COMET OF 1889-1896-1903 233 



This is exactly the effect necessary to reconcile the discordances 

 noted in the three appearances. Consequently, I undertook to 

 determine by computation the effect of the perturbations of 

 Venus upon the comet. 



Owing to the rapid motion of Venus about the sun the per- 

 turbations change sign frequently and the interval of computa- 

 tion should be very short. The perturbations were computed 

 for the entire interval 1889 to 1903 with a 20-day interval from 

 January 1889 to April 1890, from April 1896 to May 1897 

 and from May 1903 to December 1903; these three periods 

 being the times of closest approach of the comet to the planet. 

 During the remaining times the comet was at such a distance 

 from Venus that the direct action of that planet upon the comet 

 became very small and during these periods the interval of com- 

 putation was made 40 days. 



The computations were carried out in a manner entirely sim- 

 ilar to those of 9 : the adopted mass of Venus being 



390,000 



The values of the integrated perturbations for the intervals be- 

 tween the various appearances are given in the following table ; 

 the date of osculation being, 1896 October 23.5 : 



Perturbations by Venus. 



1S96-1889. 1 896- 1 903. 



AZu — 28.623 -(-79.160 



With these additional perturbations Elements II were carried 

 back to 1889 and forward to 1903 and a normal place formed 

 representing each of the three appearances. From these nor- 



