Search

Using the PMAS Integral Field Unit on the Calar Alto 3.5m telescope we observed the southern component (Markarian 59) of the `cometary' starburst galaxy NGC 4861. Mrk 59 is centred on a giant nebula and concentration of stars 1 kpc in diameter. Strong $\rm H\alpha$ emission points to a star-formation rate (SFR) at least 0.47 $\rm M_{\odot}yr^{-1}$. Mrk 59 has a very high [OIII]$\rm\lambda5007/H\beta$ ratio, reaching 7.35 in the central nebula, with a second peak at a star-forming hotspot further north. Fast outflows are not detected but nebular motion and galaxy rotation produce relative velocities up to 40 km $\rm s^{-1}$. Spectral analysis of different regions with `Fitting Analysis using Differential evolution Optimisation' (FADO) finds that the stars in the central and `spur' nebulae are very young, $\rm \leq125~Myr$ with a large $\rm <10~Myr$ contribution. Older stars ($\rm \sim 1~Gyr$) make up the northern disk component, while the other regions show mixtures of 1 Gyr age with very young stars. This and the high specific SFR $\rm\sim 3.5~Gyr^{-1}$ imply a bimodal star formation history, with Mrk 59 formed in ongoing starbursts fuelled by a huge gas inflow, turning the galaxy into an asymmetric `green pea' or blue compact dwarf. We map the HeII$\lambda4686$ emission, and identify a broad component from the central nebula, consistent with the emission of $\sim 300$ Wolf-Rayet stars. About a third of the HeII$\lambda$4686 flux is a narrow line emitted from a more extended area covering the central and spur nebulae, and may have a different origin., Comment: 19 pages, 24 figures. Accepted for publication in MNRAS

We observe the radio galaxy PKS 1934-63 (at $z=0.1825$) using MUSE (Multi Unit Spectroscopic Explorer) on the Very Large Telescope (VLT). The radio source is GigaHertz Peaked Spectrum and compact (0.13 kpc), implying an early stage of evolution ($\leq 10^4$ yr). Our data show an interacting pair of galaxies, projected separation 9.1 kpc, velocity difference $\Delta(v)=216$ km $\rm s^{-1}$. The larger galaxy is a $\rm M_{*}\simeq 10^{11}M_{\odot}$ spheroidal with the emission-line spectrum of a high-excitation young radio AGN, e.g. strong [OI]6300 and [OIII]5007. Emission-line ratios indicate a large contribution to the line luminosity from high-velocity shocks ($\simeq 550$ km $\rm s^{-1}$) . The companion is a non-AGN disk galaxy, with extended $\rm H\alpha$ emission from which its star-formation rate is estimated as $\rm 0.61~M_{\odot}yr^{-1}$. Both galaxies show rotational velocity gradients in $\rm H\alpha$ and other lines, with the interaction being prograde-prograde. The SE-NW velocity gradient of the AGN host is misaligned from the E-W radio axis, but aligned with a previously discovered central ultraviolet source, and a factor 2 greater in amplitude in $\rm H\alpha$ than in other (forbidden) lines (e.g. [OIII]5007). This could be produced by a fast rotating (100-150 km $\rm s^{-1}$) disk with circumnuclear star-formation. We also identify a broad component of [OIII]5007 emission, blueshifted with a velocity gradient aligned with the radio jets, and associated with outflow. However, the broad component of [OI]6300 is redshifted. In spectral fits, both galaxies have old stellar populations plus $\sim 0.1\%$ of very young stars, consistent with the galaxies undergoing first perigalacticon, triggering infall and star-formation from $\sim 40$ Myr ago followed by the radio outburst., Comment: 24 pages, 35 figures, MNRAS in press

