This paper is part of a more extensive study of the large voids in spatial volumes both wide and deep. Three samples of optical tracers with spectroscopically measured redshifts in a volume with b≥30° and z≤0.14 have been extracted from the catalogue of Lebedev and Lebedeva (1986) and the NASA/IPAC Extragalactic Database (NED):

1. (1) a homogeneous sample of 277 rich Abell clusters

2. (2) an inhomogeneous sample of 969 objects: rich and poor Abell clusters, Zwicky clusters, other clusters, and groups

3. (3) a sample of 18 623 NED galaxies.

We performed two surveys at the Calar Alto Observatory to identify low luminosity galaxies (LLG) in 4 fields towards nearby voids. While the central parts of the voids remain empty, we found about 20 very isolated (nearest neighbour distances of DNN ≥ 4Mpc) galaxies along the rims of some (but not all) voids (Hopp et al. 1995, Hopp & Kuhn 1995, Kuhn et al. 1996, Popescu et al. 1996). Many of them are dwarfs, a few are giants. CCD surface photometry revealed normal properties (Vennik et al. 1996), HI-observations show a tendency that the isolated dwarfs are overabundant in neutral gas compared to sheet and cluster galaxies of the same luminosity (Huchtmeier et al. 1996).

Log N - log S diagrams are being used as powerful diagnostic tools to probe evolutionary properties of different extragalactic populations. In the optical/near-infrared regime the discrepancy between near-infrared number counts which follow theoretical predictions and counts in the B band which show an excess density revealed a new population of galaxies. The nature of this population of faint blue galaxies is still unknown.

The Hubble (magnitude-redshift) diagram for brightest cluster galaxies (BCGs) is a classic cosmological tool, widely studied because of the remarkably small dispersion (∼ 0.3 mag) in the absolute optical magnitudes of low redshift BCGs (Postman and Lauer 1995). Extending the BCG Hubble diagram to higher redshifts would greatly enhance its role as a cosmological probe, but this has been frustrated by several technical problems:

1. – the conventional means of cluster selection in the optical become increasingly compromised by projection effects at z > 0.1

2. – at higher redshifts the interpretation of optical magnitudes becomes increasingly complicated by the effects of possible star formation.

The Shapley Concentration is a prominent supercluster in the southern sky. It is interesting not only for its relevance in the peculiar motion problem (it seems to be responsible of ∼ 30% of the acceleration acting on the Local Group of galaxies), but also because it is the most remarkable feature which appears studying the distribution of the Abell-ACO clusters of galaxies: Zucca et al. (1993) found that at every density contrast the Shapley Concentration stands out as the richest supercluster in the sky. In particular, the central part of this supercluster is dominated by a complex containing the three ACO clusters A3556, A3558 and A3562 and the poor cluster SC1329-314, which form a structure elongated ∼ 3° along the East-West direction. We are carrying on a long term study of the Shapley Concentration in order to describe its dynamical state and to determine its mass. The project consists of redshift determinations (with the ESO telescopes at La Silla) for galaxies both in the clusters (Bardelli et al. 1994) and in the inter-cluster field of this supercluster, X-ray observations (ROSAT) of the hot gas in the clusters (Bardelli et al. 1996) and radio observations (ATCA, MOST and VLA, Venturi et al. 1996) of the radiogalaxies.

We have used the MX multifiber spectrometer on the Steward 2.3m measuring redshifts of up to 25 galaxies in each of about 90 Abell cluster fields with richness class R ≥ 1 and mag10 ≤ 16.8 (estimated z ≤ 0.12) and no more than one previously-measured redshift. This work has resulted in a deeper, more complete sample for two-point correlation and other studies of large-scale structure. To date, we have collected such data for 110 clusters (a few not in our sample). For most, we have seven or more cluster members with redshifts, enough to add significantly to the available sample of cluster velocity dispersions for other studies of cluster properties.

The ESO Slice Project (ESP) is a galaxy redshift survey we have recently completed as an ESO Key-Project over about 30 square degrees, in a region near the South Galactic Pole (Vettolani et al., submitted to A&A). The survey is nearly complete to the limiting magnitude bJ = 19.4 and consists of more than three thousands galaxies with reliable redshift determination.

