26 



NATURE 



[March 3, 19 10 



By a combination of three sets of correcting magnets, 

 two horizontal and one vertical, Kelvin obtained complete 

 neutralisation of the disturbing effect of the ship's per- 

 manent magnetism, both as respects. semicircular error in 

 change of the ship's course and heeling error as she heels 

 or rolls. From time to time, if the condition of perfect 

 compensation is to be maintained, the position of these 

 various correctors has to be altered, because of changes 

 which take place in the so-called permanent magnetism of 

 the ship. .The navigator has always to be on the look-out 

 for the gradual development of errors from this cause, 

 however perfectly the first adjustment has been carried 

 out. 



We have ne.xt to consider the effects of induced 

 magnetism. The most important of these arise from the 

 fact that the ship is a long body of magnetisable material 

 turning in a horizontal plane, and therefore subject to the 

 inductive influence of the horizontal component of the 

 earth's magnetic field. Think of what would happen if we 

 >yere to take a pivoted compass needle and place it above 

 or below a bar of soft iron, and slowly turn the bar round 

 in a horizontal plane. We are to think of the bar as 

 having no appreciable magnetic hysteresis, so that in every 

 position it is the induced effect only with which we have 

 to do. What will be the nature of the deviation ? When 

 the bar points north, and again when it points south, there 

 is no deflection of the needle, for though the magnetism of 

 the bar is then at its strongest, the field due to it is in 

 the line with the undisturbed earth field ; also when the 

 bar points east or west there is no deflection, for the bar 

 then takes up no magnetism ; but between these points, 

 namely, when the bar is pointing N.E., S.E., S.W., or 

 N.W., the deflection is at its maximum. So in a ship's 

 compass this error, due to the purely transient magnetism 

 induced by the horizontal component of the earth's field, 

 has its maximum on these four courses, once in each quad- 

 rant, and for that reason it is called the quadrantal error. 



It is due, as we have seen, to the ship's being a long 

 body, extending fore and aft, and it is corrected by 

 balancing this excess of fore and aft iron by other iron, 

 placed quite near the compass and on either side of it. 

 The two balls which you see on the side of the Kelvin 

 binnacle are the correctors for quadrantal error. They are 

 adjusted, in the first place, by selecting a suitable size of 

 ball, and then placing them nearer to or further from the 

 compass until, on swinging the ship, the quadrantal error 

 disappears. The possibility of correcting the quadrantal 

 error in this way had been pointed out by Airy as earlv 

 as 1840; but with the old form of compass card and 

 needles it could not be done, because of the excessive length 

 and large magnetic moment of the needles. To apply the 

 method to a compass of the old pattern would have needed 

 globes of impracticable size, not a few inches in diameter 

 as -these are, but weighing tons. Kelvin, with his short 

 needles on a light card, made it possible to carry out the 

 process, and so gave the world, for the first time, a com- 

 pass that would point truly to the magnetic north, not- 

 withstanding all the perturbations due to permanent and 

 induced magnetism in the iron of the ship. 

 ^ One more of these disturbing causes remains to be men- 

 tioned.^ The vertical component of the earth's field induces 

 magnetism as well as the horizontal component, and gives 

 rise to an additional error of two kinds, namely, a further 

 .semicircular error and a further heeling error. These are 

 distinct from the semicircular error and heeling error due 

 to permanent magnetism, and the right way to correct 

 them is to fix a bar of soft iron in a vertical position ' 

 near the binnacle, so that the magnetism induced on it will 

 act as a counter-balance. This is the Flinders bnr. so 

 called because its use was pointed out by Captain Flinders 

 a.s early as 1801. It has generally to be fixed in front of 

 the binnacle, and in Kelvin's compass it is made in several 

 separate lengths of soft iron, which can be put together 

 to make up a bar giving any necessary amount of correct- 

 ing effect. 



The main function of the Flinders bar is to correct the 

 semicircular error due to induced vertical magnetism. So 

 far as the heeling error is concerned it also .helps, but in 

 practice it is found convenient to correct a part of the 



i That is to say, vertical when the ship is on even keel or perpendicular to 

 the deck. 



