We highlight the power of the Gaia DR2 in studying many fine structures of the Hertzsprung-Russell diagram (HRD). Gaia allows us to present many different HRDs, depending in particular on stellar population selections. We do not aim here for completeness in terms of types of stars or stellar evolutionary aspects. Instead, we have chosen several illustrative examples. We describe some of the selections that can be made in Gaia DR2 to highlight the main structures of the Gaia HRDs. We select both field and cluster (open and globular) stars, compare the observations with previous classifications and with stellar evolutionary tracks, and we present variations of the Gaia HRD with age, metallicity, and kinematics. Late stages of stellar evolution such as hot subdwarfs, post-AGB stars, planetary nebulae, and white dwarfs are also analysed, as well as low-mass brown dwarf objects. The Gaia HRDs are unprecedented in both precision and coverage of the various Milky Way stellar populations and stellar evolutionary phases. Many fine structures of the HRDs are presented. The clear split of the white dwarf sequence into hydrogen and helium white dwarfs is presented for the first time in an HRD. The relation between kinematics and the HRD is nicely illustrated. Two different populations in a classical kinematic selection of the halo are unambiguously identified in the HRD. Membership and mean parameters for a selected list of open clusters are provided. They allow drawing very detailed cluster sequences, highlighting fine structures, and providing extremely precise empirical isochrones that will lead to more insight in stellar physics. Gaia DR2 demonstrates the potential of combining precise astrometry and photometry for large samples for studies in stellar evolution and stellar population and opens an entire new area for HRD-based studies.

Accounting for ~20% of the total QSO population, Broad Absorption Line QSOs are still an unsolved problem in the AGN context. They present wide troughs in the UV spectrum, due to material with velocities up to 0.2 c toward the observer. The two models proposed in literature try to explain them as a particular phase of the evolution of QSOs or as normal QSOs, but seen from a particular line of sight. We built a statistically complete sample of Radio-Loud BAL QSOs, and carried out an observing campaign to piece together the whole spectrum in the cm wavelength domain, and highlight all the possible differences with respect to a comparison sample of Radio-Loud non-BAL QSOs. VLBI observations at high angular resolution have been performed, to study the pc-scale morphology of these objects. Finally, we tried to detect a possible dust component with observations at mm-wavelengths. Results do not seem to indicate a young age for all BAL QSOs. Instead a variety of orientations and morphologies have been found, constraining the outflows foreseen by the orientation model to have different possible angles with respect to the jet axis.

The origin of broad-absorption-line quasars (BAL QSOs) is still an open issue. Accounting for ~20% of the QSO population, these objects present broad absorption lines in their optical spectra generated from outflows with velocities up to 0.2c. In this work we present the results of a multi-frequency study of a well-defined radio-loud BAL QSO sample, and a comparison sample of radio-loud non-BAL QSOs, both selected from the Sloan Digital Sky Survey (SDSS). We aim to test which of the currently-popular models for the BAL phenomenon - `orientation' or 'evolutionary' - best accounts for the radio properties of BAL quasars. Observations from 1.4 to 43 GHz have been obtained with the VLA and Effelsberg telescopes, and data from 74 to 408 MHz have been compiled from the literature. The fractions of candidate GHz-peaked sources are similar in the two samples (36\pm12% vs 23\pm8%), suggesting that BAL QSOs are not generally younger than non-BAL QSOs. BAL and non-BAL QSOs show a large range of spectral indices, consistent with a broad range of orientations. There is weak evidence (91% confidence) that the spectral indices of the BAL QSOs are steeper than those of non-BAL QSOs, mildly favouring edge-on orientations. At a higher level of significance (\geq97%), the spectra of BAL QSOs are not flatter than those of non-BAL QSOs, which suggests that a polar orientation is not preferred.

Broad Absorption Line Quasars (BAL QSOs) have been found to be associated with extremely compact radio sources. These reduced dimensions can be either due to projection effects or these objects might actually be intrinsically small. Exploring these two hypotheses is important to understand the nature and origin of the BAL phenomenon because orientation effects are an important discriminant between the different models proposed to explain this phenomenon. In this work we present VLBA observations of 5 BAL QSOs and discuss their pc-scale morphology.

