by the observations of Captain Cook, the magnetic 

 equator and the terrestrial equator have at least one more inter- 

 section, independently of the eastern node, situated in the Indian 

 Sea, at about 295° of longitude, and dependent on the circular 

 part. There are therefore in all at least three nodes, and per- 

 h&ps four, if the magnetic equator, near its western node, rise 

 a little towards the north, before it descends again into the 

 south, near the archipelago of the Society Islands. Fig. 2, re- 

 presents the series of these inflexions, of which we shall presently 

 find some striking confirmations in the effects we shall derive. 



In examining the magnetic dips observed on both sides the 

 line just traced, we find that they increase in proportion as they 

 recede from it. I*" we confine ourselves to consider that half of 

 the globe where the magnetic equator appears to be exactly cir- 

 cular, and which comprehends Europe, Africa, the Atlantic 

 Ocean, and the eastern coasts of the two Americas, we find that 



* Palis is 9' 21" of time to the eastward of Greeinvich, or 2* 20' 15" of 

 space; consequently this node is in 113° 14' of west longitude from the 

 Rojal Observatory of Greenwich. 



the 



