Our ideas about the surface brightness distribution of galaxies has changed greatly since 1970. I contrast the view at that time with our present view of the subject, and then briefly discuss some topics in the studies of Low Surface Brightness Galaxies (LSBG) that I think are particularly interesting: the Tully-Fisher relation, LSBG as systems of high angular momentum, LSBG in clusters of galaxies, and the potential impact of the HIPASS survey.

I have been asked to review some of the arguments for and against the existence of a significant population of Hidden Galaxies, and to concentrate on the distant past and the possible future of this subject, leaving more recent observations to others. In particular I ask how should we correct for selection effects and how are we to resolve contradictory evidence?

Our view of the properties of galaxies is strongly affected by the way in which we survey for them. I discuss some aspects of selection effects and methods to compensate for them. One result is an estimate of the surface brightness distribution. I believe this is progress, but considerable uncertainty remains.

Surveys in the 21cm line of neutral hydrogen are testing the completeness of the catalogs of nearby galaxies. The remarkable observational fact is that the potential wells that confine gas to sufficient density that it can remain neutral in the face of ionizing radiation also provide sites for star formation, so that there are no known cases of neutral intergalactic clouds without associated star light.

Here we describe results of our attempt to determine what types of galaxies are responsible for the Lyα forest absorption systems, based on our ongoing imaging and spectroscopic survey of faint galaxies in fields of HST spectroscopic target QSOs. Our primary conclusions are that the bulk of the Lyα forest arises in more or less normal galaxies (that span the normal range of luminosity and morphology) and that any “unseen” low surface brightness galaxies are unlikely to contribute significantly to the luminosity density of the universe.

The debate about the slope and amplitude of the galaxy luminosity function at the faint end is discussed w.r.t. faint galaxies in large surveys, in particular the second CfA (CfA2) and the Las Campanas (LCRS) redshift surveys. Large surveys are necessary to determine the statistics of rare objects or objects that can only be seen out to limited volumes. Both surveys show excesses of faint galaxies over Schechter function fits, but the parent sample for the LCRS survey generally does not contain large or low surface brightness galaxies which do appear in the CfA2 survey. The objects that comprise the relatively large excess of faint galaxies in the CfA2 survey are shown to be primarily of low surface brightness and late morphological type and are generally emission line galaxies. Galaxy samples constructed like the LCRS will generally always be deficient in low luminosity galaxies and thus are not useful for constraining the faint end of the galaxy luminosity function.

The local space density of galaxies as a function of their basic structural parameters -like luminosity, surface brightness and scalesize-is still poorly known. Our poor knowledge is mainly the result of strong selection biases against low surface brightness and small scalesize galaxies in any optically selected sample. We show that in order to correct for selection biases one has to obtain accurate surface photometry and distance estimates for a large (≳ 1000) sample of galaxies. We derive bivariate space density distributions in the (scalesize, surface brightness)-plane and the (luminosity, scalesize)-plane for a sample of ~1000 local Sb-Sdm spiral galaxies. We present a parameterization of these bivariate distributions, based on a Schechter type luminosity function and a log-normal scalesize distribution at a given luminosity. We show how surface brightness limits and (1+z)4 cosmological redshift dimming can influence interpretation of luminosity function determinations and deep galaxy counts.

Recent results on the determination of the shape of the faint end of the galaxy luminosity function in rich clusters are discussed. There is increasing evidence that in many cases the faint end of the function is steep, indicating a large population of dwarf, possibly low surface-brightness, galaxies. In addition, the magnitude at which the turn-up appears is approximately constant with richness and distance. However, it is clear that not all clusters show such a feature.

We present preliminary measurements of the bJ and K-band luminosity functions (LFs) of field galaxies obtained from optical and K-band imaging of a sample of galaxies selected from the Stromlo-APM Redshift Survey. The bJ LF is consistent with that previously published from photographic data. The K-band LF has been estimated over a range of 12 magnitudes and is reasonably well fit by a Schechter function with faint-end slope α = −1.2.

We examine the constraints that can be placed on the space density of low-surface-brightness galaxies from deep HST images. Such images, while covering only a small solid angle, provide enough depth and spatial resolution to detect LSB galaxies at moderate redshift and distinguish them from galaxies of higher surface brightness.

We consider five simple models of the non-evolving or slowly-evolving population of LSB galaxies, motivated by various discussions in the recent literature. The basic results are (1) models with a large space-density of giant LSB galaxies at moderate redshift do not look like the real world and, (2) models with a large space-density of dwarf LSB galaxies are consistent with HST data (that is, they do not produce more faint LSB galaxies per unit solid angle than are detected at magnitudes I ≳ 23), but these LSB dwarf galaxies do not contribute much to faint galaxy counts unless they formed their stars in a rapid burst.

We describe a new, deep photographic survey of the Virgo Cluster which uses multiple exposures on Tech Pan film with the United Kingdom Schmidt Telescope to probe the dwarf population to fainter surface brightness limits than previous surveys. We have identified galaxies having sizes (≥ 3 arcsec scale length) and surface brightnesses (≤ 24.5 R mag arcsec−2) characteristic of those expected for dwarf spheroidal galaxies in the cluster. The survey is providing substantial samples of extremely low luminosity galaxies outside the environment of the Local Group and nearby groups for the first time. An initial study of two small areas has found dwarf spheroidal candidates in large numbers (500 deg−2) which indicate a steep, continuously rising luminosity function at low luminosities.