We investigate the spiral galaxy NGC 5394, which is strongly interacting with the larger spiral NGC 5395 (the pair is Arp 84), using optical integral-field spectroscopy from the CALIFA survey. Spatially-resolved equivalent-widths, emission-line ratios and kinematics reveal many features related to the interaction, which has reshaped the galaxy. $\rm H\alpha$ maps (with other diagnostic emission lines) show a concentrated central ($r<1$ kpc) starburst and three less luminous star-forming regions (one knot far out in the northern arm), and we estimate the dust-corrected total star-formation rate as 3.39 $\rm M_{\odot}yr^{-1}$. However, much of the galaxy, especially the outer tidal arms, has a post-starburst spectrum, evidence of a more extensive episode of star-formation a few $\times 10^8$ yr ago, triggered by the previous perigalacticon. The $\rm [NII]6584/H\alpha$ ratio is high in the nucleus, reaching 0.63 at the centre, which we interpret as related to high electron density ($n_e\simeq 750$ $\rm cm^{-3}$ from the $\rm [SII]{6717\over 6731}$ ratio). We find a central region of strong and blueshifted NaI(5890,5896) absorption, indicative of a starburst-driven outflow from the nucleus at an estimated velocity $\sim 223$ km $\rm s^{-1}$. The CALIFA data also show an annular region at radii 2.25--4 kpc from the nucleus, with elevated ratios of [NII], [OI]6300 etc. to the Balmer lines -- this is evidence of shock excitation, which might be the result of interaction-triggered gas inflow., Comment: 17 pages, 29 figures, MNRAS, accepted and in print

We performed long-slit optical spectroscopy (GTC-OSIRIS) of 6 radio-loud QSOs at redshifts $2

We discuss the ability of the planned Euclid mission to detect deviations from General Relativity using its extensive redshift survey of more than 50 Million galaxies. Constraints on the gravity theory are placed measuring the growth rate of structure within 14 redshift bins between z=0.7 and z=2. The growth rate is measured from redshift-space distortions, i.e. the anisotropy of the clustering pattern induced by coherent peculiar motions. This is performed in the overall context of the Euclid spectroscopic survey, which will simultaneously measure the expansion history of the universe, using the power spectrum and its baryonic features as a standard ruler, accounting for the relative degeneracies of expansion and growth parameters. The resulting expected errors on the growth rate in the different redshift bins, expressed through the quantity f\sigma_8, range between 1.3% and 4.4%. We discuss the optimisation of the survey configuration and investigate the important dependence on the growth parameterisation and the assumed cosmological model. We show how a specific parameterisation could actually drive the design towards artificially restricted regions of the parameter space. Finally, in the framework of the popular "\gamma -parameterisation", we show that the Euclid spectroscopic survey alone will already be able to provide substantial evidence (in Bayesian terms) if the growth index differs from the GR value \gamma=0.55 by at least \sim 0.13. This will combine with the comparable inference power provided by the Euclid weak lensing survey, resulting in Euclid's unique ability to provide a decisive test of modified gravity., Comment: 18 pages, 15 figures, accepted by MNRAS

We investigate the prospects of extending observations of high redshift QSOs to z>8 by means of a very wide-area near-infrared slitless spectroscopic survey, e.g. the planned survey with the European Space Agency's Euclid telescope. For any QSOs at z>8.06 the strong Lyman-alpha line will enter the wavelength range of the Euclid near-infrared spectrometer and imaging photometer (NISP). We perform a detailed simulation of Euclid NISP slitless spectroscopy (with the parameters of the wide survey) in an artificial field containing QSO spectra at all redshifts up to z=12. The simulation spectra are analysed with an automated redshift finder, and a detection rate estimated as a function of H magnitude and redshift. Spectroscopic identification of QSOs would reach deeper limits for the redshift ranges where either H-alpha (0.678.06) is visible. Furthermore, if photometrically-selected z>8 spectra can be re-examined and re-fitted to minimize the effects of spectral contamination, the QSO detection rate in the Lyman-alpha window will be increased by an estimated 60% and will then be better here than at any other redshift, with an effective limit H=21.5. With an extrapolated rate of QSO evolution, we predict the Euclid wide (15000 sq. deg.) spectroscopic survey will identify 20-35 (19-33 with a small correction for lineless objects) very luminous QSOs at z>8.06., Comment: 16 pages, 19 figures, accepted for publication in MNRAS

