274 STAFF COMMANDER EVANS AND MR. A. SMITH ON THE MAGNETIC 
hard iron in the ship acting in the fore-and-aft line, and multiplied by the reciprocal 
of the earth’s horizontal force, and also by the factor ^ . The last part may be considered 
as itself consisting of two parts ; one, of the subpermanent magnetism induced while the 
ship was building by the vertical component of the earth’s force, and which probably 
bears some relation to the transient magnetism induced by the same vertical component ; 
another, of the subpermanent magnetism induced while the ship was building by the 
headward component of the earth’s horizontal force. 
C theoretically consists of similar parts acting towards the sides of the ship ; but as 
the iron may in general be considered as symmetrically arranged on each side of the 
compass, the value of C is probably, in all cases when the ship is upright and the com- 
pass is amidships, to be attributed to subpermanent magnetism induced while the ship 
was building by the transverse component of the earth’s horizontal force. The part of 
B consisting of transient induced magnetism varies as the tangent of the dip. The other 
part of B and C vary inversely as the earth’s horizontal force. As regards changes 
which take place after launching, without a change of geographical position, there are 
differences between the several parts of B and C which require notice. 
When the ship is launched, notwithstanding that her head is no longer kept in one 
fixed direction, the forces which cause the two first-mentioned parts of B still act in 
precisely the same direction as before, and these two parts probably undergo little 
change. 
With the third part of B and the whole of C the case is very different. The forces 
which cause these parts cease to act in the same direction as at first. If the vessel is 
allowed to swing at her anchors, or is under sail or steam, she will probably on an 
average be nearly as much on one point as on another ; or, which would come to nearly 
the same thing, if she is lying in a tideway she may be alternately for six hours in one 
direction and for six hours in the opposite direction. A great portion of the C and of 
that part of the B which arose from horizontal force thus become dispelled. 
The symmetry which gives C its character ceases the moment the ship heels. An 
addition is then made to C proportional to the angle of heel, and this addition consists 
in fact of two parts, corresponding to the two parts of B which, as we have seen, do not 
exist in the original C, viz. a part consisting of transient magnetism induced by the 
vertical force, and a part consisting of subpermanent magnetism induced by the same 
force. These will be more conveniently considered when we come to discuss the heel- 
ing error. 
The semicircular deviation may be put under the form \/B 2 +C 2 sin(^'+a), in which 
v/B 2 -J-C 2 represents the maximum of semicircular deviation, a f tan «= -j the angle to 
the right of the ship’s head of the force causing this deviation; for convenience, these 
two quantities are tabulated in the eleventh and thirteenth columns. 
The terms D sin 2£'-}-E cos 2£' make up what is called the “quadrantal deviation.” 