A complete sample of spiral galaxies in the Ursa Major cluster is imaged at various optical wavelengths and in the Near-Infrared K′-band. HI rotation curves were obtained for all gas rich systems. The Near-Infrared surface brightness distribution of disk galaxies turns out to be bimodal; galaxies avoid a domain around mag/arcsec2. This bimodality is particularly striking when only the more isolated, non-interacting systems are considered. The Luminosity Function of the HSB family of galaxies is truncated well above the completion limit while the Luminosity Function of the LSB family is still sharply rising at our limiting magnitude. Near-Infrared mass-to-light ratios suggest that HSB galaxies are close to a kinematic maximum-disk situation while LSB galaxies are dark matter dominated at all radii. Assuming equal Near-Infrared mass-to-light ratios for both HSB and LSB systems, we find that the gap in the surface brightness distribution corresponds to a situation in which the baryonic mass is marginally self-gravitating. We finally conclude that the luminosity-line width relation is a fundamental correlation between the amount and distribution of dark matter mass and the total luminosity, regardless of how the luminous mass is distributed within the dark mater halo.

The systematic mapping of obscured and optically invisible galaxies behind the Milky Way through complementary surveys are important in arriving at the whole-sky distribution of complete galaxy samples and therewith for our understanding of the dynamics in the local Universe. In this paper, a status report is given of the various deep optical, near infrared (NIR), and systematic blind H I-surveys in the Zone of Avoidance, including a discussion on the limitations and selection effects inherent to the different multi-wavelength surveys and first results.

By applying a digital filtering technique to 1550 deg2 of the POSS-II in the vicinity of M31, we found two previously unidentified very low surface brightness dwarf galaxies which we designate And V and VI. Follow-up imaging with the KPNO 4-m telescope resolved these into stars easily. The V- and I- band images of And V indicate a distance similar to that of M31, and <[Fe/H]> ~ −1.5. All evidence strongly supports its classification as a dwarf spheroidal companion to M31. Data for And VI are being analyzed, but preliminary indications support a similar conclusion. Our search for more dwarfs, including follow-up observations of numerous candidates found via digital filtering, is incomplete; thus, further identifications may be forthcoming.

The Fornax Spectroscopic Survey is a large optical spectroscopic survey of all 14 000 objects with 16.5 < BJ < 19.7 in a 12 deg2 area of sky centered on the Fornax Cluster. We are using the 400-fibre Two Degree Field spectrograph on the Anglo-Australian Telescope: the multiplex advantage of this system allows us to observe objects conventionally classified as “stars” as well as “galaxies”. This is the only way to minimise selection effects caused by image classification or assessing cluster membership.

In this paper we present the first measurements of low surface brightness (LSB) galaxies we have detected both in the Fornax Cluster and among the background field galaxies. The new cluster members include some very low luminosity (MB ≈ −11.5 mag) dwarf ellipticals, whereas the background LSB galaxies are luminous (−19.6 < MB < −17.0 mag) disk-like galaxies.

We are using the 2dF spectrograph to make a survey of all objects (“stars” and “galaxies”) in a 12 deg2 region towards the Fornax cluster. We have discovered a population of compact emission-line galaxies unresolved on photographic sky survey plates and therefore missing in most galaxy surveys based on such material. These galaxies are as luminous as normal field galaxies. Using Hα to estimate star formation they contribute at least an additional 5% to the local star formation rate.

Given the difficulties in detecting even a single LSB, surveying statistically significant numbers of LSBs presents a daunting task. Large, systematic surveys with well understood selection criteria are necessary for assessing the full cosmological significance of the LSB population. Here I briefly review some of the progress which has been made in the last ten years, and suggest strategies which may prove fruitful in the future.

In this paper we compare the radial distribution of Fornax cluster galaxies selected at different surface brightness values. The bright galaxy (RC3) sample surface brightness distribution peaks at a μBx = 21.5, the Low Surface Brightness (LSB) sample at μBx = 23.7 and the very Low Surface Brigthness (VLSB) sample at μBx = 25.7. The bright galaxy surface density decreases with an exponential scale length of α = 0.5°, the LSB galaxies with α = 1.3° and the VLSB galaxies with α = 2.0°. Thus the LSB and VLSB populations define a mass scale size some 2.5 and 4 times larger than the bright galaxies respectively. The contribution of the VLSB population to the total cluster luminosity is comparible to that of the brighter galaxies. If their mass to light ratios are significantly larger than than that of the bright galaxies they will dominate the mass and possibly account for the missing mass in clusters.

We have performed a digital survey for Low Surface Brightness (LSB) galaxies in the spiral-rich Cancer and Pegasus clusters as well as the low density regime defined by the Great Wall. A total of 127 galaxies were found with μB(0) > 22.0 mag arcsec‒2, 119 of which were previously unidentified.

Consistent with other surveys, we find a significant number of galaxies with μB(0) > 23.0 mag arcsec‒2 which suggests that the space density of galaxies as a function of μB(0) is not strongly peaked. To more rigorously test this hypothesis we compare the actual surface brightness distribution from our survey with that from two different types of Monte-Carlo based sky images and show that it is not possible to distinguish between the flat distribution and the Gaussian one as the proper description of the underlying surface brightness distribution for this survey beyond 24.0 mag arcsec−2.

The colors of the survey galaxies range continuously from very blue to the first discovery of very red LSB galaxies. It also includes a group of old galaxies which show evidence for recent star formation. This continuous range of colors clearly shows that LSB galaxies at the present epoch define a wide range of evolutionary states.