We examine the colour-magnitude relation of approximately 5000 Brightest Cluster Galaxies (BCGs) in the Sloan Digital Sky Survey, and compare with non-BCG E/S0 galaxies. The colour-magnitude and colour-sigma (velocity dispersion) relations are flatter in slope (by a factor of about 2) for BCGs than for non-BCG E/S0s, and the BCGs also tend to be redder by 0.01 magnitudes in g-r. We investigate radial colour gradients in both samples, using the ratio of the de Vaucouleurs radii in the g and r bands. We find BCGs have significantly flatter (by 23%) mean colour gradients than other high luminosity E/S0s. In early-type galaxies, the colour gradients are strongest at intermediate luminosities of Mr=-22. Colour gradients in E/S0s increase with radius (up to 10kpc) and are negatively correlated with 10sigma + Mr (velocity dispersion relative to luminosity) and with mass density. The gradients also tend to decrease with increasing stellar age. These trends are weak or not seen in BCGs, in which the mean colour gradient is low whatever the other properties. We discuss possible explanations, which involve a greater amount of dry merging in the formation history of the BCGs., Comment: 16 pages, 24 figures, revised, accepted for publication in MNRAS.

We select a sample of 70378 E/S0 (early-type) galaxies at 0

We present first results from the Southern Cosmology Survey, a new multiwavelength survey of the southern sky coordinated with the Atacama Cosmology Telescope (ACT), a recently commissioned ground-based mm-band Cosmic Microwave Background experiment. This article presents a full analysis of archival optical multi-band imaging data covering an 8 square degree region near right ascension 23 hours and declination -55 degrees, obtained by the Blanco 4-m telescope and Mosaic-II camera in late 2005. We describe the pipeline we have developed to process this large data volume, obtain accurate photometric redshifts, and detect optical clusters. Our cluster finding process uses the combination of a matched spatial filter, photometric redshift probability distributions and richness estimation. We present photometric redshifts, richness estimates, luminosities, and masses for 8 new optically-selected clusters with mass greater than $3\times10^{14}M_{\sun}$ at redshifts out to 0.7. We also present estimates for the expected Sunyaev-Zel'dovich effect (SZE) signal from these clusters as specific predictions for upcoming observations by ACT, the South Pole Telescope and Atacama Pathfinder Experiment., Comment: 12 pages, 8 figures, accepted in ApJ. Reflects changes from referee as well as a new Table providing mass estimates and positions for all clusters in the survey

We investigate Arp 295, a pair of interacting spirals at z=0.023. We measure scalelengths 5.24 kpc for Arp 295a and 2.52 kpc for 295b. There is a much smaller Im galaxy associated with the larger spiral. Arp 295b is asymmetric with the disk more extended eastwards but with the brightest star-formation regions on the west side of the nucleus. The spectrum of Arp 295b shows strong emission lines with [OII]3727 prominent. We measure the total H-alpha line luminosities of Arp 295b and Arp 295c as 4.69x 10^{41} and 6.76x 10^{40} ergs/s, corresponding to star-formation rates 3.7 and 0.53 solar masses/year. We map the kinematics of Arp 295b in H-alpha with the Manchester Echelle Spectrograph. The maximum disk rotation velocity is 252.6 km/s. The rotation curve of Arp 295b is very asymmetric. The east (approaching) side has a higher radial velocity than the west with the maximum difference 5 arcsec from the nucleus, where it is 88 km/s or a factor 1.675. The slowed rotation on the western side of the nucleus may have locally enhanced star-formation., Comment: 11 pages, 10 figures, revised version. Rev Mex Ast, in press. To appear vol 43, no. 1, April 2007

We perform UBR imaging and optical spectroscopy of the interacting galaxy pair Arp 104, at z=0.0098. This consists of NGC5218, a disturbed Sb barred spiral with an inclined outer shell, the round spheroidal NGC5216, a connecting bridge of length 50 kpc and a curved plume. Neither galaxy shows emission lines. NGC5218 has strong Balmer lines and appears to have undergone a major starburst some 0.2 Gyr ago, triggered by the last close passage of the two galaxies. The galaxy is very red in its centre, suggesting it is dusty, but its outer regions, and the bridge connecting the two galaxies, have the blue colours of 0.2-0.4 Gyr old stars. NGC5216 lacks strong Balmer lines but outside its centre is blue in U-B, suggesting it experienced a star-formation episode only about 40 Myr ago. This could have been fuelled by gas from NGC5218, transfered through the bridge. The bridge passes through NGC5216 to emerge as a plume extending 14 kpc to the SW. The plume, from its colours, is very young and may be a site of ongoing star-formation or the formation of a tidal dwarf galaxy., Comment: 12 pages, 11 figures