Observations that radio-quiet QSOs exist in average galaxy cluster environments (Smith et al. 1995 and references therein) demonstrate that QSOs can be used to derive important information on the structure of the Universe at the largest scales. Previous studies of QSO clustering have been frustrated by the lack of large QSO redshift surveys. Although QSO clustering is detected in the largest existing QSO catalogues (see Shanks & Boyle 1994), it is difficult to place strong limits on the cosmological evolution of QSO clustering or the level of clustering at large scales (> 10h–1 Mpc) with current QSO catalogues.

In this paper we present the results of a redshift survey of IRAS galaxies behind the whole southern Milky Way region 210° < l < 360° at |b| < 15°. The galaxy candidates in the southern Milky Way were selected from IRAS Point Source Catalogue (IPSC) by applying flux density and infrared colour criteria (Strauss et al. 1990; Rowan-Robinson et al. 1991). The selected sample is flux limited with f60 > 0.6Jy. A visual search for galaxy-like objects have been carried out in UK-Schmidt Infrared and IIIaJ films in the selected IRAS positions. In spite of the large Galactic extinction in this region, 966 galaxy candidates have been identified (Yamada et al. 1993).

In two strips of 22° × 1° and 5° × 1° near the SGP Vettolani et al. have made a deep redshift survey as an ESO Key Project (the ESO Slice Project galaxy redshift survey). All the galaxies down to bJ ∼ 19.4 were observed with the OPTOPUS multi-fibre spectrograph on the 3.6 m telescope in La Silla, yielding 3348 redshifts.

To study galaxy populations and their evolution at the highest possible redshifts, a small area of the sky, the Hubble Deep Field (HDF) was imaged to an unprecedented sensitivity of R = 29.5 (Williams et al. 1996). As a complement to the HST observations, we have used the VLA at 8 GHz to image an area 5.′4 in diameter (FWHM) centered on the HDF to an rms sensitivity of 2 μJy. With a radio resolution of about 3″, we have 33 sources above 9.5 μJy, seven in the 4 arcmin2 HDF field of which six have clear optical IDs. There are an additional 12 IDs in the HST flanking fields. The optical counterparts of the radio sources are a mixture of ellipticals, spirals, and irregulars, consistent with earlier surveys of comparable depth (Windhorst et al. 1995). With a median redshift <z> ∼ 1, the radio galaxies we are sampling are somewhat more distant than the classical starbursting galaxies which dominate less sensitive radio surveys. Our HDF identifications are predominately with post-starburst galaxies, moderate power AGN, and blue irregulars (Fomalont et al. 1996).

The evolving class of objects responsible for the majority of the cosmic X-ray background (XRB) remains at large nearly three and a half decades after the discovery of the XRB. Surveys of sources selected on the basis of their X-ray properties alone provide an unbiased picture of the X-ray sky, but to date they have not been ideal for the discovery of rare types of X-ray sources at faint fluxes: large-area X-ray surveys have been restricted to bright sources, while deep X-ray surveys have been limited to very small patches of sky. X-ray selection coupled with another selection criterion, e.g., a radio or infrared detection, complements “pure” X-ray surveys by (1) permitting the exploration of large areas of sky to faint flux limits for types of extragalactic X-ray sources not well represented in other surveys, and (2) assisting the location of the optical counterparts of these X-ray sources. Using this approach, I have searched for new components of the XRB among the faintest X-ray sources detected by the Einstein Observatory.

Multiwavelength surveys present a variety of challenging statistical problems: raw data processing, source identification, source characterization and classification, and interrelations between multiwavelength properties. For these last two issues, we discuss the applicability of standard and new multivariate statistical techniques. Traditional methods such as ANOVA, principal components analysis, cluster analysis, and tests for multivariate linear hypotheses are underutilized in astronomy and can be very helpful. Newer statistical methods such as projection pursuit, multivariate splines, and visualization tools such as XGobi are briefly introduced. However, multivariate databases from astronomical surveys present significant challenges to the statistical community. These include treatments of heteroscedastic measurement errors, censoring and truncation due to flux limits, and parameter estimation for nonlinear astrophysical models.