We aim to obtain a complete sample of redshift > 3.6 radio QSOs from FIRST sources having star-like counterparts in the SDSS DR5 photometric survey (r<=20.2). We found that simple supervised neural networks, trained on sources with SDSS spectra, and using optical photometry and radio data, are very effective for identifying high-z QSOs without spectra. The technique yields a completeness of 96 per cent and an efficiency of 62 per cent. Applying the trained networks to 4415 sources without DR5 spectra we found 58 z>=3.6 QSO candidates. We obtained spectra of 27 of them, and 17 are confirmed as high-z QSOs. Spectra of 13 additional candidates from the literature and from SDSS DR6 revealed 7 more z>=3.6 QSOs, giving and overall efficiency of 60 per cent. None of the non-candidates with spectra from NED or DR6 is a z>=3.6 QSO, consistently with a high completeness. The initial sample of z>=3.6 QSOs is increased from 52 to 76, i.e. by a factor 1.46. From the new identifications and candidates we estimate an incompleteness of SDSS for the spectroscopic classification of FIRST 3.6<=z<=4.6 QSOs of 15 percent for r<=20.2.

We present multi-frequency observations of a sample of 15 radio-emitting Broad Absorption Line Quasars (BAL QSOs), covering a spectral range between 74 MHz and 43 GHz. They display mostly convex radio spectra which typically peak at about 1-5 GHz (in the observer's rest-frame), flatten at MHz frequencies, probably due to synchrotron self-absorption, and become steeper at high frequencies, i.e., >~ 20 GHz. VLA 22-GHz maps (HPBW ~ 80 mas) show unresolved or very compact sources, with linear projected sizes of <= 1 kpc. About 2/3 of the sample look unpolarised or weakly polarised at 8.4 GHz, frequency in which reasonable upper limits could be obtained for polarised intensity. Statistical comparisons have been made between the spectral index distributions of samples of BAL and non-BAL QSOs, both in the observed and the rest-frame, finding steeper spectra among non-BAL QSOs. However constraining this comparison to compact sources results in no significant differences between both distributions. This comparison is consistent with BAL QSOs not being oriented along a particular line of sight. In addition, our analysis of the spectral shape, variability and polarisation properties shows that radio BAL QSOs share several properties common to young radio sources like Compact Steep Spectrum (CSS) or Gigahertz-Peaked Spectrum (GPS) sources.

We selected from VLA-FIRST a sample of 94 objects starlike in SDSSS, and with APM colour O-E>2, i.e. consistent with their being high-z QSOs. 78 candidates were classified spectroscopically from published data (mainly SDSS) or observations reported here. The fractions of QSOs (51/78) and z > 3 QSOs (23/78) are comparable to those found in other photometric searches for high-z QSOs. We confirm that O-E>2 ensures inclusion of all QSOs with 3.7 < z < 4.4. The fraction of broad absorption line (BAL) QSOs for 2 < z < 4.4 is 27+-10 per cent (7/26), and the estimated BAL fraction for radio loud (RL) QSOs is at least as high as for optically selected QSOs (about 13 per cent). The high BAL fraction and the high fraction of LoBALs in our sample are likely due to the red colour selection. The space density of RL QSOs for 3.7 < z < 4.4, MAB (1450 A) < -26.6 and P(1.4 GHz) > 10^25.7 W Hz^(-1) is 1.7+-0.6 Gpc^(-3). Adopting a RL fraction 13.4+-3 per cent, this corresponds to rho = 12.5+-5.6 Gpc^(-3), in good agreement with the SDSS QSO luminosity function in Fan et al. (2001). We note the unusual QSO FIRST 1413+4505 (z=3.11), which shows strong associated Lyalpha absorption and an extreme observed luminosity, L about 2 x 10^(15) solar luminosities.

