Côté (1995) surveyed the nearby Centaurus A (Cen A) and Sculptor (Sci) group regions in the search for gas-rich, dwarf irregular galaxies (Irr) which are associated with these galaxy aggregates. 36 Irrs have been found and confirmed as new group members via 21cm/Hα redshifts. In finding this large number of Irrs, the question rises whether systems from the other dwarf galaxy family, the gas-poor dwarf ellipticals (dE), also exist in the two groups. So far nothing is known about dE membership but taking into account the morphology-density relation for dwarfs (Binggeli et al. 1990) a number ratio Irr/dE≈ 1 - 2 would be expected. Prompted by these “missing” dEs we started a large program to identify and study dEs in Cen A and Sci. The main scientific goals are (1) to find dE candidates and to establish group membership via velocity or distance and (2) to estimate the luminosity function (LF) of dwarf galaxies to the faintest possible luminosities. For the two groups, being respectively at a mean distance of 4.0 Mpc and 2.5 Mpc, we will be able to reach dwarf galaxies 4-5 magnitudes fainter than in the previous studies in the Virgo and Fornax clusters and the five groups Leo I, Dorado, NGC1400, NGC5044, and Antlia (Sandage et al. 1985, Ferguson &: Sandage 1990). We briefly outline here first results from the CenA group. All details about the complete survey including Scl will be presented in Jerjen et al. (1997)

We have carried out a large area survey for nearby LSB galaxies to try and answer the following questions:

1. 1. How are LSB galaxies distributed relative to the normal HSB galaxies? Do they have the same clustering properties suggesting that they trace the same mass distribution?

2. 2. What are the implications of the clustering properties on galaxy formation scenarios?

3. 3. Can they account for faint blue galaxies at higher redshifts?

4. 4. Is there any difference between the field and cluster LSB galaxies?

2. The Survey

We have devised an automated technique to pick out local LSB galaxies using 65 UKSTU photographic plates corresponding to 2187 sq.°. We have imposed stringent selection criteria in order to minimise the background contamination. Our galaxies have μ0 ≥ 22.5 B mag arcsec-2 α ≥ 3.0 arcseconds and BT ≤ 20.0. Our survey areas are confined to the Fornax Cluster, the NGC1400, Sculptor and Dorado Groups and the fields between them. The survey, which was carried out in three stages over a period of 2 years, has produced detections of 2435 LSB galaxies (32 galaxies per field). We have estimated the background contamination in three ways: first, by a limited redshift sample, secondly, by Ferguson’s (1989) classification on the Fornax plate and finally, by a numerical simulation of background galaxies. The first method produced an estimate of 14 galaxies per field, while the second and the third methods produced the same estimate of 19 galaxies.

Generally the density of the environment is important for the formation and evolution of galaxies: bright elliptical galaxies reside in the centers of galaxy clusters, whereas late type galaxies preferentially populate small groups, filaments or the outskirts of rich clusters of galaxies. Furthermore, fainter galaxies define smaller voids than bright galaxies. Our investigation of void catalogues in the Northern Local Void region shows that large voids are interlaced by systems of faint galaxies dividing them into smaller subvoids (Lindner et al., 1995). This void hierarchy is an important property of the Large Scale Structure in the nearby Universe and can help to solve problems concerning the distribution of dwarf galaxies and BCDGs.

NGC 595 is a giant Hɪɪ region located in the western part of the spiral galaxy M 33. It is the second in importance in this galaxy, after NGC 604. At 0.84 Mpc, HST is able to resolve its stellar content. Malumuth et al. (1996) obtained HST UV, U, B and V images of this region and derived an ionizing luminosity of 5 × 1050 phots-1 and an average reddening EB-V = 0.36±0.28 mag. The stars are mostly concentrated in the central part of the region, where little emission of gas is seen (the ionized gas lies more in a shell around the stars, figure 1a). Wilson & Scoville (1993) showed the molecular gas to be situated in the south-east part of the region, just outside of the bright knot of stars. Viallefond et al. (1986) found a reddening gradient in the north-east/south-west direction by observing the Hi gas, which was confirmed by Malumuth et al. (1996) with stellar photometry.

We obtained ISO images for NGC 595 in the 5.0 to 8.5 μm range. The emission in this spectral range is dominated by the so-called PAH bands. Current interpretation of these has them originating from stochastically heated molecules. Two of these bands are located in the range observed, at 6.2 μm and 7.7 μm. Stochastic heating implies that the in-band flux is directly proportional to the number of photons absorbed by the molecules. For typical HII regions, Cohen et al. (1989) found 0.58 for the I6.2/I7.7 in-band ratio. However many processes, ionization, dehydrogenation, can modify this ratio. Furthermore, an underlying continuum is present though its exact origin is unknown.

