X 



able to compare the actual and calculated deviations of the compass 

 from the north for all azimuths of the ship. He soon found that 

 almost the whole deviation of the compass was accounted for by the 

 permanent magnetism, and that the residual part followed nearly the 

 quadrantal law. He thence deduced a simple rule by which the 

 compasses could be practically corrected to a first approximation. 

 Putting the ship's length (1) north and south, and (2) east and west, 

 he showed how to place two permanent magnets near the compass so 

 that it indicated true magnetic north in each position. Placing next 

 the ship's length north-east and south-west, the effect of the quad- 

 rantal deviation became prominent, and this he corrected by a mass 

 of soft iron, whose own induced magnetism, when properly placed, 

 counterbalanced that of the ship. These corrections being disturbed 

 when the ship heeled over, another magnet was added. On applying 

 this method to the " Rainbow," and trying the compass with the 

 ship's head in different positions through the circumference, it was 

 sensibly perfect: the deviation, which at first had exceeded 50°, 

 sometimes on one side and sometimes on the other, was at once re- 

 duced to half a degree. The great development of iron-built ships 

 soon rendered some modification in these corrections necessary ; im- 

 provements were made by their author, and other mathematicians 

 also made a special study of the deviations of the compass. Mr. 

 Airy wrote several other papers in connexion with this subject, such 

 as those in 1840, 1856, 1860, and 1862, and in 1865 he delivered a 

 course of three lectures at the School of Naval Architecture and 

 Marine Engineering at South Kensington. The principle that the 

 compass ought be corrected by magnets or otherwise has not been 

 universally received ; it was contended that it was better to use a 

 table of errors. The Astronomer Royal maintained that the former 

 course was the proper one, while Mr. Archibald Smith has been the 

 champion of the latter. The question has been much discussed, but 

 cannot be entered into here. 



Mr. Airy formed one of an important Commission for the restora- 

 tion of the standards of weights and measures which had been injured 

 by the fire at the House of Commons. Contrary to the opinion preva- 

 lent in France, the Commission recommended that the standard measure 

 should be defined by the length of a certain rod preserved in some place 

 of safety, and not by any natural standard, such as the length of the 

 seconds pendulum. Contrary, also, to the method adopted by Bessel 

 for the Prussian standard, the yard is defined by the distance between 

 two points marked on the bar and not by the length of the bar. The 

 history of standards is given by Mr. Airy in a long paper in the 

 ' Phil. Trans.' for 1857. 



About 1841 Mr. Airy turned his attention to the theory of tides. 

 He wrote several papers on this subject, discussing separately the 