We have performed a deep spectroscopic survey of extremely red galaxies on the GOODS-South field, using GMOS on Gemini South. We present here spectra and redshifts for 16 ERGs at 0.87

Hard X-ray surveys have uncovered a large population of heavily obscured AGN. They also reveal a population of quasars with moderate obscuration at both visible and X-ray wavelengths. We use Chandra selected samples of quasars from the ELAIS Deep X-ray Survey (EDXS) and CDF-N to investigate the obscuration towards the nuclei of moderately obscured AGN. We find an inverse correlation between the optical to X-ray flux ratio and the X-ray hardness ratio which can be interpreted as due to obscuration at visible and X-ray wavelengths. We present detailed optical and near-infrared data for a sample of optically-faint (R>23) quasars from the EDXS. These are used to constrain the amount of rest-frame UV/optical reddening towards these quasars. It is found that optically-faint quasars are mostly faint due to obscuration, not because they are intrinsically weak. After correcting for reddening, the optical magnitudes of most of these quasars are similar to the brighter quasars at these X-ray fluxes. Combining with gas column densities inferred from the X-ray observations we consider the gas-to-dust ratios of the obscuring matter. We find that the quasars generally have higher gas-to-dust absorption than that seen in the Milky Way - similar to what has been found for nearby Seyfert galaxies. We consider the possible existence of a large population of X-ray sources which have optical properties of Type 1 (unobscured) quasars, but X-ray properties of Type 2 (obscured) quasars. We show that the observed distribution of optical-to-X-ray flux ratios of quasars at z>1 is skewed to low values compared to the intrinsic distribution..., Comment: 12 pages, 10 figures, ApJ, in press

We identify a deep sample of 198 extremely red objects (EROs) in (50.4 sq. arcmin of) the Chandra Deep Field South, selected as I-K>3.92 galaxies to a limit K=22. The ERO number counts remain well below the predictions for pure luminosity evolution, and fall below even a non-evolving model, suggesting the comoving number density of passive/very red galaxies decreases with redshift. The angular correlation function of these EROs indicates stronger clustering than that of other galaxies at the same magnitudes, and is best-fitted by models in which the EROs have a comoving correlation radius 12.5/h Mpc, or 21.4/h Mpc in a stable clustering model. We find a 40-arcsec diameter grouping of 10 EROs, centered on the Chandra source (and ERO) XID:58, with colours suggesting a cluster of mostly passive EROs at approx. z=1.5. The 942 ksec Chandra survey detected 73 X-ray sources in the area of our ERO sample, of which 17 coincide with EROs. Of these sources, 13 have X-ray properties indicative of obscured AGN, while the faintest 4 may be starburst galaxies. In addition, we find evidence that Chandra sources and EROs are positively cross-correlated at non-zero (2-20 arcsec) separations, implying that they tend to trace the same large-scale structures., Comment: 17 pages, latex, 14 figures, accepted for publication in MNRAS. Major revisions from original version, with a new, reselected sample of EROs

Using the HST WFPC2 we perform deep I-band imaging of 9 radio-selected (limit 14 microJanskys at 8.5 GHz) faint galaxies from Roche, Lowenthal and Koo (2002). Two are also observed in V. Six of the galaxies have known redshifs of 0.40.2 Gyr, to a late stage of the interaction. One of these 9 galaxies may be an exception in that it is a large red elliptical and its strong radio emission is more likely to be from an obscured AGN., Comment: 12 pages, latex, 8 figures, accepted for publication in MNRAS

