Among the various phenomena observed in interacting galaxies is the ejection due to tidal forces of stellar and gaseous material into the intergalactic medium and its subsequent rearranging which can lead to the formation of self-gravitating tidal dwarf galaxies (TDGs). We investigate this process with a detailed multiwavelength study of the interacting system Arp 245 and a numerical model of the collision computed with a Tree-SPH code. Our observations consist of optical/near-infrared broad band imaging, Hα imaging, optical spectroscopy, H I VLA cartography and CO line mapping. The system, composed of the two spiral galaxies NGC 2992 and NGC 2993, is observed at an early stage of the interaction, about 100 Myr after perigalacticon, though at a time when tidal tails have already developed. The VLA observations disclose a third partner to the interaction: an edge-on, flat galaxy, FGC 0938, which looks strikingly undisturbed and might just be falling towards the NGC 2992/93 system. Our H I map shows prominent counterparts to the optical tails. Whereas the stellar and gaseous components of the plume that originates from NGC 2992 match, the stellar and H I tails emanating from NGC 2993 have a different morphology. In particular, the H I forms a ring, a feature that has been successfully reproduced by our numerical simulations. The H I emission in the system as a whole peaks at the tip of the NGC 2992 tail where a gas reservoir of about 10 9 M ⊙ , about 60% of the H I towards NGC 2992, coincides with a star-forming -2optical condensation, A245N. The latter tidal object exhibits properties ranging between those of dwarf irregular galaxies (structural parameters, gas content, star formation rate) and those of spiral disks (metallicity, star formation efficiency, stellar population). Although it is likely, based on our analysis of the HI and model datacube, that A245N might become an independent dwarf galaxy, the dynamical evidence is still open to debate. Prompted by the questions raised for this particular object, we discuss some issues related to the definition and identification of TDGs and highlight some specific conditions which seem required to form them. We finally outline what is needed in terms of future numerical simulations in order to further our understanding of these objects.

Context. In Papers I and II of this series, we have found clear indications of the existence of two distinct populations of stars in the solar neighborhood belonging to the metal-rich end of the halo metallicity distribution function. Based on high-resolution, high S/N spectra, it is possible to distinguish between "high-alpha" and "low-alpha" components using the Results. The "high-alpha" halo stars have ages 2-3 Gyr larger than the "low-alpha" ones, with some probability that the thick-disk stars have ages intermediate between these two halo components. The orbital parameters show very distinct differences between the "high-alpha" and "low-alpha" halo stars. The "low-alpha" ones have r max 's to 30-40 kpc, z max 's to ≈18 kpc, and e max 's clumped at values greater than 0.85, while the "high-alpha" ones, r max 's to about 16 kpc, z max 's to 6-8 kpc, and e max values more or less uniformly distributed over 0.4-1.0. Conclusions. A dual in situ-plus-accretion formation scenario best explains the existence and characteristics of these two metal-rich halo populations, but one remaining defect is that this model is not consistent regarding the r max 's obtained for the in situ "high-alpha" component; the predicted values are too small. It appears that ω Cen may have contributed in a significant way to the existence of the "low-alpha" component; recent models, including dynamical friction and tidal stripping, have produced results consistent with the present mass and orbital characteristics of ω Cen, while at the same time including extremes in the orbital parameters as great as those of the "low-alpha" component.

We explore test particle orbits in the orbital plane of eccentric stellar binary systems, searching for ‘invariant loops’: closed curves that change shape periodically as a function of binary orbital phase as the test particles in them move under the gravity of the stars. Stable invariant loops play the same role in this periodically varying potential as stable periodic orbits do in stationary potentials; in particular, when dissipation is weak, gas will most likely follow the non‐intersecting loops, while nearby particle orbits librate around them. We use this method to set bounds on the sizes of discs around the stars, and on the gap between those and the inner edge of a possible circumbinary disc. Gas dynamics may impose further restrictions, but our study sets upper bounds for the size of circumstellar discs, and a lower bound for the inner radius of a circumbinary disc. We find that circumstellar discs are sharply reduced as the eccentricity of the binary grows. For the disc around the secondary star, the tidal (Jacobi) radius calculated for circular orbits at the periastron radius gives a good estimate of the maximum size. Discs change in size and shape only marginally with the binary phase, with no strong preference to increase or decrease at any particular phase. The circumstellar discs in particular can be quite asymmetric. We compare our results with other numerical and theoretical results and with observations of the α Centauri and L1551 systems, finding very good agreement. The calculated changes in the shapes and crowding of the circumstellar orbits can be used to predict how the disc luminosity and mass inflow should vary with binary phase.

With the aim of studying the nonlinear stellar and gaseous response to the gravitational potential of a galaxy such as the Milky Way, we have modeled 3D galactic spiral arms as a superposition of inhomogeneous oblate spheroids and added their contribution to an axisymmetric model of the Galactic mass distribution. Three spiral loci are proposed here, based in different sets of observations. A comparison of our model with a tight-winding approximation shows that the self-gravitation of the whole spiral pattern is important in the middle and outer galactic regions. A preliminary self-consistency analysis taking Ω p = 15 and 20 km s −1 kpc −1 for the angular speed of the spiral pattern, seems to favor the value Ω p = 20 km s −1 kpc −1 . As a first step to full 3D calculations the model is suitable for, we have explored the stellar orbital structure in the midplane of the Galaxy. We present the standard analysis in the pattern rotating frame, and complement this analysis with orbital information from the Galactic inertial frame. Prograde and retrograde orbits are defined unambiguously in the inertial frame, then labeled as such in the Poincaré diagrams of the non-inertial frame. In this manner we found a sharp separatrix between the two classes of orbits. Chaos is restricted to the prograde orbits, and its onset occurs for the higher spiral perturbation considered plausible in our Galaxy. An unrealistically high spiral perturbation tends to destroy the separatrix and make chaos pervasive. This may be relevant in other spiral galaxies.