We present observations of the most radio-luminous BAL quasar known, 1624+3758, at redshift z = 3.377. The quasar has several unusual properties: (1) The FeII UV191 1787-A emission line is very prominent. (2) The BAL trough (BALnicity index 2990 km/s) is detached by 21000 km/s and extends to velocity v = -29000 km/s. There are additional intrinsic absorbers at -1900 and -2800 km/s. (3) The radio rotation measure of the quasar, 18350 rad/m^2, is the second-highest known. The radio luminosity is P(1.4GHz) = 4.3 x 10^27 W/Hz (H0 = 50 km/s/Mpc, q0 = 0.5), radio loudness R* = 260. The radio source is compact (<~ 2.8 kpc) and the radio spectrum is GHz-peaked, consistent with it being relatively young. The width of the CIV emission line, in conjunction with the total optical luminosity, implies black-hole mass ~ 10^9 M(sun), L/L(Eddington) ~ 2. The high Eddington ratio, and the radio-loudness, place this quasar in one corner of Boroson's (2002) 2-component scheme for the classification of AGN, implying a very high accretion rate, and this may account for some of the unusual observed properties. A complex mini-BAL absorber at v = -3000 km/s is detected in each of CIV, NV and OVI. The blue and red components of the CIV doublet happen to be unblended, allowing both the covering factor and optical depth to be determined as a function of velocity. Variation of covering factor with velocity dominates the form of the mini-BAL...

The application of supervised artificial neural networks (ANNs) for quasar selection from combined radio and optical surveys with photometric and morphological data is investigated, using the list of candidates and their classification from White et al. (2000). Seven input parameters and one output, evaluated to 1 for quasars and 0 for nonquasars, were used, with architectures 7:1 and 7:2:1. Both models were trained on samples of ~800 sources and yielded similar performance on independent test samples, with reliability as large as 87% at 80% completeness. For comparison the quasar fraction from the original candidate list was 56%. The accuracy is similar to that found by White et al. using supervised learning with oblique decision trees. This performance probably approaches the maximum value achievable with this database. Probabilities for the 98 candidates without spectroscopic classification in White et al. are presented and compared with the results from their work, showing a good agreement. This is the first analysis of the performance of ANNs for the selection of quasars. Our work shows that ANNs provide a promising technique for the selection of specific object types in astronomical databases.

We obtained spectra of 60 red, starlike objects (E< 18.8) identified with FIRST radio sources, S_{1.4GHz} > 1 mJy. Eight are QSOs with redshift z> 3.6.Combined with our pilot search (Benn et al 2002), our sample of 121 candidates yields a total of 18 z > 3.6 QSOs (10 of these with z > 4.0). 8% of candidates with S_{1.4GHz}< 10 mJy, and 37% of candidates with S_{1.4GHz}> 10 mJy are QSOs with z > 3.6. The surface density of E < 18.8, S_{1.4GHz} > 1mJy, z> 4 QSOs is 0.003 deg^{-2}. This is currently the only well-defined sample of radio-loud QSOs at z ~ 4 selected independently of radio spectral index. The QSOs are highly luminous in the optical (8 have M_B < -28, q_0 = 0.5, H_0 = 50 kms^{-1}Mpc^{-1}). The SEDs are as varied as those seen in optical searches for high-redshift QSOs, but the fraction of objects with weak (strongly self-absorbed) Ly alpha emission is marginally higher (3 out of 18) than for high-redshift QSOs from SDSS (5 out of 96).

We present new XMM-Newton X-ray observations of the z=1.246 narrow-line radioquasar RX J1011.2+5545 serendipitously discovered by ROSAT. The flat X-ray spectrum previously measured by ROSAT and ASCA is shown to be the result of a steep Gamma~1.8 power law spectrum seen through a moderate intrinsic absorbing column NH~4E21 cm^-2. The position of the X-ray source is entirely coincident with the nucleus of the radio source that we have resolved in new sensitive VLA observations at 3.6 and 6 cm, implying that scattering in the radio lobes is not responsible for the bulk of X-ray emission. In the EPIC pn image, a faint patch of X-ray emission is apparent 14'' to the NE of the main X-ray source. The former is positionally coincident with an apparently extended optical object with R~21.9, but there is no associated radio emission, thus ruling out the possibility that this represents a hotspot in a jet emanating from the primary X-ray source. No reflection features are detected in the X-ray spectrum of the narrow-line radioquasar, although an Fe line with equivalent width of up to 600 eV cannot be ruled out.