The SCUBA 8-mJy survey is the largest submm extragalactic mapping survey undertaken to date, centred on the Lockman Hole and ELAIS N2 regions. Here, we present new 1.4-GHz imaging of the depth and resolution necessary to reliably identify radio counterparts for 18 of 30 submm sources. Armed with this greatly improved positional information, we present and analyse new optical, near-IR and XMM-Newton X-ray imaging to identify optical/IR host galaxies to half of the submm-selected sources in those fields. As many as 15% of the submm sources detected at 1.4 GHz are resolved by the 1.4'' beam and a further 25% have more than one radio counterpart, suggesting that radio and submm emission arise from extended starbursts and that interactions are common. We note that less than a quarter of the submm-selected sample would have been recovered by targeting optically faint radio sources, underlining the selective nature of such surveys. At least 60% of the radio-confirmed optical/IR host galaxies appear to be morphologically distorted; many are composite systems - red galaxies with relatively blue companions; just over half are found to be very red or extremely red; contrary to popular belief, most are sufficiently bright to be tackled with spectrographs on 8-m telescopes. We estimate the median redshift of the S(850um) ~ 8 mJy submm galaxy population: if the radio/far-IR correlation holds at high redshift, and our sample is unbiased, we derive a conservative median z >= 2.0, or >= 2.4 using spectral templates more representative of known submm galaxies [abridged]., Comment: In press at MNRAS. Extra IR imaging now included. 27 pages; 14 figures, mostly colour. The resolution of Figs 1 and 2 has been degraded. Higher quality versions of paper available at http://www.roe.ac.uk/~rji/8mjy.ps.gz (a4) and http://www.roe.ac.uk/~rji/8mjy_letter.ps.gz (letter)

We investigate the number counts, clustering, morphology and radio/X-ray emission of extremely red objects (EROs), defined as galaxies with R-K>5.0. This criterion will select old, passive ellipticals at z>0.9. To K=21 we identify a total of 158 EROs in 81.5 sq. arcmin. The ERO number counts are lower than predicted by a pure luminosity evolution model, but higher than predicted by current CDM-based hierarchical models. They can be fitted by a combination of moderate merging with a decrease with redshift in the comoving number density of passive galaxies. We investigate the angular correlation function and find that the EROs have a much higher correlation amplitude than full K-limited samples of galaxies. This implies that the intrinsic clustering of EROs is stronger than that of present-day E/S0 galaxies. We estimate the comoving correlation radius as 10-13 Mpc. We estimate seeing-corrected angular sizes and morphological types for a K<19.5 subsample (31 galaxies), and find a 3:2 mixture of bulge and disk profiles. 1/4 appear to be interacting, disturbed or irregular. The angular sizes of the bulge-profile EROs are consistent with passively evolving ellipticals at 0.9

We investigate the nature of the faintest radio sources detected in 3 VLA surveys, to faint limits 8 microJy at 8.5 GHz. Using the Keck Low Resolution Imaging Spectrograph in BRI and the Near Infra Red Camera in K, we image 51 radio sources, and identify probable optical counterparts for 50. With LRIS spectroscopy, we successfully acquire new redshifts for 17 sources, and 9 have redshifts from previous data. Of this sample of 26, about 60 per cent are disk galaxies with high radio to optical ratios, indicating ongoing or very recent starbursts, at redshifts z=0-1. Others are normal, non-interacting galaxies at z<0.4, QSOs and radioluminous giant ellipticals (obscured AGN?). All of the 14 non-QSO galaxies with spectra show emission lines, but with luminosities corresponding to star-formation rates about an order of magnitude lower than estimated from the high radio luminosities (dust extinction?). Three have strong H-delta absorption (`e+a galaxies'). The Keck spectra also provide approximate emission line widths and rotation curves. Most (12/14) have kinematics implying large dynamical masses, and moderate excitation ratios consistent with L* spirals. The other two, which have sub-L* optical luminosities, appear to be low-mass galaxies with much higher excitations and may be more similar to HII galaxies., Comment: 20 pages, latex, 16 figures, accepted for publication in MNRAS

