6 Comparison with other data and discussion

Table 2 gives the differences in the system of the ICRS with VLBI radio positions taken from the list of Ma, in the sense radio minus this work (CL). The residuals obtained are consistent with the precision of our optical data, and show that none of these CERS can be ruled out as possible link objects (source 1148-671 was not included in Table 2). Although the number of comparison objects is small, we nonetheless plotted the residuals as a function of the coordinates to search for possible systematic effects; no clear evidence in this sense was found.

The optical data presented in Table 1 were also compared with, mostly unpublished, optical positions with respect to the IRS Catalogue we have obtained for these objects. The same plate material was used for both determinations. This comparison yielded the following statistics: $\overline{\Delta\alpha \cos \delta}$ = -0.06$^{\prime\prime}$, $\sigma(\overline{\Delta\alpha \cos \delta})$=0.17$\hbox{$^{\prime\prime}$}$; $\overline{\Delta\delta} = 0.10\hbox{$^{\prime\prime}$}$,$\sigma(\overline{\Delta\delta}) = 0.23\hbox{$^{\prime\prime}$}$; n = 24. Plots of the (Hipp-IRS) residuals as a function of the coordinates show clear systematic effects similar to those reported in Paper IV. This is not surprising considering that our Hipparcos referred optical positions are linked to the ICRS, therefore a comparison of these positions with IRS referred positions brings out the deviations of the IRS(FK5) system from the ICRS.