Ever since their discovery, blue compact dwarf (BCD) galaxies have been thought of as excellent candidates for being young galaxies, i.e. systems presently undergoing one of their very first bursts of star formation. This is mainly because BCDs are very metal-deficient, the metallicities of their ionized gas ranging between Zʘ/50 and Zʘ/3, which makes them the least chemically evolved galaxies in the Universe. Other evidence, such as the very high fractional neutral hydrogen gas content and the lack of an evident underlying old stellar population on optical images, also point to the relative youth of some BCDs.

Thuan, Izotov h Lipovetsky (1997) have argued SBS 0335-052 to be a young galaxy on the basis of the following evidence: 1) HST imaging of the BCD shows its underlying extended low surface brightness component to have an irregular and filamentary structure, suggesting that a significant part of the emission (~1/3) comes not from an underlying stellar population, but from ionized gas. Any underlying stellar population must be younger than ~108yr. Propagating star formation occurs in a chain of 6 super-star clusters with ages ranging between 4 and 30 Myr. 2) The underlying component shows unusually blue colors consistent with gaseous emission, in contrast to most BCDs which possess an underlying red component. 3) VLA 21 cm observations show the BCD to be embedded in an extraordinarily large HI cloud with dimensions some 64 × 24 kpc (the typical size of HI envelopes around BCDs is more like a few kpc in each dimension).

NGC 4449 is a bright (B = 9.5), nearby (4 Mpc) irregular galaxy with a high surface brightness and strong star-formation activity; for photos see Hunter and Gallagher (1989). Using the Effelsberg 100-meter dish, van Woerden et al. (1975) - and later also Bajają et al. (1994) - found its HI distribution to extend, at the level 2 × 1019 atoms cm-2 over a full degree, or ten times the D25 diameter. Van Woerden et al. also mentioned that the HI velocity gradient in this outer disk was opposite to that measured at Westerbork over the innermost 4 arcminutes.

With the VLA D-array, we have used a mosaic of 3 × 3 pointings to map the HI at resolutions of 1.0’ and 10 kms-1 over a field (FWHM) of 1° diameter (Hunter et al. 1998). The HI distribution (Fig. 1) shows the following features: 1) a bright inner body, centred on the optical galaxy and dominated by an annulus of 5 kpc diameter; 2) a symmetric elliptical - probably highly inclined - disk of 35 × 15 kpc diameter, with strong condensations at its NE and SW ends; 3) a set of narrow streamers, consisting of straight segments making abrupt angles, and containing several bright clouds. The HI mass in the inner body and elliptical disk together is 12 × 10s Mʘ the streamers (about 7 × 108 Mʘ of HI) are surrounded by a diffuse HI distribution, missed by the VLA interferometers but detected by the Effelsberg single-dish.

In recent years, color-magnitude diagrams have revealed that nearby dwarf galaxies have a fascinating variety of star formation histories. The fact that the dwarfs have periods of star formation activity and inactivity raises several questions. How do the properties of the interstellar medium (ISM) in a galaxy affect its star formation rate? Are the interstellar media of dwarf galaxies similar to those in spirals? And what can we learn about dwarf galaxies’ evolution from their gas contents?

Recent studies (Puche & Westpfahl 1994, Young & Lo 1996) have shown that the distribution of HI in some extreme low luminosity dwarf irregular galaxies (e.g. M81dwA, Holmberg I, Leo A) tends to have a ring-like (or shell-like) distribution which suggests that a single burst of star formation could expell most of the remaining ISM (or at least a large fraction of it) from the system. In view of this, Puche & Westpfahl (1994) suggested that in dwarf spheroidal galaxies, the HI should be found at large radii since no young stellar population is observed in most of them.

The line flux ratio of the [C II] 158 μm fine-structure (2P3/2 → 2P3/2) emission to CO rotational emission is a probe of the intensity of the interstellar UV radiation in galaxies. The largest values for the [C II]/CO (J = 1-0) line ratios are found in starburst galaxies, none of which, however, show values larger than 6 ×103 (Stacey et al. 1991). The exception ([CII]/CO ≃ 2 × 104) are galaxies of Magellanic type which have low metallicity, an intense interstellar UV field, and irregular morphology (Table 1).

