New radio data has been presented for PKS 1718-649 and existing radio and optical data from the literature have been reviewed. Two interpretations for the nature of PKS 1718-649 based on these data have been discussed. The first interpretation, that PKS 1718-649 is a core-jet radio source which is core dominated, receives little support although it cannot be ruled out completely. The major reason for rejection of this interpretation is the lack of evidence for a beamed component of synchrotron emission at radio or optical wavelengths, and the morphology of the pc-scale radio source.
The second interpretation appears to be more plausible, that PKS 1718-649 is a GHz Peaked-Spectrum radio source associated with a galaxy. The major evidence supporting this interpretation is that PKS 1718-649 fulfils all of the major criteria for GPS sources bar one, that GPS sources are highly luminous. This last point can be readily explained in terms of the selection biases on current samples of GPS radio sources. PKS 1718-649 also provides qualitative support for the suggestion of radio source confinement.
However, this interpretation is still somewhat uncertain. Further observations would be helpful for confirmation. Future VLBI observations will allow a determination of the speed of separation of the pc-scale components. If the source is a core-jet, a substantial apparent motion is expected; if a GPS source, no substantial motion is expected (c.f. Tzioumis et al. 1989). Monitoring of the radio flux density simultaneously at low and high radio frequency will determine if variability is present.
If PKS 1718-649 can be confirmed as a GPS source it will be an important discovery. At a red shift of z=0.014, it would be by far the nearest GPS source and allow unprecedented opportunities for investigation of the radio source and its environment with high spatial resolution and high sensitivity observations. Models for the suggested confinement of the radio source have been, up to recently, largely qualitative in their predictions and untestable due to the difficulty of observing the generally very distant GPS sources. However, models are becoming more rigorous (e.g. Bicknell and Dopita 1996, in preparation). High quality observations of a low red shift GPS source and its environment would be very valuable for testing new detailed models.