The major photographic surveys of the sky (undertaken with the ESO, Palomar and UK Schmidt telescopes) still hold enormous potential regarding the information content of the original plate material. The problem to date has been to extract that information in a meaningful and quantitative way, hence giving rise to the large number of digitisation programmes currently underway and which are reported elsewhere in this volume.

The UK 1.2m Schmidt Telescope (UKST) at Siding Spring Observatory in Australia has been routinely using panchromatic Tech-Pan emulsion on film as a detector since 1992. Details of how the telescope was adapted to take film and how hypersensitization techniques have been modified for film have been described by Parker (1992). The film has a much finer grain than its equivalent spectroscopic emulsion on glass (IIIa-F) and consequently records images typically one magnitude fainter than IIIa-F when properly hypersensitized. The improved depth, resolution and low noise also permit better differentiation between stars and galaxies (see Parker et al. 1994). Consequently Tech-Pan film has become one of the most useful and commonly used emulsions at the UKST for a variety of passbands and may in time be used for a full or partial sky survey.

The next generation of spectroscopic surveys, both Galactic and extra-galactic (e.g., SDSS, 2dF), present the challenge of classifying spectra in an efficient and objective manner. The standard approach to this problem has been to visually classify spectra based on a number of spectral features (e.g., the equivalent widths of emission lines). The size of new spectral surveys (> 106 galaxies) and the desire to compare the luminosity and environments of galaxies with their spectral properties make these techniques infeasible. We describe here an automated classification scheme that is being developed for the SDSS.

The Guide Star photometric Catalog (GSPC-I; Lasker et al. 1988) is an all-sky set of photoelectrically determined BV sequences created to provide photometric calibrators for the Guide Star Catalog (Paper-I: Lasker et al. 1990, Paper-II: Russell et al. 1990, Paper-III: Jenkner et al. 1990). Although the GSPC-I has been the basis of preliminary photometric calibrations for the Digitized Sky Survey (DSS; Doggett et al. 1995), its relatively bright cutoff at about 15th magnitude limits its capability to support calibration of sky surveys, e.g., the new GSC-II (McLean et al. 1996, this volume, p. 431).

Over the last few years there has been considerable progress in developing successful algorithms for obtaining astrometric quality positions from Schmidt plates which compensate for deficiencies of the polynomial approach; the sub-plate method (Taff 1989), the mask method (Taff et al. 1990), the collocation method (Bucciarelli et al. 1993) the filter method (Röser et al. 1995) and a filter weighting according to the method of infinitely overlapping circles (Morrison et al. 1996b.) However, none of these nor any other studies have investigated the magnitude dependence of the position estimates outside the magnitude range of standard reference catalogs. Often Schmidt plates cover the magnitude range from 6m to 19m (for the GSC it is 6m to 15m). Presently available reference catalogs, however, have a limiting magnitude of V ≈ 10m. Therefore, no magnitude dependent term for fainter stars can be reliably found by reducing the measurements based only on comparisons with reference stars.

We attempted to quantify the magnitude-dependent systematics in a sample of Schmidt plates by comparison to positions from the Yale/San Juan Southern Proper Motion program which offers star positions and absolute proper motions down to B = 18 with a mean density of about 50 stars per square degree and a positional accuracy of 0.1″ (Platais et al. 1995).

Instrumental parameters and astrometric results from five telescopes are summarized in Table 1. The KPNO 0.9m has field-corrector optics. The CTIO 0.9m is a classical Cassegrain. The 4-meter telescopes both have doublet field-correctors. The USNO 0.2m (8-inch) astrograph has a 5-element lens which is designed for a 9° flat field of view for photographic plates. A bandpass of 570-650 nm is used at the USNO 0.2m while most frames with the other telescopes have been taken through a Gunn r (600-710 nm) filter. Stellar images on the CCD frames have been fitted with a 2-D circular symmetric Gaussian profile, giving the centroiding error σfit . Plots of σfit vs. instrumental magnitude look similar in shape for all telesopes. The asymptotic fit precision, σafp , is the limit in σfit achieved for bright stars, given in milli pixels (mpx) in Table 1. This centering error is overestimated by an amount depending on the deviation of the real image profile from the model function (Winter 1997).