IN THE IRON SHIPS OP THE ROYAL NAVY. 
355 
a separate and permanent pillar or stand : it is by this superior elevation that the strong 
magnetic power of the u'on beams and adjoining topsides are correspondingly lessened. 
As every piece of iron not composing a part of, and hammered in the fabrication of 
the hull, — such as the rudder, funnel, boilers and machinery, tanks, cooking galleys, 
fastenings of deck houses, &c., — are all of a magnetic character differing from the hull 
of a ship, their proximity should be avoided, and, so far as possible, the compass should 
be placed so that they may act as correctors of the general magnetism of the hull. 
A compass placed out of the middle line of the deck is affected by the nearest top- 
side, and its deviations must necessarily be much increased if that topside has the domi- 
nant polarity, as in ships built East or West. 
Experience has proved that the practical value of mast or elevated compasses has in 
some cases been overrated; they are, in fact, affected by the ship’s magnetism to an 
amount depending on their elevation and the direction of the ship’s build ; thus in ships 
built North or South, but especially the latter — the compass being on the mizen mast — 
the deviations will be large comparatively. In ships built East or West the deviations 
will be comparatively small, from the topside, which would affect a deck compass, being 
more directly under the mast-compass ; they may therefore be useful in the latter cases, 
and valueless in a ship built head to the South. The wear and tear on the pivots and 
agate caps of mast-compasses, from the increased motion due to their elevation, require 
constant attention when they are employed. 
3. On various Sources of Error affecting a Compass placed under favourable conditions. 
Errors arising from changes of geographic position, as also incidental causes of error 
due to anomalous rather than general conditions, have been brought under review in 
the general progress of this Eeport. There is, however, one source of compass-error — 
that arising from the heeling of the ship — which has not been alluded to, as the ship 
in all the points hitherto reviewed is assumed to be on an even keel. 
The few experiments made in ships of the Eoyal Navy will be found in Table I., and 
they tend to prove, as also does the test of experience, that when the original compass 
deviations are small, the errors from heeling are generally small in proportion ; and 
conversely, that exaggerated errors from heeling are the consequence of exaggerated 
errors while on an even keel. Ample elevation from the deck, in order to raise the 
compass above the level of the topsides and adjacent deck beams, is one of the chief 
conditions for reducing this source of error. 
The action of the topside nearest to the compass when the ship is heeled is well 
marked in the examples of the ships Bloodhound and Sharpshooter, the former with 
-1-B, or head built to South, and the latter with — B, or head built to North; in the 
Bloodhound the north end of the needle is dravvm to leeward, in the Sharpshooter to 
windward. In the iron sailing-brig Recruit, with +B, the effect is the same as in the 
Bloodhound. 
From the diagrams illustrative of the polarity of an iron ship’s topsides, the action of 
3 B 2 
