An EVN+MERLIN map of 3C287 at 5 GHz is presented by Fanti et al. (1989). They found that the most compact A component - which is the brightest one - is in the centre of the source. The source brightness decreases smoothly along the curved jet, and has a secondary maximum at 15 mas to East and 45 mas to North from the centre. This component marked with C in their paper. The two point spectral index between 18 cm and 6 cm (S ) changes from -0.5 to -1.5 near the brightness minimum, then flattens again up to -0.5 in the C component (Fanti et al. 1989).
The most remarkable difference between the map of Fanti et al. (1989) and our results is that we cannot find emission above the 3 noise level (1 mJy) at 15 mas to East and 45 mas to North from the center in Fig. 1. Our array did not include MERLIN which is very powerful in mapping faint, extended emission. However, the peak brightness of component C was 55 mJy/beam using 7 mas beam (see Fig. 3 of Fanti et al. 1989) and the corresponding feature was clearly identified at lower frequencies as well (Fanti et al. 1985; Nan et al. 1988, 1991a). If present at the epoch of our observation, such a bright feature should have been identified in our map as well. We also note that the baseline sensitivity in our experiment was a factor of four better than that of Fanti et al. (1989) due to the larger bandwidth used.
The near circular track of the jet bridge line suggested earlier by Fanti et al. 1989 was based on their C component. 1.6 GHz observations support this regularly curving jet scenario (e.g. Nan et al. 1988). Unfortunately, we cannot decide whether component C moved away or faded below our 3 image noise. Observation of moving discrete 'bullets' of jet material would help us to study the jet kinematics. On the other hand, by comparing the observed structures in more details, we can identify features that cannot be explained by simple precession models. All of the VLBI maps published earlier show two features which do not follow the curving jet ridge line. These correspond to our components f and g in Table 2. (see also Fig. 1.).
Recent VLBA observations at 8.4 GHz also failed to detect component C (Dallacasa et al. 1998). There is no compact core at a few mJy level even at this frequency. Component f is identified just above the image noise, while component g also appears at about 1 mJy/beam contour level using 7 mas restoring beam (Dallacasa 1998, priv. comm.).
© European Southern Observatory (ESO) 1998
Online publication: September 17, 1998