Ofi Refraction. 391 



but if he should employ his alchemical powers in converting the 

 gold of others into brass, it would not be surprising if it should 

 again be reported, as it was of old of the same Midas, that 

 " Auriculas asi?ii Mida rex habet ;" and that the punishment 

 should be remembered even longer than the alchemy. 



Supposing for a moment that Mr. Ivory is right in condemning 

 the series which has been employed for the computation of re- 

 fraction, and supposing that his observations on the series " do 

 not bear at all upon the table of refractions published in the 

 Nautical Almanac for 1822 ; it is still scarcely credible that any 

 mathematician, besides Mr. Ivory, will be hardy enough to assert 

 that " the formula and the table must both be considered as 

 entirely empirical." Phil. Mag. Sept. p. 167. 



Can it possibly be believed that the most skilful empiric, even 

 with the advantage of all the refined methods of empiricism that 

 were cultivated by Lagrange, could ever have deduced, from a 

 table of refractions only, a formula so complicated as this ; 



s ' s« s^ s* 



The values of B and C only are determined empirically ; but was 

 Eulcr's theory of the moon entirely empirical, because some of 

 the coefficients were determined from observation only ? 



Again, the only improvement, which the author of the table 

 considers as of practical importance, is the introduction of an 

 accurate and convenient mode of correcting for the heights of 

 the barometer and thermometer. Will Mr. Ivory call this ap- 

 proximate correction w any degree empirical ? He may assert 

 that this computation is independent of the former : but it is, at 

 least, an essential part of the table : a table on which even 

 Mr. Ivory has condescended to bestow the conditional commen- 

 dation, that " if it has a real and solid foundation, it must be 

 allowed that no greater or more honourable testimony can be 

 given in favour of — the French Astronomers .'" 



The French Astronomers really stand so little in need of this 

 honourable testimony, that they will probably forgive the author 

 of the table if he persists in withholding it; and its claim to the 

 confidence of the public may safely be allowed to rest upon the 