Mochizuki et al. (1994) showed that the two conditions, a low metallicity and an intense UV field are expected to result in the large [C II]/CO ratios in the Magellanic types. These conditions are likely to be common in more massive galaxies as well, but at an early epoch in their evolution. This suggests that bright [C II] emission can be a good indicator of young galaxies. However, the metal-poor galaxies previously observed in the [C II] line were exclusively Magellanic types. Observations of different types of dwarf galaxies are necessary to help us understand the physical conditions in the interstellar medium of low-metallicity systems.

The Local Group contains 37 currently known or probable member galaxies within a radius of 1.8 Mpc (Grebel 1997a). 31 of these galaxies are dwarf galaxies with MB > -17 mag. The most frequent galaxy types are low-mass, gas-poor dwarf spheroidal (dSph) galaxies, which are dominated by old and intermediate-age populations, and more massive, gas-rich dwarf irregular (dIrr) galaxies with recent star formation. DSph galaxies have been found mostly in close vicinity to the large spiral galaxies in the Local Group, while dirr galaxies tend to be more distant and isolated.

Stellar Evolution theory is based upon well understood physics and provides clear predictions as to how a Colour-Magnitude Diagram (CMD) will change due to effects of age and metallicity. The theory has been tested by looking at nearby coeval star clusters. The power of applying CMD analysis to galaxies has been demonstrated in studies of the Carina dSph (Smecker-Hane et al. 1996). In Carina the observation of separate, distinct Main Sequence (MS) Turnoffs has forced us to believe that this small, nearby companion of our Galaxy has had a very complex star formation history. No similar direct evidence for “bursting” behaviour on a global scale has been seen indisputably in larger systems. Partly this is due to the greater distances of larger systems, but also to the complications in distinguishing old star formation events (> 1 Gyr old) in systems which are currently forming stars. Where we lack MS turnoffs we have to resort to statistical modeling of the CMD. This has been applied using a a number of different approaches, but all generally based on Monte-Carlo techniques (e.g. Tosi et al. 1992; Bertelli et al. 1992; Tolstoy &; Saha 1996). However, even using these more sophisticated analysis techniques, it is difficult to find unique solutions. This is mostly due to the age-metallicity degeneracy on the Red Giant Branch (RGB). The RGB is usually the most populated, easiest to observe phase of stellar evolution. The Carina CMD reveals the dangers of blindly interpreting the RGB, because from the RGB alone it is impossible to extract the information revealed by the MS Turnoffs.

The results of an investigation of the evolutionary status of fifteen gas-rich, low surface brightness dwarf galaxies (LSBDGs) are presented. LSBDGs are defined by unusually high values of MH/LB and the presence of extended HI envelopes. At both the present epoch and the past, the star formation process appears to be inefficient: the LSBDGs are underluminous for their HI mass and very little active star formation is currently observed. Analysis of global optical colors and elemental enrichments indicate that these objects are not “young” systems; rather, star formation has been occurring for several Gyr. One clue to the star formation history of these objects is that the global gas density is significantly lower than the Toomre instability criterion throughout the gas disk. Local peaks in the HI surface density, however, approach the instability criterion and are correlated with sites of active star formation. Thus, while star formation appears to be inhibited globally, the local gas distribution plays a crucial role in regulating the star formation activity in these low mass galaxies.

The M81 group is one of the nearest groups of galaxies, but its properties are quite different from those of the Local Group. It has therefore provided a different environment for the evolution of its member galaxies. We have carried out a CCD survey of the M81 group to search for analogs to Local Group dwarf elliptical (dE) galaxies. All the M81 dwarfs previously identified in photographic surveys were recovered and we also discovered several new systems whose surface brightnesses fall within the range found for Local Group dE’s.

We have obtained HST WFPC2 images through the F555W and F814W filters of two M81 group dE’s: BK5N and a new system, designated F8D1. The resulting color-magnitude diagrams show the upper two magnitudes of the red giant branch. The I magnitudes of the red giant branch tip in both galaxies yield distances that are consistent with membership in the M81 group. Surface brightness and total magnitude measurements indicate that BK5N and F8D1 have similar central surface brightness (24.5 and 25.4 mag arcsec-2 in V, respectively), but F8D1 is larger length scale results in it being 3 magnitudes more luminous than BK5N. BK5N lies on the relation between central surface brightness and absolute magnitude defined by Local Group dwarf ellipticals, but F8D1 does not. F8D1 is more luminous for its central surface brightness than the relation predicts, similar to the large low surface brightness dwarf galaxies found in, for example, Virgo. The mean color of the giant branch is used to establish the mean abundance of each galaxy. F8D1, the more luminous galaxy, is significantly more metal rich ([Fe/H] ≈ -1.0) than BK5N ([Fe/H] ≈ -1.7). Both BK5N and F8D1 lie on the relation between absolute magnitude and metal abundance defined by Local Group dwarf ellipticals. However, as regards the relation between central surface brightness and metal abundance, BK5N again follows the Local Group dwarfs, while F8D1 deviates significantly from this relation. This suggests that the total amount of luminous matter is more fundamental in controlling metal enrichment than the surface density of luminous matter. We have also used the color width of the giant branch compared with the photometric errors to establish abundance ranges in both galaxies, the sizes of which are comparable to those in Local Group dE’s.

