Until recently, the study of the most distant galaxies was very restricted: unresolved ground-based imaging gave measurements of the global colours; noisy spectra revealed AGN- or starburst-driven emission lines; and WFPC 2 imaging with the Hubble Space Telescope (HST) showed the shape of high-redshift galaxies in their unfamiliar rest-frame ultraviolet light. However, our knowledge has radically increased through the light-gathering power of the Keck telescopes, and the availability of the near-infrared NICMOS camera on HST. Deep Keck spectra showing the rest-ultraviolet continuum and stellar- and ISM-absorption features in z ~ 3 galaxies (e.g., Steidel et al. 1996b; Pettini et al. 2000) has opened the door to spectroscopic study of stellar populations in the distant Universe. A complementary approach to exploring the composition of high-redshift objects comes from multi-waveband HST imaging from the ultraviolet to the near-infrared, which provides the resolution necessary to study their spatially-resolved stellar populations.
Morphology offers a window on the evolutionary status of
galaxies. However, because of band-shifting effects, the interpretation
of the shape of galaxies hinges on knowledge of their redshifts and
spectral energy distributions (SEDs). In this article we explore this
through high resolution optical/near-infrared imaging and deep
spectroscopy of distant galaxies. To resolve the issue of whether the
peculiar galaxies which dominate the number counts at faint magnitudes
are the counterparts of local irregulars or whether their morphological
peculiarity is due mainly to band-shifting effects at high redshift, an
unbiased study of the rest-frame optical morphological properties is
demanded, where the effects of dust and recent star formation are less
dominant than in the rest ultraviolet. In
Section 2 we address this by analysing the
IDT-NICMOS images of the northern Hubble Deep Field (HDF). In
Section 3 we present a case study of the
brightest z ~
3 galaxy in the HDF. This example, HDF 4-555.1 (known as the "Hot
Dog"), is a highly elongated system and is sufficiently extended to
allow resolved ground-based spectroscopy, which we have obtained with
Keck/LRIS
(Oke et al. 1995).
We use a similar approach to explore the stellar populations in a pair
of z
4 gravitationally-lensed
arcs behind the cluster Abell 2390, which is described in
Section 4. Throughout, we assume a cosmology
where
h50 = H0 / 50 km s-1
Mpc-1, q0 = 0.5 and
= 0, unless otherwise stated.
{All magnitudes in this paper are with respect to the AB system
(Oke & Gunn 1983)
where mAB = -48.57 - 2.5 log10
f
/ (erg
cm-2 s-1 Hz-1).