High-redshift clusters of galaxies are amongst the largest cosmic structures. Their properties and evolution are key ingredients to our understanding of cosmology: to study the growth of structure from the inhomogeneities of the cosmic microwave background; the processes of galaxy formation, evolution, and differentiation; and to measure the cosmological parameters (through their interaction with the geometry of the universe, the age estimates of their component galaxies, or the measurement of the amount of matter locked in their potential wells). However, not much is yet known about the properties of clusters at redshifts of cosmological interest. We propose here a radically new method to study large samples of cluster galaxies using microslits to perform spectroscopy of huge numbers of objects in single fields in a narrow spectral range-chosen to fit an emission line at the cluster redshift. Our objective is to obtain spectroscopy in a very restricted wavelength range (~100 A in width) of several thousands of objects for each single 8x8 square arcmin field. Approximately 100 of them will be identified as cluster emission-line objects and will yield basic measurements of the dynamics and the star formation in the cluster (that figure applies to a cluster at z~0.50, and becomes ~40 and ~20 for clusters at z~0.75 and z~1.00 respectively). This is a pioneering approach that, once proven, will be followed in combination with photometric redshift techniques and applied to other astrophysical problems.

We define a new complete sample of 13 optically-luminous radio quasars M_AB(1450 Angstrom) < -26.9 mag and log P_1.4 GHz(W/Hz) > 25.7 with redshift 3.8 < z < 4.5, obtained by cross-correlating the FIRST radio survey and the APM catalogue of POSS-I. We measure the space density to be 1.0 +/- 0.3 /Gpc^3, a factor 1.9 +/- 0.7 smaller than the space density of similar quasars at z=2. Using a new measurement of the radio-loud fraction of quasars we find that at z=4 the total space density of quasars with M_AB(1450 Angstrom) < -26.9 is 7.4 +/- 2.6/Gpc^3. This is a factor 1.8 +/- 0.8 less than the space density at z=2, found by the 2dF quasar survey. This (z=2)/(z=4) ratio, consistent with that of the radio-loud quasars, is significantly different from the ratio of about 10 found for samples including lower-luminosity quasars. This suggests that the decline of the space density beyond z=2 is slower for optically-luminous quasars than for less-luminous ones.

We present results from a deep mid-infrared survey of the Hubble Deep Field South (HDF-S) region performed at 7 and 15 micron with the CAM instrument on board the Infrared Space Observatory (ISO). The final map in each band was constructed by the coaddition of four independent rasters, registered using bright sources securely detected in all rasters, with the absolute astrometry being defined by a radio source detected at both 7 and 15 micron. We sought detections of bright sources in a circular region of radius 2.5 arcmin at the centre of each map, in a manner that simulations indicated would produce highly reliable and complete source catalogues using simple selection criteria. Merging source lists in the two bands yielded a catalogue of 35 distinct sources, which we calibrated photometrically using photospheric models of late-type stars detected in our data. We present extragalactic source count results in both bands, and discuss the constraints they impose on models of galaxy evolution models, given the volume of space sampled by this galaxy population.