In the Local Group few Hɪɪ regions exhibit nebular Heɪɪ λ4686 emission, indicative of unusually high excitation. Approximately eight such nebulae are known, all located in low metallicity environments (IC 1613, SMC, LMC) except the Galactic ring nebula G2.4+1.4 (Esteban et al. 1992). The best studied case is the nebula S 3 surrounding the WO3 star DR 1 (Garnett et al. 1991, Kingsburgh &: Barlow 1995). Except for two cases (N 44C, N159 F; Stasińska et al. 1986, Pakull & Angebault 1986) the nebulae are asociated with early WN (cf. Niemela et al. 1991) and WO stars.

So far approximately 50 to 60 extragalactic Hɪɪ regions showing nebular Heɪɪ λ4686 are known (Campbell et al. 1986, Izotov et al. 1994, 1997a). Most of them are found in H II galaxies, BCDs and related objects from low metallicity samples used for determinations of the primordial helium abundance. The most prominent case is the low metallicity (Z) record holder I Zw 18. He II emission is typically on the order of 1-5 % of Hβ.

Different ionization mechanisms (photoionization by hot stars, shock excitation, photoionization by X-rays) have been put forward to explain the required hard spectrum (Garnett et al. 1991).

The first quantitative approach to address this problem are the evolutionary synthesis models of Schaerer (1996) and Schaerer & Vacca (1997), which include well tested evolutionary tracks and recent spherically expanding non-LTE atmosphere models appropriate for WR and O stars.

Diffraction limited near-infrared H-band (1.6 μm) NICMOS HST images are scheduled to be obtained in mid-October 1997 of the young cluster NGC 2070 (age 3.5 Myr) in the 30 Dor giant HII region in the LMC. The aim is to search for the low-mass (M < 2 Mʘ) low-luminosity, red pre-Main Sequence stellar population and to establish the H-band infrared luminosity function. With the NICMOS we can now determine whether the IMF in this prototypical extragalactic starburst cluster is deficient in subsolar low-mass stars or not. The best ground-based data can sample only M > 2 Mʘ. In principle, NICMOS in the H-band (F160W) is sensitive enough to reach a magnitude of ~ 23.5 in a relatively short integration time, which indeed corresponds to the fantastic possibility to detect young stellar objects with masses near the hydrogen burning limit (M=0.1 Mʘ) according to pre-Main Sequence evolutionary models. Even if we could reach only H = 22.5 (i.e. M=0.4 Mʘ), our observations will still go a long way in directly answering, by star counts, whether the IMF in starburst galaxies is low-mass deficient or not, with all the corresponding far-reaching implications.

The observations would also tell us whether the 30 Dor cluster can be regarded as a prototype young globular cluster. This possibility would be ruled out, if we found NGC 2070 to be low-mass deficient, because old globular clusters do have a rich population of low-mass stars.

We use a sample of 45 low-metallicity H II regions to determine the primordial helium abundance Yp. We have carefully investigated the physical effects which may make the He I line intensities deviate from their recombination values such as collisional and fluorescent enhancements, underlying He I stellar absorption and absorption by Galactic interstellar Na, I. By extrapolating the Y vs. O/H linear regression to O/H = 0, we obtain Yp = 0.245±0.002. Our Yp gives 0.06±0.01.

