Galaxy interactions were much more frequent and the starbursts they can trigger if gas is involved were much stronger in the early universe. Yet their little brothers and sister in the local universe can be studied to much greater detail - and have revealed a number of surprises in the last few years. Using GALEV models in comparison to integrated photometry and spectra of starburst galaxies - from the Blue Compact Dwarfs (BCDs) through the spectacular "big mergers" like NGC 7252 or the Antennae galaxies, it has been possible - by including all available information from other wavelength regions like radio-HI, mm-CO, and X-ray maps - to determine burst strengths and star formation efficiencies (SFEs), showing a clear trend of decreasing burst strengths with increasing total galaxy mass for the (apparently fairly isolated) BCDs on one hand and burst strengths and SFEs up to two orders of magnitude higher in massive gas-rich mergers (Krüger et al. 91, 92, 93, 94, 95, Fritze & Gerhard 94a, b, Fritze 05). For the extreme nuclear starbursts in Ultraluminous Infrared galaxies (ULIRGs) extremely high SFEs have independently been reported, together with a molecular cloud structure fundamentally different from the one known in our Milky Way in terms of much higher fractions of molecular gas having the high densities of molecular cloud cores. This raises the question whether SF is one universal process with a tremendous range of scaling or whether there are two different modes of "violent" and "normal" SF. Hydrodynamical models predict a high ambient pressure in gas-rich mergers which can account for larger and denser molecular clouds and higher SFEs. High SFEs, in turn, are a prerequisite for the formation of massive and long-term bound star clusters (SCs) that evolve into globular clusters (GCs). Our analysis of the ~1 Gyr old/young SC population in NGC 7252 showed that GC formation in this gas-rich spiral-spiral merger produced a secondary population of GCs about as rich in number as the inherited population from both spirals (Fritze & Burkert 95).

Even in the nearest major merger, the Antennae, the molecular cloud mass spectrum or the molecular cloud structure cannot be resolved today. Before ALMA gets operational, the young massive SCs are our best proxy for high SFEs and massive molecular cloud cores in interacting galaxies. Since the ages of star clusters can be determined very accurately (together with their extinctions and metallicities), young star clusters are tracers of the dynamical evolution of starbursts (nuclear vs global, contracting, propagating). The ages of intermediate age/old GCs track major mergers, their metallicities give clues to the galaxy types involved (Fritze 04). 

Pixel-by-pixel analyses of the ACS Early Release data for the Tadpole and Mice interacting galaxy systems with GALEV models have revealed the spatial   distribution of stars of various ages and, surprisingly, found star formation (SF) and even star cluster formation (SCF) all along the ~200 kpc long tidal tails. Other instances of extragalactic SF are found in deep H-alpha imaging (Ferguson+98), in the UV with Galex (Thilker+05). In collaboration with J. Gallagher (Madison) and E. Wehner (Canada) we investigate SF in low surface brightness structures around galaxies, the extremely low SFR density regime (Wehner+06).

Another intriguing feature of strong galaxy interactions is the formation of so-called Tidal Dwarf Galaxies (TDGs) in the extended tidal tails thrown out of spiral disks. P. Weilbacher analysed the first sample of TDGs in his Diploma and PhD theses in Göttingen. Combining optical and NIR data, he investigated the ratio of stars formed in situ and stars inherited from the spiral and could, for the first time, show by means of VLT tilted slit commissioning spectra that some of these systems already show rotation, i.e. dynamical decoupling from their surrounding tail, in combination with infall (still in formation) (Weilbacher+00, 01, 02, 03). The cosmological implications of this recycling mode of dwarf galaxy formation still going on in the local universe are being explored in collaboration with P. Weilbacher (now Potsdam).

In collaboration with P.-A. Duc (Saclay) and his group, E. Brinks (Hertfordshire for HI) and J. Braine (Marseille for CO) we explore the SF process in the expanding low surface density tidal tails of galaxies. Like in the shells around galaxies, the HI density in the tidal tails is far below Kennicutt's threshold for SF, yet the HI is dynamically very cold. A spectacular recent result from the comparison of stellar and total dynamical mass in one TDG showed there must be a substantial fraction of DM in this TDG (Bournaud+07), unexpected due to its origin in disk material. The most plausible assumption about the nature of this DM is some form of very cold molecular gas, that might make up a significant mass fraction of the outer disks of spirals. Clearly more TDGs need to be examined this way before firm conclusion can be drawn.

References:
- Bournaud+07
- Ferguson et al, 98
- Fritze 04
- Fritze 05
- Fritze & Burkert 95
- Fritze & Gerhard 94a
- Fritze & Gerhard 94b
- Krüger et al. 91
- Krüger et al. 92
- Krüger et al. 93
- Krüger et al. 94
- Krüger et al. 95
- Thilker+05
- Wehner+06
- Weilbacher+00, 01, 02, 03