DEVIATIONS OF THE COMPASS. 
435 
the sine of the inclination to the cosine of the inclination, or proportional to 
the tangent of the inclination. Such a bar, placed in a ship vvitli its upper 
end at the same level as the compass, will, as the ship is swung, produce no 
deviation when north or south of the compass, will produce a maximum 
easterly deviation when east, and a maximum westerly deviation when west 
of it, these bearings being always those by the disturbed compass. 
If the bar of soft iron be horizontal, and in an east and west position, it is 
at right angles to.the line of force, and will produce no effect; i! it be hori¬ 
zontal and in any other position, its south end will attract the north end of 
the compass-needle. If such a bar be placed in a ship frt the same level as 
the needle, and directed towards it, then in four positions ns the ship is 
viz. when the bar is E., W., N. or S. of the needle, it will produce 
do deviation; in the first two positions, because it is not magnetic; in the 
tat two, because, though magnetic, it attracts the north end of the needle in 
the direction in which it already points. Such a bar will produce a maxi¬ 
mum easterly deviation when N.E. or 8.W. of the cornpass-ueedle, a maxi¬ 
mum westerly deviation when S.E. or N.W. of it. 
1 he attraction of such a horizontal bar on the needle when a change m 
the geographical position takes place, is proportional to the total magnetic 
intensity multiplied by the cosine of the inclination ; the directive force oft ie 
earth on the needle, which it has to overcome in order to produce a deviation, 
w P ro portional to the same quantity. The deviation produced is conse- 
quenily independent of the force and of the inclination, and is the same in 
ail magnetic latitudes. . 
t *''' r (> f hard iron magnetized acts on the needle, when there is noc ange 
a magnetic latitude, precisely as a vertical bar of soft iron ; that is to say, 
,f re aro two opposite positions in which it causes no deviation, an wo 
at distances from them of 90° by the disturbed compass, at which 
T h* maximum deviations, at the one easterly, at the other westerly. 
e . re L ,8 ’ however, an important difference between the effect o an g 
; ,; <1 bar of hard iron and a vertical bar of soft iron when there is » 8 
cha? n . et,C latitude - The attraction of the bar of hard iron ^maua. 
to ov 8 l ° n a - cha "S° magnetic latitude; the directive forcewhdutbas 
,n . ordep to produce a deviation is proportional to 
w„5 u c '“tenmty imdtipiied by the cosine of the inclination, or m oti 
i 2e( i to l * ie horizontal magnetic intensity. Hence the effect of a 0 
horizon! * ard ! ron iu causiug deviation is inversely proportioi« 
Imii V' agnet ' c 'alensity at the place. „.,nd the 
iujiiunt ° ases ' vl, ' fi h wo have considered, as the ship is swung r ’ 
m”«ts eastc p rl y and of westerly deviations arc exactly equal; lj,lt “ ,rag 
ptedomii ^ ° N °^ ' ron >na y easily he imagined in which one o 
TIhn s and l hat to any extent. . . i u ars 0 f 
•oft i ro i 'J' e “ u Ppo»e two similar and similarly placed horizon 
f C" t r 1 1 ,i,e «”r ss ' »" e N - ;v ,d s ; yitaXS 
these two h?*’ 8 ' -,‘ e °^ M!r *y in 8 and W. with its L. end N. . . n „ 
will remain 18 W .! * Produce an easterly deviation, which, as the s | 
rbi ain sensibly constant * • c 
h°n ofUmV’ 6 the sha P e position of the iron in a ship, if f[^consd- 
j er *hle di8ta, W ° » xtrome qualities we have mentioned, ia “ i( - t h e 
Aviation 8 a ' Ce Trmu the needle compared to the size of the needle,, i ne 
* hi pCinf 8mall ‘ if 'vo suppose (as is done throughout) tha^tne 
taj n of u le „ 0D 1 . evcn heel, the following formulae for expressing e 
^l-atiorm “Sf® in ^muth may be deduced from P^son ^^ 
ri8> See Col. Sabine’s ‘Contributions to Terrestrial Magnet* 
