Tielens et al.(1979) mapped this source and concluded it was a 260" double. Our
VLA observations show that the two components in the Tielens map are
actually a pair of
physically distinct double lobed sources, each with its own optical
identification. The two
USS sources are nearly north-south, so we have labeled them as 4C26.38N
and 4C26.38S
(Figure 5a). The southern source 4C26.38S has a redshift of ,
but we have not
been able to obtain a spectroscopic redshift of 4C26.38N. There is a
very faint and diffuse
object in our R band image centered between the lobes of 4C26.38N,
but it does not have
any strong emission lines in the observed bandpass.
The western radio lobe (A) of 4C26.38N is diffuse and the eastern
lobe (B) is resolved
into at least two components, the brighter component being closer to
the center (Fig 5b).
(Note that the map is distorted by beam smearing due to the angular
distance from the
pointing center.) These features make 4C26.38N similar to the low
surface brightness
lobes in USS sources in intermediate redshift clusters shown in Paper I,
but there is no
equivalent optical identification in the field of 4C26.38N. Other
distant diffuse FRI type
sources objects were found in Paper I, and it is very likely that
4C26.38N is beyond .
The existence of two USS sources so close together is surprising.
For the 4C catalog as
a whole one would expect a few unrelated pairs at this separation.
But the USS objects are
only about 5 percent of the 4C catalog. There are only 300 USS
sources at these flux
levels in the sky, so the probability of a pair of these within 260"
is .
The number of pairs one can form from N sources is N(N-1)/2, so the number of
unrelated pairs one would expect to see this close in projection in a
random sample of
300 objects is
.
If the small angular separation is not a projection coincidence and
4C26.38N is near
the same distance as 4C26.38S, then their minimum physical separation
is 1 Mpc. One
possible explanation for the extraordinary coincidence of two USS
objects close together
is that the mechanism for steepening the radio spectrum is operating
on the scale of a
few Mpc. For example, if the hot IGM of a rich cluster is responsible
for confining the
radio plasma, then spectral aging of the electron distribution will
steepen the synchrotron
spectrum. The low redshift x-ray cluster Abell 2256 has multiple
ultra steep spectrum
sources where this mechanism appears to be operating (e.g. Rottgering
et al 1994b). If a
common IGM is responsible for confining both 4C26.38N and 4C26.38S
this would indicate
a very substantial cluster (supercluster?) at a redshift of .
Another feature of
both 4C26.38N and 4C26.38S may be related to the nature of the IGM.
Both sources have a
large (and similar!) angular size, near the high end of the LAS
distribution for high redshift
sources (Rottgering et al 1994). Thus whether or not the two sources
are related, the IGM
around each must be smooth enough to allow the development of 200 kpc
structures. If
4C26.38N and 4C26.38S exist together in a common environment then it
is astonishingly
evolved at a very early epoch. Spectroscopy of 4C26.38N is urgently
required to determine if this might be the case.
The southern radio source 4C26.38S, has a normal FRII edge brightened double structure (Figures 5c-5f). Both lobes are roughly equivalent in brightness at 20 cm and 6 cm, and there is extended emission on both sides along the axis towards the lobes. Lobe A is more extended at a higher flux level then lobe B (more than one hot spot?); at low flux levels a jet feature is brighter on the B side. Polarization is detected in both lobes at both frequencies (Figure 5e and 5f). Lobe A is strongly depolarized and B is weakly depolarized. On the basis of the Laing-Garrington effect, the presence of a jet together with the polarization features suggests that B is the foreground lobe and A is the background lobe. In the higher resolution (0.4") 6 cm map (Figure 5g) lobe A is nearly unresolved and B is only slightly resolved. No obvious radio core is detected, but see below.
Infrared, optical, and narrow band images of 4C26.38S are shown in
Fig. 5h-51. The
centroid of all these images is displaced away from the geometrical
center of the radio lobes,
towards lobe A. The Ly emission is extended over at least 5",
and precisely aligned along
the radio axis. Within the emission line nebula, the line intensity is
peaked in the center,
with a faint plateau of extended emission along the radio axis in both
directions. A single
separate faint knot of line emission is seen at the geometrical center
of the radio lobes. The
broad B band contains the redshifted Ly
line and is a
combination of line and continuum
emission. However, the B band morphology is smoother than the emission
line morphology
with aligned emission extending along the radio axis for at least 7".
The B band peak is
offset to the northeast along the radio axis by 1.8" from the peak of
the Ly
emission. This
difference indicates a luminous and aligned UV continuum. The V band
image is similarly
aligned and extended but is bifurcated into two unequal components.
From the position of
this bifurcation the V band continuum extends towards the northeast
lobe B without fading
for 2". This northeastern aligned component is well matched to the B
band morphology.
The alignment is interrupted southwest of the bifurcation in the V band
towards the A lobe.
The peak in this component is displaced away from the radio axis to
the north, although
there is faint diffuse emission continuing along the radio axis.
The bifurcation is spatially
coincident with the peak in the Lyman
emission. (It is also
spatially coincident with a
2
feature in the 6 cm radio map - Fig 5d.) In I and K band
only the northeastern aligned
component is detected. In the I band image this northeastern
component is resolved into
a clumpy morphology. The object is compact but still aligned in K
band. Comparing
the B band with the K band, the peak in B is displaced to the
northeast along the radio
axis. In short, the V band and Ly
narrow band morphologies
overlap each other but are
inverted: the line emission peaks where the V band emission
disappears. The B,I, and K
band morphologies show an increasingly blue continuum along the
radio axis towards the northeastern lobe B.
If the core is at the center of the bifurcation, the natural interpretation is that the bifurcation is due to an obscuring disk, and the aligned continuum is primarily on the foreground side. Confirmation of the optical bifurcation feature, detection of any submillimeter continuum, and detection of the radio core would be extremely useful.
A possible faint companion object (Z) is visible to the north of the field center. It is detected in all bands (except the narrow band) and has similar colors to 4C26.38S.