We present results from a deep mid-IR survey of the Hubble Deep Field South (HDF-S) region performed at 7 and 15um with the CAM instrument on board ISO. We found reliable optical/near-IR associations for 32 of the 35 sources detected in this field by Oliver et al. (2002, Paper I): eight of them were identified as stars, one is definitely an AGN, a second seems likely to be an AGN, too, while the remaining 22 appear to be normal spiral or starburst galaxies. Using model spectral energy distributions (SEDs) of similar galaxies, we compare methods for estimating the star formation rates (SFRs) in these objects, finding that an estimator based on integrated (3-1000um) IR luminosity reproduces the model SFRs best. Applying this estimator to model fits to the SEDs of our 22 spiral and starburst galaxies, we find that they are forming stars at rates of ~1-100 M_sol/yr, with a median value of ~40M_sol/yr, assuming an Einstein - de Sitter universe with a Hubble constant of 50 km/s/Mpc, and star formation taking place according to a Salpeter (1955) IMF across the mass range 0.1-100M_sol. We split the redshift range 0.0<z<0.6 into two equal-volume bins to compute raw estimates of the star formation rate density contributed by these sources, assuming the same cosmology and IMF as above and computing errors based on estimated uncertainties in the SFRs of individual galaxies. We compare these results with other estimates of the SFR density made with the same assumptions, showing them to be consistent with the results of Flores et al. (1999) from their ISO survey of the CFRS 1415+52 field. However, the relatively small volume of our survey means that our SFR density estimates suffer from a large sampling variance, implying that our results, by themselves, do not place tight constraints on the global mean SFR density.

UBVR photometry of a sample of B3-VLA radio quasars, about 80 per cent complete, is used to analyse their spectral energy distribution (SED). The SEDs are generally well fitted with power-laws, with an average slope alpha=-0.39 (S_nu propto nu^alpha). Two quasars appear clearly differenciated, exhibiting redder colours that the rest, and they have redshifts z=0.50 and 1.12. Broad-band composite SEDs in the range 1300-4500 AA were obtained from the remaining quasars and they show the CIV1549 line and a break at around 3000 A, where the SED changes from alpha_blue=0.11+-0.16 at lambda>3000 A to alpha_UV=-0.66+-0.15 at lambda<3000 A. The continuum shape depends on redshift. For the quasars with z<1.2 we find alpha_blue=0.21+-0.16 and alpha_UV=-0.87+-0.20, and for z>1.2 alpha_UV is more flat, -0.48+-0.12. A similar trend is found between alpha_UV and luminosity at 2400 A, L_2400, with luminous quasars exhibiting a bluer spectrum. In addition, an intrinsic correlation is found between L_2400 and radio power at 408 MHz. The correlations alpha_UV-z, alpha_UV-L_2400 and L_2400-z are consistent with accretion disc models with approximately constant black hole mass and accretion rates increasing with redshift. If the trends L_2400-z and P_408-z are predominantly related to the radio-flux limits of the sample, only one of the correlations alpha_UV-L_2400 or alpha_UV-z needs to be intrinsic.

We have discovered an obscured active galaxy at redshift z = 1.246 identified with the ROSAT X-ray source RX J1011.2+5545. We report on multiwavelength observations of this source and discuss its X-ray, optical and radio properties. This is the first X-ray selected, obscured active galaxy at high redshift to be shown to be radio-loud, with a radio counterpart exhibiting a classical double-lobe morphology.

We present K-band imaging and photometry of a sample of 52 radio loud quasars (RQs) selected from the B3 survey with flux densities above 0.5 Jy at 408 MHz. The optical completeness of the sample is 90% and the quasars cover the redshift range 0.4 - 2.3. For ~57% of the sources for which the quality of the images allowed a detailed morphological study (16/28) resolved extended emission was detected around the QSO, and its K flux was measured. Interpreting this ``fuzz'' as starlight emission from the host galaxy, its location on the K-z plane at z<1 is consistent with radio quasars being hosted by galaxies similar to radio galaxies (RGs) or giant ellipticals (gEs). At higher redshifts the detected host galaxies of RQs are more luminous than typical RGs and gEs, although some weak detections or upper limits are consistent with a similar fraction of RQs being hosted by galaxies with the expected luminosities for RGs or gEs. We found a significant correlation between radio power and nuclear infrared luminosity indicating a direct link between the radio synchrotron emission and the nuclear emission in K. This correlation is more tight for the steep-spectrum sources (99.97% significance). In addition, a trend is found between radio power and infrared luminosity of the host galaxy (or mass), in the sense that the most powerful quasars inhabit the most luminous galaxies.