Because of their low gravitational energy, low-mass galaxies are seriously affected by energetical processes in their interstellar medium, such as supernova explosions, or by gravitational perturbations, e.g., by neighbouring galaxies. This can reasonably explain their variety of morphological types. If the evolutionary timescales of galaxies are predominantly determined by internal processes, the multi-phase character as well as star-gas interactions and phase transitions have to be taken into account. For this purpose we have developed a numerical treatment of the dynamical behaviour of gas and stars, which also accounts for the metal dependence of some processes and which can trace the chemical evolution for different elements. This so-called chemodynamical treatment is described in detail in Theis et al. (1992) and Samland et al. (1997). It considers three stellar components and devides the gas into clouds (CM, with a mass spectrum) and a hot intercloud medium (ICM). Since the element enhancement of the interstellar medium is produced by different processes with different lifetimes of their progenitors, O, Fe, and N are used as tracer elements to represent supernovae type II (SNell), type la (SNela), and planetary nebulae (PNe) contributions. While supernovae form the ICM, PNe only attribute to the CM so that only mixing effects of both gas phases can alter abundance ratios.

Due to limited computer capacities the first chemodynamical simulations of dwarf galaxies could be performed only one-dimensionally so far (see e.g., Hensler et al. 1993, 1998). The recently developed two-dimensional chemodynamical code CoDEx (Samland 1994) was first applied to massive disk galaxies and produced models of which a particular one could represent various chemical and structural observations of the Milky Way with striking agreement (Samland et al. 1997).

The life and evolution of galaxies are dramatically affected by environmental effects. Interactions with the intergalactic medium and collisions with companions cause major perturbations in the morphology and contents of galaxies: in particular stars and gas clouds may be gravitationally pulled out from their parent galaxies during tidal encounters, forming rings, tails and bridges. This debris of collisions lies at the origin of a new generation of small galaxies, the so-called “tidal dwarf galaxies” (hereafter TDGs). Such an exotic way of forming galaxies was put forward by Schweizer (1978) and by Mirabel et al. (1992), who clearly observed the genesis of a star-forming object, out of material tidally expelled from the interacting system NGC 4038/39 (“The Antennae”). Recent studies, based on optical and HI observations, have shown that TDGs actually form a class of “recycled” objects with some properties similar to the more classical dwarf irregulars (dIrr) and blue compact dwarf galaxies (BCDs).

Dwarf galaxies at high redshifts are important to study for a variety of reasons. By dwarf, we mean galaxies with low-mass, though galaxies with low-luminosities or small sizes are also commonly referred to as dwarfs. Several groups have suggested that such galaxies may be major contributors to faint blue galaxies, whose nature remains unclear. Dwarfs are viable candidates for faint blue galaxies if many undergo strong bursts of star-formation at redshifts z ~ 1 (Babul and Ferguson 1996) or even lower redshifts z ~ 0.3 (Cowie, Songaila, and Hu 1991; Broadhurst et al. 1988) and fade or disappear by today; if they have a much steeper luminosity function (Driver et al. 1994) than generally adopted in faint galaxy models; or if they represent small pre-merger fragments of larger galaxies today (Guiderdoni and Rocca-Volmerange 1990; Broadhurst, Ellis, and Glazebrook 1992), as might be expected in standard cold dark matter models with hierarchical galaxy formation.

We study the global properties of 51 compact field galaxies with redshifts z ~ 0.2 - 1.3 and apparent magnitudes I814 < 23.74 in the flanking fields of the Hubble Deep Field. All these galaxies have angular half-light radii re < 0.5 arcsec. Keck spectra covering ~4000-9000 Å, combined with HST І814 images and Keck V-band images, were used to derive redshifts, V606 - I814 colors, absolute blue magnitudes (MB), linear half-light radii (Re ), blue average surface brightnesses within Re (SBe ), velocity widths (σ), virial masses (M), mass-to-light ratios (M/L), excitations (O[III]/Hβ), and star formation rates (SFR). The results of this study can be summarized as follows:

1. (i): Only 12% of the 51 compact galaxies have absorption-line dominated spectra, while 88% show strong, narrow emission lines, similar to the so-called CNELGs (e.g., Koo, this volume).

2. (ii): Despite being very luminous (i.e., LB ~ L*; see figure la), compact emission-line galaxies are low-mass stellar sytems (i.e., M ≤ 1010 Mʘ, typically; see figure lb).

3. (iii): Roughly 60% of the compact emission-line galaxies have colors, sizes, surface brightnesses, luminosities, velocity widths, excitations, star formation rates, and mass-to-light ratios characteristic of young, star-forming HII galaxies (see figures 1 and 2). The remaining 40% form a more heterogeneous class of evolved starbursts, similar to local disk starburst galaxies.

4. (iv): Without additional star formation, galaxy evolution models predict that HII-like distant compacts will fade to resemble today’s spheroidal galaxies such as NGC 205 (Koo, this volume).