T E R 



239 



T E R 



observed by Col. Sabine, that the quantities of magnetism 

 in the two hemispheres are nearly equal ; but as all the 

 four poles lie in one hemisphere of the earth, which would 

 be formed if the latter were cut by the plane of a meridian 

 passing through points whose longitudes are 100 and 280 

 (80 W.), it must follow that the hemisphere which con- 

 tains the poles, that is, the hemisphere which contains 

 America and the Pacific Ocean, must have a greater quan- 

 tity than the other. 



The isodynamic lines present the appearance of double 

 flexures, like those of equal dip ; and in both systems of 

 curves there is a like tendency to form two foci, or centres 

 of greatest attraction ; and the bends gradually become 

 less strongly marked as the lines approach the equatorial 

 regions of the earth, but it is obvious that the lines in the 

 two systems are far from being parallel to one another : 

 in the southern hemisphere the isodynamical line 1 crosses 

 the line of no dip in two places at least ; and each of the 

 dip-lines, bli, re, dd, in No. 1, would pass through several 

 of the intensity-lines in No. 2. Even within the limits of 

 the British Isles the deviations of the two systems of lines 

 from parallelism are very sensible ('Memoir,' by Maj. 

 Sabine, in the Eighth Report of the. British Association) ; 

 and it may be inferred that, at least in the northern hemi- 

 spere, the pole of maximum intensity is quite distinct from 

 the pole of the dip, the distance between them in latitude 

 being probably as much as 20 degrees. Q and R in the 

 cut No. 2 are the presumed places of the two intensity- 

 poles in the northern hemisphere. It is at present quite 

 uncertain whether or not there is a corresponding difference 

 between the poles of maximum intensity and of maximum 

 dip in the southern hemisphere, but the circumstance is 

 probable, from the fact that the highest observed inten- 

 sities in both are equal at places (New York and Van Die- 

 men's Land) where the dips are also equal, and where they 

 want 20 degrees of being the greatest. 



Professor Hansteen, in his treatise on the magnetism of 

 the earth (1819 , has shown, from a comparison of the ob- 

 served places of the four poles of the dip at different times, ' 

 that each of them has a slow movement about the axis of 

 the earth. Not much dependence can be placed on the 

 computed periods of the revolutions, but M. Hansteen 

 assigns for that of the North American pole 1890 years, 

 anil for that of the Siberian pole 8UO years. 



The existence of two magnetic poles in each hemisphere 

 is thus evident, and Gauss of Gottingen observes that 

 there must also be a third point between each pair, which 

 possesses the character of both, and therefore is a true pole. 

 nil T/irnry of Terrestrial Magnetism, translated in 

 Taylors ' Scientific Memoirs,' pt. vi.) This is indeed ob- 

 vious : for if a dipping-needle were carried from one pole 

 towards the other, it would begin to deviate from the ver- 

 tical direction towards the pole- it had quitted : and, 

 coming near the other, it would be found to deviate from 

 the vertical towards the pole which it was approaching ; 

 and, as these deviations are in opposite directions, there 

 must be an intermediate point at which the needle would 

 assume a vertical position. 



Almost as soon as a few observations on the phenomena 

 of terrestrial magnetism were collected, Dr. Halley (1701) 

 propounded a theory in order to account for them. He 

 conceived that the earth itself might be a shell, containing 

 v ithin it a globe which revolved with it about the same 

 centre of gravity and the same axis ; the outer globe, or 

 shell, lieinar supposed to perform its rotation in twenty-four 

 hours, and the other in a time rather greater or less. Each 

 globe was supposed to have a magnetic axis passing through 

 the common centre, but the two axes were supposed to 

 he inclined to each other and to that of the diurnal rota- 

 tion ; and consequently there were supposed to be, in all, 

 four magnetic poles. 



The deviation of these magnetic axes from that of the 

 earth's diurnal rotation was supposed to be the cause of 

 the ceneral variation (declination) of the compass-needle, 

 and the siuw deviation of the magnetic axes IVom each 

 other was sup|x>sed to be the cause of that continual varia- 

 tion ot'the declination which is observed at every place on 

 the earth's surface. The theory is highly ingenious, and 

 I' the epicycles, by which, in the infancy of 

 astronomy, it was attempted to account for the variations 

 in the movements of the planets: but when observations 

 were multiplied, and the variations of the needle at con- 

 siderable interval* of time were compared together, it was 



found to be incapable of representing the phenomena; 

 and Mayer, of Gottingen, without gaining any advantage, 

 modified the hypothesis by assuming that the centre of 

 the small magnet was placed at a certain distance from that 

 of the earth. 