We investigate a very luminous Lyman-alpha emitting galaxy, associated with a z=2.3 damped absorption system in the spectrum of the QSO PHL 957. The galaxy was observed with the HST WFPC2, through a narrow filter (F410M) corresponding to rest-frame Lyman -alpha, for 41.2 ksec, with shorter exposures in F555W and F814W. The galaxy is resolved into a close (0.35 arcsec) pair of two components, CFgA and CFgB. The profile of CFgA is consistent with an exponential disk, but CFgB is closer to a bulge galaxy. Fr om the observed colours we estimate rest-frame Lyman-alpha equivalent widths of 151 Angstroms for CFgA and 33 for CFgB. From the F814W and F555W magnitudes we estimate rest-frame blue-band absolute magnitudes of -23.12 for CFgA and -23.24 for CFgB, signi ficantly brighter than local galaxies of the same size. CFgA shows a remarkable 3.9 magnitudes of surface brightness enhancement relative to local spirals, and we speculate that this may be the result of a very luminous starburst, triggered by the merger of the two components. The Lyman-alpha emission could be from a brief phase early in the starburst in which most of these photons can escape. We search for other z=2.3 galaxies with strong Lyman-alpha emission, selecting these by a colour F410M-F555W<-0.2. Eight candidate sources, all fainter than CFgA, are identified. One is a point-source and may be an AGN; the others are irregular a nd of low surface brightness, appearing typical of Lyman break galaxies, but with Lyman-alpha equivalent widths near 100 Angstroms., Comment: 17 pages, Latex, 17 figures. Accepted for publication in MNRAS

We model galaxy evolution in the 6.75 and 15 micron passbands of the ISO satellite, by combining models of galaxy evolution at optical wavelengths (which are consistent with the optical galaxy counts) with observed spectral energy distributions in the infra-red. Our model is consistent with the local 12 micron galaxy luminosity, if 3.5 per cent of spirals at z=0 are in interacting pairs with mid-infra-red luminosities enhanced by major starbursts. Source counts from deep ISO surveys exceed non-evolving predictions but are more consistent with our evolving model. The steep number count of 6.75 micron sources appears to be explained primarily by evolving early-type galaxies, whereas at 15 microns the main contributors are star-forming spirals and starbursting interacting/merging galaxies. The 31 per cent of 15 micron sources which are visibly interacting galaxies have high mid-infra-red/optical flux ratios indicating major starbursts combined with dust extinction. The numbers and high mean redshift of these sources suggest the merger-starbursts are increasing in luminosity with redshift, approximately as (1+z)^2, in addition to undergoing (1+z)^2 number evolution reflecting the optically observed increase with redshift in the fraction of interacting galaxies., Comment: 12 pages, Latex, 12 figures, submitted to MNRAS

On five K-band Omega camera images, covering a total of 162.2 sq. arcmin to K=19.5, we investigate (i) the clustering environment of 5 radio galaxies at 0.71.5) redshifts. We also find a significant excess of 1.5--5.0 arcsec separation pairs of galaxies compared to the expectation from the inward extrapolation of omega(theta). To K=19.5, we estimate that 11.0\pm 3.4 per cent of galaxies are in close pairs in excess of omega(theta). This can be explained if the local rate of galaxy mergers and interactions increases with redshift as $\sim (1+z)^m$ with $m=1.33_{-0.51}^{+0.36}$., Comment: 14 pages, latex, 8 figures, submitted to MNRAS

We investigate the angular correlation function, omega(theta), and third and fourth-order hierarchical moments of a survey with approx. 70000 faint galaxies detected on a total area 1.74 sq. deg. to a limit of R=23.5. The omega(theta) amplitudes and scaling with magnitude limit are most consistent with approximately stable clustering (epsilon=0). On the larger of our fields, the amplitude of omega(theta) is significantly higher at 2-5 arcsec separations than at >5 arcsec, indicating an excess of physically close galaxy pairs above the number expected from omega(theta) seen at larger scales. This is consistent with the number of interacting galaxy pairs seen locally combined with a merger/interaction rate evolving as (1+z)^m with m=2.01_{-0.69}^{+0.52}. We also estimate the hierarchical moments s_3 and s_4 from the galaxy counts-in-cells. For 18.5

