xix 



It is impossible to consider in detail the numerous additions which 

 Mr. Airy has made to our knowledge at various times. It is necessary 

 to pass over with merely a mention such an important memoir as that 

 in which he discusses theoretically the stresses in t\e interior of a 

 rectangular beam. In another paper he mentions a method of 

 correcting the chromatic dispersion of the atmosphere in observing 

 transits of Venus. He has also written several papers on the com- 

 parison of Earth currents and magnetic disturbances, on the diurnal 

 and annual inequalities of terrestrial magnetism, and on some lunar 

 magnetic inequalities. We must hasten on to his last work. 



In 1874 the Astronomer Royal brought before the notice of the 

 Astronomical Society a new mode of treating the lunar theory. After 

 giving a rapid survey of the methods hitherto employed, he remarks 

 that the nature of the steps has compelled the investigators to decide 

 the succession of their terms, not by numerical magnitude, but by 

 algebraical order, and that this has produced great inequality of con- 

 vergence. The mental labour cannot be alleviated by an amanuensis, 

 and he quotes a remark of M. Plana, " Quelquefois ces calculs me font 

 presque perdre -la tete." He proposes, as a new method, to begin 

 with Delaunay's final numerical expressions for the longitude, latitude, 

 and parallax with a symbolical term attached to every number for 

 contingent correction. These corrections are so small that it is 

 sufficient to retain only their first power. The expressions are then 

 substituted in the equations of motion with the time for independent 

 variable, and the result of the substitution is a great number of equa- 

 tions for determining the numerical values of a great many small 

 quantities. In this theory the orders of the terms are numerical and 

 equally accurate throughout ; the details are so easy, that a great part 

 can be intrusted to a mere computer. Though he was then seventy- 

 three years of age, he had aleady begun the work. He says that, 

 though it is sufficiently possible that he may not be able to complete 

 it, he desires to have it in such a state that a successor may be able 

 to take it up. For this reason in each of the following years he gives 

 further details as to the theory, and describes how far he had ad- 

 vanced in the approximations. Finally, in 1886 the numerical lunar 

 theory was given to the world in the form in which it was left by the 

 author ; a wonderful monument of what a man can do at the age of 

 eighty-five. He explains in the preface how the work was delayed 

 by the heavy pressure of business, not only in the ordinary conduct 

 of the Observatory, but also in completing the calculations for the 

 transit of Venus in 1874, and in preparing for that in 1882. He 

 then remarks on some serious discordances which remain to be ac- 

 counted for ; "I cannot conjecture," he adds, "whether I may be 

 able to examine sufficiently into this matter." He never was able. 

 A long catalogue of the honours and titles which Sir Gr. Airy 



c 2 



