78 



Staff-Commander E. W. Creak. [Mar. 15, 



The constants for soft iron provide a means of predicting probable 

 changes of deviation on change of magnetic latitude for certain 

 vessels of the following classes, and others of similar construction : — 



1. Iron armour-plated. 



2. Iron cased with wood. 



3. Iron troop ships. 



4. Steel and iron ships cased with wood. 



5. Composite-built vessels. 



6. Wooden ships with iron beams and vertical bulkbeads. 



These vessels were all in a state of magnetic stability previous to 

 the observations which have been discussed, and their compasses have 

 had the semicircular deviation reduced to small values, or corrected in 

 England by permanent bar magnets. 



This correction may be considered as the introduction of a permanent 

 magnetic force acting independently, and in opposition to the magnetic 

 forces of the ship proceeding from hard iron. 



It is now proposed to consider the effects of a change of magnetic 

 latitude on the component parts of the deviation. 



Semicircular Deviation. 



On semicircular deviation from fore and aft forces time has but 

 little effect, and the greater part of it is due to permanent magnetism 

 in hard iron, which may be reduced to zero for all latitudes by a 

 permanent magnet. 



A second but small part of the semicircular deviation proceeds from 

 sub-permanent magnetism in hard iron. It is subject to alterations 

 slowly by time, from concussion, and from the ship remaining in a 

 constant position with respect to the magnetic meridian for several 

 days, and is more intensely affected by a combination of the two 

 latter causes. 



Deviations from sub-permanent magnetism which have been tem- 

 porarily altered in value as described, return slowly to their original 

 value on removal of the inducing cause. 



The principal cause of change in the semicircular deviation on 

 change of magnetic latitude in corrected compasses, arises from 

 vertical induction in soft iron which changes directly as the tangent 

 of the dip. 



In standard compasses judiciously placed with regard to surround- 

 ing iron, this element of change is small, and similar in value for 

 similar classes of ships. 



With very few exceptions nearly the whole of the semicircular 

 deviation from transverse forces is due to permanent magnetism in 

 hard iron subject to the same laws as that proceeding from fore and aft 

 forces. 



In the exceptional cases alluded to, there is a small part due to 



