3 o SECTION PHYSICS. 



Mr. Archibald Smith, the theory of the octantal errors was worked 

 out and fully described in a paper communicated to the transactions 

 of the Royal Society of England, and published about the year 1861, 

 the title of the paper being Effect produced in the deviatiojts of the 

 compass by the length and arrangement of the compass needles ; and a 

 new mode of correcting the Quadrantal Deviation. This paper, by 

 Messrs. Smith and Evans, contains the mathematical investigation of 

 the theory of these octantal errors, and a very curious and interesting 

 result is arrived at, according to which these errors are very much 

 less with such a compass as the Admiralty standard, in which there arc 

 four needles instead of only two or one, as in many common compasses, 

 and with the needles arranged in a particular way, the two ends nearest 

 the north being 3o Q apart, and those on each side of them being, I 

 think, 30^ more, so that the four ends were placed at intervals ot 30? 

 from one another. With this particular disposition of the needles it 

 was proved that the octantal error was theoretically very much less, 

 and, by Captain Evans's experiments on the " Great Eastern," it 

 seems that, when an Admiralty compass was substituted for the great 

 compass with one needle which had been in the binnacle before, the 

 error was corrected nearly enough at all points by the correctors as 

 actually applied. That very dangerous kind of error then may be 

 considered as being rectifiable, and thus, although we cannot have 

 infinitely small needles, we can at all events have a correction which 

 will be perfect, or practically perfect, for the ship in all positions by 

 the correctors to which I have referred. 



The quadrantal error has to be corrected by placing masses of soft 

 iron on each side of the compass, in such masses and in such positions 

 as shall, by the magnetism induced in them, counteract the effect of 

 the magnetism induced in the ship. The natural history of ships of 

 all classes, which has been worked out most admirably by the 

 Admiralty Compass Department, in respect of magnetism, worked by 

 the fullest development of the mathematical theory of Archibald 

 Smith, with the most thoroughly business-like perfection of detail, 

 and with great scientific accuracy of observation, and which has been 

 going on for thirty years, has given us knowledge which is of 

 inestimable value on this subject. Among other things it shows us 

 the amount of quadrantal error we have to meet with in different 



