282 POOR 



An attempt was made to correct Elements II so as to satisfy 

 the three appearances, but without full success ; the resulting 

 residuals being too large. To satisfy the equations represent- 

 ing the appearance in 1903, /i would have to be increased : such 

 an increase, however, would throw large discordances in the 

 normal places representing the first appearance in 1889. In 

 other words, the mean motion of the comet during the interval 

 1 896-1 903 seemed to be greater than the mean motion during 

 the interval 1896- 1889. The two appearances of 1889 ^"^ 

 1896 could be brought into accord ; so could the appearances of 

 1896 and 1903, but no value of the mean motion would at 

 once satisfy all three appearances. It was not possible to 

 satisfy the three appearances by any simple variation of Ele- 

 ments II. 



This seemed to indicate an error in the computed perturba- 

 tions or in some neglected perturbation. The perturbations of 

 the four planets. Earth, Mars, Jupiter and Saturn had been 

 carefully checked and the results agree extremely well with 

 those obtained entirely independently by Neugebauer and re- 

 ferred to in 10. From the result of the comparison there given 

 it would seem as though these perturbations were accurately 

 computed. 



The perturbations of Uranus were far too small to be of any 

 account. It remained then to test the effect of Venus upon the 

 comet. A computation of these perturbations for a few dates 

 showed that they were quite as large and fully as important as 

 those of either the Earth or Mars. Another fact which led me 

 to attribute these discordances to the action of Venus, is that 

 Venus was in widely different parts of her orbit at the times the 

 comet passed perihelion in 1889 and 1896, and consequently 

 the acceleration or retardation of the comet would be radically 

 different in 1896-97 from what it was in 1889-90. While the 

 comet was passing over the arc of its orbit near perihelion in 

 1889-90, the action of Venus was such as to retard it and to 

 decrease its mean motion ; while the comet was passing over a 

 corresponding arc of its orbit in 1896-97, the action of Venus 

 was such as to accelerate it and to increase its mean motion. 



