IN THE IEON AND AEMOUR-PLATED SHIP NORTHUMBERLAND. 
495 
tion, but to under correct it, leaving about 5° uncorrected. This was the object sought 
in the application of magnets to the ‘ Northumberland ’ ; and the success with which it 
was effected, and the certainty of the process, may be seen by the following comparison 
of the deviations on the 28th February and the 2nd March. 
Standard compass . 
A 
B 
C 
D 
E 
o 
o 
o 
("February 28, 1867 
+0 01 
-36 13 
+ 7 09 
+7 19 
+6 59 
(.March 2 
-0 14 
- 4 55 
+ 0 14 
+7 13 
+1 15 
J February 28 
+0 41 
- 5 20 
+40 41 
+6 28 
+2 40 
t March 2 
-0 18 
- 4 15 
- 1 04 
+6 42 
+1 44 
In general no attempt is made to correct the quadrantal deviation, that deviation 
and the residual semicircular deviation being the subject of tabular correction. I may 
observe that in some cases Mr. Airy’s second or tentative mode of correcting, viz. by 
one or more fore and aft and transverse magnets, is adopted, but only when special cir- 
cumstances prevent the first method being applicable. In general by the process first 
described, all* that is necessary or desirable in the way of mechanical correction can be 
effected. 
In the middle of March 1867, the ‘ Northumberland,’ as before stated, was placed in a 
dry dock at Devonport, and the magnets were removed ; her head in dock was S. 84° E. 
magnetic. Such a position would, from all former experience, and especially in a newly 
launched ship, be expected to increase considerably the force to starboard, or the value 
of +6, but without much alteration in the fore and aft force, or the value of 93, in 
those compasses in which the 93 and (5 were caused by the general magnetism of the 
hull. We have continuous observations to the end of 1867 strictly comparable as 
regards the Standard compass ; but as regards the other three compasses allowance must 
be made for the introduction between August and December 1867 of five iron beams* 
to extend the light poop deck before the poop compass. 
The changes in the Standard compass while the ship lay in dock with her head 
S. 84° E. are shown in the following Table. 
Swinging at anchors 
[1867]. 
r January 26th 
99. 
-•637 
S. 
+•108 
1 February 28th 
-•637 
+ 110 
f June 26th 
-•576 
+•277 
Head S. 84° E. magnetic from middle of March 
j August 29th 
-•569 
+•256 
[ December 10th 
-•593 
+•329 
* The effect of the introduction of the five new poop-deck beams when the ship’s head was S. 84° E. mag- 
netic would be to increase (5 and ® and diminish A. The exact value of the altered (£ cannot he computed 
without knowing the altered values of ® and A ; and these cannot be ascertained till the ship is again swung. 
I think, however, it is certain that the change in ® cannot exceed ‘017 (or 1° in !D), and the change in A cannot 
exceed -020, and I have accordingly inserted in the General Table, under date 10th December 1 867, the values 
of 33 and (5 computed as well with the original values of 3) and A, as with values allowing these amounts. It 
may he safely assumed that the true values lie between the two as limits. 
MDCCCLXVIII. 3 Z 
