336 cosmos. 



lines. Never before, I believe, had any government fitted out 

 a naval expedition for an object whose attainment promised 

 such advantages to practical navigation, while, at the same 

 time, it deserved to be regarded as peculiarly scientific and 

 physico-mathematical. 



As no phenomenon can be thoroughly investigated by a 

 careful observer, without being considered in its relation to 

 other phenomena, Halley, on his return from his voyage, haz- 

 arded the conjecture that the northern light was of a magnet- 

 ic origin. I have remarked, in the general picture of nature, 

 that Faraday's brilliant discovery (the evolution of light by 

 magnetic force) has raised this hypothesis, enounced as early 

 as in the year 1714, to empirical certainty. 



But if the laws of terrestrial magnetism are to be thorough 

 ly investigated that is to say, if they are to be sought in the 

 great cycle of the periodic movement in space of the three va- 

 rieties of magnetic curves, it is by no means sufficient that 

 the diurnal regular or disturbed course of the needle should 

 be observed at the magnetic stations which, since 1828, have 

 begun to cover a considerable portion of the earth's surface, 

 both in northern and southern latitudes ;* but four times in 

 every century an expedition of three ships should be sent out, 

 to examine, as nearly as possible at the same time, the state 

 of the magnetism of the Earth, so far as it can be investiga- 

 ied in those parts which are covered by the ocean. The mag- 

 netic equator, or the curve at which the inclination is null, 

 must not merely be inferred from the geographical position of 

 its nodes (the intersections with the geographical equator), but 

 the course of the ship should be made continually to vary ac- 

 cording to the observations of inclination, so as never to leave 

 the track of the magnetic equator for the time being. Land 

 expeditions should be combined with these voyages, in order, 

 where masses of land can not be entirely traversed, to determ- 

 ine at what points of the coast-line the magnetic curves (es- 

 pecially those having no variation) enter. Special attention 

 might also, perhaps, be deservedly directed to the movement 

 and gradual changes in the oval configuration and almost con- 

 centric curves of variation of the two isolated closed systems 

 in Eastern Asia, and in the South Pacific in the meridian of 

 the Marquesas Group.t Since the memorable Antarctic ex- 

 pedition of Sir James Clark Ross (1839-1843), fitted out 

 with admirable instruments, has thrown so much fight over 

 the polar regions of the southern hemisphere, and has determ- 

 * Cosmos, vol. i., p. 190-192. t Cosmos, vol. i., p. 182. 



