Astron. Astrophys. 351, 47-58 (1999)

1. Introduction

High redshift radio galaxies (HzRG) (redshift 2) were until a few years ago the most distant "galaxies" (or proto-galaxies) we could study. They had a crucial role as probes of the distant universe. In the late 90's thanks to the Hubble Space and Keck telescopes, many star forming galaxies have been discovered at 3-4 (see for instance the recent work of Pettini, Steidel, Giavalisco). The advantage of these objects is that, unlike HzRG, the emission is not influenced by the nuclear activity. The study of the stellar component (crucial in Cosmology and galaxy formation studies) is in principle easier, since the emission is dominated by the stars.

The very blue continuum observed in HzRG suggested the presence of young stars (e.g. Lilly, Longair & McLean 1983) and it was believed that we were observing galaxies in the process of formation. However, it was discovered later that at least an important contribution to the continuum radiation is not stellar, but a consequence of the nuclear activity.

One of the most interesting properties of HzRG is the so called alignment effect (observed at 0.7): the continuum and emission line structures are extended and closely aligned with the radio axis (Chambers et al. 1987, McCarthy et al. 1987). The nature of this phenomenon is controversial but it suggests that the continuum and line emission are closely linked to the nuclear activity, rather than being due to the stellar population characteristic of a galaxy in the process of formation. Three main mechanisms have been proposed to explain the alignment effect and all of them could play a role: 1) scattered light from a quasar hidden from the line of sight (Tadhunter, Fosbury & di Serego Alighieri 1989) 2) young stars whose formation is triggered by the passage of the radio jet through the ambient gas (Rees 1989) 3) nebular continuum (Dickson et al. 1995). The discovery of polarized continuum with the electric vector perpendicular to the axis of the UV structures provides strong evidence for the existence of scattered light in many HzRG at 2 (Cimatti et al. 1996, 1997, 1998). Compelling evidence for a young stellar population in a HzRG does not exist yet, except possibly for 4C41.17 (3.8) (Dey et al. 1997). In any case, the observed properties of HzRG are a consequence (at least in the UV rest frame) of the nuclear activity.

In spite of this complexity, HzRG still have a crucial role in the understanding of galaxy formation, since they are the only way we have to study the early stages of giant ellipticals . Since all powerful radio galaxies at low redshift are giant ellipticals and there is evidence that this is also the case at 1 (Best et al. 1998), it is believed that the host galaxies of radio galaxies at higher redshifts are also giant ellipticals (Pentericci et al. 1999). It is however necessary to understand how the nuclear activity influences what we see in order to make a correct interpretation of the observed properties.

Another interesting aspect is the relationship between the rapid phase(s) of star formation and the formation and fueling of a massive black hole during the formation of these galaxies which are destined to become the massive ellipticals we see today. Exactly what this relationship is is not clear although interactions/mergers are likely to play a role.

It has been proposed that ultraluminous infrared galaxies (ULIRGs) are progenitors of the giant ellipticals of today (Kormendy & Sanders, 1992). On the other hand, Sanders et al. (1988) suggested that ULIRGs will evolve into quasars. While there is a continuing debate about what powers these galaxies (starburst or active galactic nuclei (AGN)) it is clear that 1) ULIRGS show clear evidence for interactions/mergers (e.g. Borne et al. 1999, Sanders et al. 1988)  2) some, at least, contain powerful AGN (e.g. Sanders et al. 1988).

A population of luminous galaxies in the submillimetre wavelengths has been discovered in recent years. Studies of the spectral energy distributions suggest that these are the analogues at high redshift of ULIRGs at low redshift (e.g. Sanders & Mirabel 1996). As for many local ULIRGS, it is not clear whether these galaxies are powered by starburst or active galactic nuclei. The study of distant ultraluminous submm sources can provide important information about the nature of ULIRGs and how a massive black hole forms at high redshift and coexists with a powerful starburst. The interpretation of the ISM properties of these objects in the context of distant radio galaxies then becomes crucial.

We study here the UV (rest frame) spectra of three HzRG and the hyperluminous submillimetre source SMM J02399-0136 (2.8).

© European Southern Observatory (ESO) 1999

Online publication: November 2, 1999
helpdesk.link@springer.de