Subsequently (1805) M. Biot, assuming that there were 

 two points in a supposed magnetic axis of the earth, by 

 one of which the magnetized needle was attracted and by 

 the other repelled, investigated a formula for expressing 

 the dip and variation in terms of an indeterminate dis- 

 tance between those points. On comparing the result 

 obtained by computation from the formula with the ob- 

 served phenomena, he found that the latter were repre- 

 sented with tolerable accuracy when the points of at- 

 traction and repulsion were infinitely near to each other 

 and to the centre of the earth. From the result of the 

 investigation it follows that if a plane, supposed to pass 

 through the centre of the earth perpendicularly to the 

 magnetic axis, were considered as a magnetic equator, the 

 tangent of the dip of the needle would be equal to twice 

 the tangent of the magnetic latitude of the place on the 

 earth's surface ; and a like conclusion had been previously 

 arrived at by Professor Kraft at St. Petersburgh, from such 

 observations as then existed. It is here supposed that the 

 curve of no dip is the circumference of a great circle 01 

 the sphere, and we have seen that this is far from being 

 conformable to observation, yet the rule just mentioned 

 may be advantageously employed when it is required, 

 from any observed dips of small magnitude, to determine 

 the situation of a point on the earth's surface where the 

 dip is zero. The last attempt to account for the pheno- 

 mena of terrestrial magnetism in this manner was made by 

 M. Hansteen (1811), who assumed the existence of two 

 small magnets of unequal strength at certain distances 

 from the centre of the earth. He compared the results, 

 with respect to variation, dip, and intensity, which he ob- 

 tained by computation from that assumption, with the 

 ' values ol those elements observed at different places ; and 

 though, to a certain extent, the agreement was satisfactory, 

 yet in several instances the differences were such as to 

 show that the hypothesis was erroneous or incomplete. 

 Professor Gauss of Gottingen, in his Treatise on the Ge- 

 neral Theory of Terrestrial Magnetism above quoted, has 

 investigated the elements independently of all hypotheses 

 concerning the distribution of the magnetic fluids in the 

 earth, and assuming only that the terrestrial force is the 

 collective action of all the magnetized particles in the 

 earth's mass, he has exhibited the resulting formulae in 

 converging series ; and has given, for ninety-one places on 

 the globe, a table of the values of the declination and in- 

 clination of the needle, and of the intensity of magnetism, 

 computed from his expressions for the horizontal and 

 vertical components of the force, together with the ob- 

 served elements at the same places ; and the srnallness of 

 the differences between these last and the computed ele 

 ments are satisfactory proofs of the correctness of the 

 theory. For a supposed connection between the tempera- 

 ture of the earth's surface and terrestrial magnetism, see 

 ISOTHERMAL LINKS. 



The want of complete success which has hitherto at- 

 tended the different attempts to exhibit the laws of niag- 

 netical phenomena make it evident that the time has not 

 arrived in which that can be done with respect to mag- 

 netism which Newton accomplished with respect to gravi- 

 tation. But though the hypotheses formed, in order to 

 account for the phenomena of terrestrial magnetism, have 

 not brought out formulae which will entirely satisfy the 

 observed elements, it must not be understood that they 

 are therefore without utility; since the approximative 

 rules which have been obtained from them afford the 

 means of computing small differences in the elements with 

 sufficient accuracy to allow observations made at times or 

 in places not very distant from one another to be reduced 

 to what they would have been had they been made at one 

 time or station ; and thus several observations may be 

 made to concur in the determination of a correct mean 

 value of the element. This remark may be considered as 

 applicable to most of the hypotheses which, in the phy- 

 sical sciences, have been proposed for the purpose of ex- 

 hibiting the laws of the phenomena; and it may be fur- 

 ther remarked, that the assumption of an hypothesis, by 

 indicating the fittest place for observing, or the nature of 

 the observations which are requisite for verifying it, i-: 