Webster et al (1995) claimed that up to 80% of QSOs may be obscured by dust. They inferred the presence of this dust from the remarkably broad range of B-K optical-infrared colours of a sample of flat-spectrum PKS radio QSOs. If such dust is typical of QSOs, it will have rendered invisible most of those which would otherwise been have detected by optical surveys. We used the William Herschel Telescope on La Palma to obtain K infrared images of 54 B3 radio quasars selected at low frequency (mainly steep-spectrum), and we find that although several have very red optical-infrared colours, most of these can be attributed to an excess of light in K rather than a dust-induced deficit in B. We present evidence that some of the infrared excess comes from the light of stars in the host galaxy (some, as previously suggested, comes from synchrotron radiation associated with flat-spectrum radio sources). The B-K colours of the B3 QSOs provide no evidence for a large reddened population. Either the Webster et al QSOs are atypical in having such large extinctions, or their reddening is not due to dust; either way, the broad range of their B-K colours does not provide evidence that a large fraction of QSOs has been missed from optical surveys.

We present the optical and UV emission line properties of 160 X-ray selected AGN taken from the RIXOS survey (including Halpha, Hbeta, [OIII]5007, MgII2798 and CIII]1909). This sample is believed to contain a mixture of absorbed and unabsorbed objects, with column densities up to 4e21 cm-2. Although the distribution of the [OIII] EW for the RIXOS AGN is typical of optically selected samples, the Balmer line EWs are relatively low. This is consistent with the presence of a dust absorber between the broad and narrow line regions (eg. a molecular torus), and intrinsically weak optical line emission. We find Baldwin effects in CIII] and MgII, and a positive response of the MgII line to its ionizing continuum. There is a strong correlation between the EW and FWHM of MgII, which may be similar to that seen in other samples for Hbeta. We demonstrate that this is consistent with models which suggest two line-emitting zones, a `very broad line region' (VBLR) and an `intermediate line region' (ILR). The correlation between EW and FWHM in MgII may be a physical characteristic of the ILR or it may reflect a geometric dependence. We found no correlation between the Hbeta FWHM and the slope of the X-ray spectrum, however this may be due to the effects of dust absorption which suppresses the broad Hbeta component, masking any relationship. The Halpha FWHM does tend to be narrow when alpha_X is soft, and broadens as alpha_X hardens, although the formal probability for this correlation is low (91 per cent). If the distribution of alpha_X in the RIXOS sample reflects the level of intrinsic absorption in these AGN, the data suggest a possible link between the velocity of the Balmer line-emitting region and the amount of absorbing material beyond.

We present measurements of the optical and X-ray continua of 108 AGN (Seyfert 1s and quasars) from the Rosat International X-ray/Optical Survey (RIXOS). The sample covers a wide range in redshift (0<z<3.3), in X-ray spectral slope (-1.5 <ax<2.6) and in optical-to-X-ray ratio (0.4$<aox<1.5). A correlation is found between ax and aox; similar correlations have recently been reported in other X-ray and optical samples. We also identify previously unreported relationships between the optical slope (aopt) and ax (particularly at high redshifts) and between aopt and aox. These trends show that the overall optical-to-X-ray conti nuum changes from convex to concave as ax hardens, demonstrating a strong beha vioural link between the optical/UV big blue bump (BBB) and the soft X-ray ex cess, which is consistent with them being part of the same spectral component. By constructing models of the optical-to-X-ray continuum, we demonstrate that the observed correlations are consistent with an intrinsic spectrum which is ab sorbed through different amounts of cold gas and dust. The intrinsic spectrum is the sum of an optical-to-soft X-ray `big bump' component and an ax=1 power law; the column density of the cold gas ranges from 0 to ~4E21 cm-2 and the dust-to-gas ratio is assumed to be Galactic. The `big bump' may be represented by a T~1E6 K thermal bremsstrahlung or an accretion disk with a surrounding hot corona. The scatter in the data can accommodate a wide range in big bump tempe rature (or black hole mass) and strength. A source for the absorbing gas may be the dusty, molecular torus which lies beyond the broad line-emitting regions, although with a much lower column density than observed in Seyfert 2 galaxies. Alternatively, it may be the bulge of a spiral host galaxy or an elliptical host galaxy.

