MAGNETISM. 



65 



employing three cylinders of iron, the 

 one being solid, and the other two hol- 

 low, but all equal in surface, that the 

 deviation of the compass needle, occa- 

 sioned by the attraction of soft iron, 

 depends on the extent of surface of the 

 iron, and is wholly independent of the 

 mass ; excepting a certain thickness, 

 Amounting to about two-tenths of an 

 inch, which is requisite for the full de- 

 velopment of its attractive energy *. It 

 may be remarked, by the way, that the 

 circumstance of the effective power be- 

 ing limited to the surfaces of bodies, or 

 nearly so, is another striking instance 

 of the analogy which subsists between 

 the magnetic and the electric agencies. 

 These inductive results of observation 

 are all in strict conformity with the theo- 

 retical deductions of Poisson already 

 adverted to, 163. 



Introducing into the general forr 

 mula this new variable quantity, 

 namely, the diameter, or radius, of the 

 sphere of iron, which we shall express 

 by r : it becomes 



tan. ^ ""- 2* cos-*- 

 Ac? 3 



(256.) These rules and formulae are 

 capable of being applied in another 

 manner ; for, instead of conceiving the 

 imaginary sphere to surround the iron 

 ball, we may imagine a similar sphere 

 concentric with the point of suspension 

 of the needle ; and it will then be obvious 

 that the centre of the ball will have the 

 same relative position in the latter sphere, 

 as the pivot of the compass has with re- 

 spect to the former ; so that the refer- 

 ence may be made indifferently to either: 

 and when the mass of iron is irregular, 

 which is the more usual case, it will be 

 more convenient to refer the common 

 centre of attraction of the iron to an 

 imaginary sphere circumscribing the 

 compass. 



(257.) It must be observed, however, 

 that in every instrument a limit exists 

 within which the above law ceases to 

 obtain. This limit arises from the influ- 

 ence which the inductive power of the 

 needle may exert upon the iron pre- 

 sented to it ; lor we have already seen 

 that the consequence of this induction 

 is attraction of the adjacent pole of the 

 magnet, whichever pole that may hap- 

 pen to be. Hence it follows that when 

 the compass is brought so near to the 



iron, as to act upon it by induction, the 

 laws above determined are superseded 

 by those dependent on this latter cause, 

 and are therefore no longer applicable. 

 In all the experiments made by Mr. 

 Barlow, care was taken that the dis- 

 tances should be such as to be entirely 

 exempt from this disturbing cause. 



(258.) Having established the law of 

 action on the compass, as far as regards 

 masses of iron of regular geometric 

 forms, the next object was to determine 

 whether the same law obtains with 

 masses of irregular shapes. This would 

 evidently not be the case, if the popular 

 notion were true, that the poles of apiece 

 of iron, under the influence of terres- 

 trial induction, reside exclusively at the 

 opposite extremities of the mass ; where- 

 as if the entire action admits of being 

 referred to one common centre of at- 

 traction, in the same manner as the com- 

 bined effect of the gravitation of all the 

 particles of a body of irregular figure 

 may be considered as directed on a sin- 

 gle point, known by the name of the 

 centre of gravity, it is reasonable to 

 expect that the same laws are common 

 to both. Experiments tried, with this 

 view, upon a twenty-four pounder, 

 showed the existence of a plane of 

 neutrality in the most irregularly- shaped 

 masses of iron, and completely esta- 

 blished the identity of the operation of 

 the attractive and repulsive forces in all 

 cases, whether the iron was presented 

 in isolated masses, or dispersed in every 

 variety of situation throughout the ship. 



(259.) The actual amount of deviation 

 produced in the ship's compass by its 

 local attraction, will, of course, be dif- 

 ferent in different vessels. With an 

 easterly or westerly course, it has been 

 observed in these latitudes to vary from 

 five to twelve or fourteen degrees : it is 

 of greater amount as the ship is in 

 higher latitudes ; and diminishes, with- 

 out however vanishing, at the equator ; 

 and again increases as we approach the 

 south pole. Mr. Barlow, iu a paper 

 lately published in the ' Philosophical 

 Transactions*,' gives the following 

 table of the deviation observed in dif- 

 ferent ships, on the best authorities, 

 from which a general idea may be 

 formed of the extent of error that may 

 thus arise, and also of its average 

 amount. 



Philosophical Transactions for 1819, p, 129. 



For 1831, p. 217. 



