6 Comparison with Taylor (1998)

The rms's, and means are given in arcseconds in Table 6 where S.E.M. is the standard error of the mean, $N_{\rm u}$is the number of observations used and $N_{\rm r}$ is the number of observations rejected. 11 outliers with residuals $\gt 0\hbox{$.\!\!^{\prime\prime}$}15$ were rejected out of 1028 measures.

  

Table 6: Comparison with the numerical integration of Taylor (1998)

In deriving these figures no change was made to the initial conditions of Taylor (1998). These values are comparable to the photographic observations of Veiga & Vieira Martins (1995a) but smaller than the CCD observations of Veiga & Vieira Martins (1995b). The significance of these differences should not be stressed because Veiga and Vieira Martins compared their observations to Laskar & Jacobson (1987) where we compared with Taylor (1998).