NO. 2105, VOL. 83] 



heeling error due to induced magnetism by means of the 

 same, kind of permanent magnet correctors as I have i 

 already described in speaking of the heeling error due to j 

 permanent magnetism, namely, vertical magnet bars placed/ 

 in. a can in the binnacle directly under the centre of thej 

 compass card. The number and height of these bars has; 

 therefore to be altered from time to time, as the ship moves] 

 to regions where the vertical force is different. When the! 

 heeling error is fully corrected we escape one cause of thej 

 unsteadiness which a compass shows when a ship rolls, 

 for we escape, the magnetic cause of oscillation, namely, 

 the alternate magnetic pull to port and starboard; but a 

 purely dynamical cause of unsteadiness necessarily remains, 

 arising from the fact that the point of suspension , of a 

 compass card must be placed some way from the centre 

 of gravity to hold the card level against the dipping action 

 of the earth's magnetic field. Consequently, every, roll to 

 either side, applies a mechanical couple tending to set up 

 oscillation, and if the period of the roll were the same, or 

 nearly the same, as the period of oscillation of the card, 

 the disturbance would become so great as to make steer- 

 ing by compass. impossible. It was to secure steadiness in 

 this sense that Kelvin strove to give his compass card a 

 long period of oscillation, recognising that the right way 

 to obtain steadiness was to make the period much longer 

 than the period of the slowest rolling motion liable to 

 occur in a ship, at the same time keeping the. friction as 

 small as possible. The problem of securing a steady, 

 frictionless compass was a problem where, as in the inven- 

 tion of the mirror galvanometer, his genius for practical 

 dynamics guided him to the right solution. In the case 

 of the compass it was rendered difficult by the fatt thai 

 other conditions, apparently antagonistic, had at the same 

 time to be satisfied in order that the correction of magnetic 

 errors might be completely carried out. 



The evolution of the Kelvin compass, in its maii. 

 features, took about five years ; but a longer task la> 

 before the inventor in overcoming the professional, con- 

 servatism of sailors, the objections of the so-called practical 

 man, active hostility in some quarters, and the passive 

 resistance of oflicial inertia. Gradually the compass came 

 to be used in merchant vessels of the best appointed class. 

 Enlightened navigators such as Captain Lecky, the author 

 of the well-known " Wrinkles," became its enthusiastic 

 advocates. Foreign admiralties took it up, and in our own 

 service individual oflicers were quick to see its merits. 

 Captain Fisher, now Admiral of the Fleet Ix)rd Fisher, was 

 warm in its praise after observing its behaviour in ships 

 under his command, first in the Northampton in rough 

 weather and afterwards in the Inflexible during the firing 

 of heavy guns in the bombardment of .Alexandria. That 

 was in 1882 ; but it was not until November, 1889, that the 

 superintendent of the Compass Department of the 

 Admiralty was in a position to inform Lord Kelvin that 

 his lo-inch compass was to be adopted as the standard 

 compass for the Navy. This was twelve years after the 

 date of his patent, and more than eleven years after he 

 had laid the invention formally before the First Lord. The 

 way of the inventor, like that of the transgressor, may 

 still be hard, but I trust it is not so hard now as it was 

 then. One does not care to dwell on the spectacle of a 

 Kelvin spending his strength in disheartening effort as the 

 sea beats against a cliff. It is painful to read the corre- 

 spondence and discussions of these weary years. One does 

 it with increased admiration of the infinite patience which 

 at last secured to us the benefits of his practical genius. 



The use of the Kelvin compass may now be said to be 

 universal, except that in the Navv a modified form, due 

 to Captain Chetwynd, with a card immersed in liquid, is- 

 taking the place of the Kelvin dry card in the newer ships 

 as being steadier still under gun-fire. The system of 

 correction remains substantially unchanged, and the com- 

 pass continues to embody the same mechanical features as 

 formed the basis of Kelvin's invention. 



In the navigational sounding machine we have another 

 invention of first-rate importance, second only to the com- 

 pass in practical value to sailors, and remarkable for its 

 extreme simplicity. It was his cable-laying experience that 

 first led Kelvin to take an interest in deep-sea sounding. 

 The process, as then carried out, was a laborious one. 

 The line was a rope an inch and a half in circumference. 