A sample of 198 soft X--ray selected active galactic nuclei (AGN) from the ROSAT International X--ray Optical Survey (RIXOS), is used to investigate the X--ray luminosity function and its evolution. RIXOS, with a flux limit of 3E-14 erg s-1 cm-2 (0.5 to 2.0 keV), samples a broad range in redshift over 20 deg^2 of sky, and is almost completely identified; it is used in combination with the Einstein Extended Medium Sensitivity Survey (EMSS), to give a total sample of over 600 AGN. We find the evolution of AGN with redshift to be consistent with pure luminosity evolution (PLE) models in which the rate of evolution slows markedly or stops at high redshifts z>1.8. We find that this result is not affected by the inclusion, or exclusion, of narrow emission line galaxies at low redshift in the RIXOS and EMSS samples, and is insensitive to uncertainties in the conversion between flux values measured with ROSAT and Einstein. We confirm, using a model independent Ve/Va test, that our survey is consistent with no evolution at high redshifts.

The influence of a foreground QSO on the lyman a forest of another QSO with higher redshift has been investigated by analyzing the spectra of three such objects at redshifts $z=2 - 2.7$. This influence is not contaminated by any projection effects, as opposed to the inverse effect along the line of sight, where incomplete coverage of the QSO continuum emitting region by the lyman a clouds could contribute to the relative lack of lines. Our results are consistent with the existence of a proximity effect due to the foreground QSO, but due to its weakness we can only reject the absence of such effect at 1 sigma level. By modelling this proximity effect assuming that lyman a clouds are low-density highly ionized objects we find that the best value for the UV photoinizing intensity at those redshifts is 10^{-20.5}(...) at the Lyman limit which is consistent with previous estimates of the background based on the inverse effect. We also find an absolute lower limit (at 95% confidence) to the UV intensity at the level of 10^{-21.8}(...) which means the rejection of a number of models for the UV background.

We present the first moderate resolution (approx. 40--120 km/s) spectroscopic observations of the bright (V<16.7) high-redshift QSOs Q1107+487 (z_{em} = 2.965) and Q1442+295 (z_{em} = 2.669) (Sanduleak & Pesch 1989). The relatively high signal to noise reached in the spectra along with an extensive wavelength coverage of the Lyman alpha and the Lyman beta forest allowed us to obtain, through profile fitting, column densities and Doppler parameters of the Lyman clouds towards these QSOs. The spectral coverage of regions longward of the Lyman alpha emission line of the QSOs at the expected wavelengths of CIV 1548,1550 at the redshifts of the Lyman alpha forest allowed us to identify some heavy element absorption systems (hereafter HEASs) towards these QSOs. We have found no Lyman absorption system or HEAS towards these QSOs for which our data allow a deuterium measurement or to provide an interesting upper limit for the D/H ratio. The reason for this is that the Lyman lines with high column density detected towards these QSOs belong to absorption systems showing velocity structure.

We present the results of our study on three close pairs of QSOs. Our results are consistent with the existence of a proximity effect due to the foreground QSO, but due to its weakness we can only reject the absence of such effect at approx. 1 sigma level. By modelling this proximity effect in terms of a simple photoionisation model, we find the best value for the UV ionizing background to be approx. 10**(-20.5) erg cm**(-2) s**(-1) Hz**(-1) srad**(-1) at the Lyman limit, and an absolute lower limit (95\% confidence) of 10**(-21.8) in the same units. This lower limit rejects a number of models for the UV background where it is mostly contributed by QSOs and absorption by Lyman limit systems is taken into account.