We investigate the angular correlation function, $\omega(\theta)$, of the galaxi es detected in the K-band on 17 fields (101.5 square arcmin in total), each containing a $z\sim 1.1$ radio galaxy. There is a significant detection of galaxy clustering at $K\sim 20$ limits, with an amplitude higher than expected from simple models which fit the faint galaxy clustering in the blue and red passbands, but consistent with a pure luminosity evolution model i f clustering is stable and early-type galaxies have a steeper correlation function than spirals. We do not detect a significant cross-correlation between the radio galaxies and the other galaxies on these fields, obtaining upper limits consistent with a mean clustering environment of Abell class 0 for $z\sim 1.1$ radio galaxies, similar to that observed for radio galaxies at $z\sim 0.5$. At $K\leq 20$, the number of galaxy-galaxy pairs of 2--3 arcsec separations exceeds the random expectation by a factor of $2.15\pm 0.26$. This excess suggests at least a tripling of the local merger rate at $z\sim 1$., Comment: 13 pages, 3 tables, 7 postscript figures, TEX, submitted to MNRAS

We use $K-band imaging to investigate the angular size and morphology of 10 6C radio galaxies, at redshifts $1\leq z\leq 1.4$. Two appear to be undergoing mergers, another contains two intensity peaks aligned with the radio jets, while the other seven appear consistent with being normal ellipticals in the K-band. Intrinsic half-light radii are estimated from the areas of each radio galaxy image above a series of thresholds. The 6C galaxy radii are found to be significantly smaller than those of the more radioluminous 3CR galaxies at similar redshifts. This would indicate that the higher mean K-band luminosity of the 3CR galaxies results from a difference in the size of the host galaxies, and not solely from a difference in the power of the active nuclei. The size-luminosity relation of the $z\sim 1.1$ 6C galaxies indicates a 1.0--1.8 mag enhancement of the rest-frame R-band surface brightness relative to either local ellipticals of the same size or FRII radio galaxies at $z<0.2$. The 3CR galaxies at $z\sim 1.1$ show a comparable enhancement in surface brightness. The mean radius of the 6C galaxies suggests that they evolve into ellipticals of $L\sim L^*$ luminosity, and is consistent with their low redshift counterparts being relatively small FRII galaxies, a factor $\sim 25$ lower in radio luminosity, or small FRI galaxies a factor of $\sim 1000$ lower in radio luminosity. Hence the 6C radio galaxies may undergo at least as much optical and radio evolution as the 3CR galaxies., Comment: 17 pages, 7 postscript figures, TEX, submitted to MNRAS

Optical spectroscopy has enabled us to identify the optical counterparts to over 200 faint X-ray sources to a flux limit of S_(0.5-2keV)=4x10^-15 erg s^-1 cm^-2 on 5 deep ROSAT fields. Here we present a spectral analysis of all the X-ray sources to investigate claims that the average source spectra harden at faint X-ray flux. From a hardness ratio analysis we confirm that the average spectra from 0.5-2 keV harden from an equivalent photon index of $\Gamma=2.2$ at S_(0.5-2keV)=1x10^-13 erg s^-1 cm^-2 to $\Gamma=1.7$ below 1x10^-14 erg s^-1 cm^-2. These spectral changes are due to the emergence of an unidentified source population rather than the class of X-ray QSOs already identified. The 128 QSOs detected so far show no evidence for spectral hardening over this energy range and retain a mean photon index of $\Gamma=2.2$. Recent work suggests that many of the remaining unidentified sources are X-ray luminous galaxies. Taking a subset identified as the most likely galaxy candidates we find that these show significantly harder spectra than QSOs. The emission line galaxies in particular show spectra more consistent with the residual X-ray background, with $\Gamma=1.51 \pm 0.1$ from 0.1-2 keV. Individually the galaxies appear to be a mixture of absorbed and unabsorbed X-ray sources. Combined with recent cross-correlation results and work on the source number count distribution, these results suggest that we may be uncovering the missing hard component of the cosmic X-ray background., Comment: MNRAS Accepted. (References revised and updated). The preprint is also available at http://www.ast.cam.ac.uk/preprint/PrePrint.html or http://www.ast.cam.ac.uk/~omar/astro.html