In recent years, major progress has been made in several areas in the investigation of galaxy halos:
|(1)||It has been firmly established that the ISM in the halos of spiral galaxies is complex and comprises the same phases as the disk ISM. Therefore, observational studies should best be based on a multi-wavelength approach.|
|(2)||In some galaxies, there is kinematical evidence for outflows of gas from the disk into the halo. Models describing these outflows are generally consistent with the data.|
|(3)||The energy sources creating and maintaining gaseous halos are massive stars, their winds, and subsequent SNe in the galaxy disks.|
|(4)||Several heating processes can contribute to the energy balance of the ISM. The dominant energy sources of the diffuse halo gas are photo-ionization, shocks, and cosmic rays. Superwinds of starburst galaxies are probably predominantly thermally driven, while the Milky Way - which is thought to be in the galactic fountain mode - might have a CR-driven halo.|
|(5)||The existence and the morphology of the halos depend on the local level of star formation in the underlying disk.|
|(6)||Different models can be regarded as basic representations of ISMs with varying levels of energy input (sorted by increasing energy input rate per unit surface area: 2-phase, fountain, chimney, 3-phase).|
Although models like the chimney model (NI89) and the galactic fountain model (Shapiro & Field 1976) are good qualitative descriptions of the ISM as depicted by current observations, open questions remain, primarily in the following areas: the ionization structure of extraplanar diffuse ionized gas and the primary heating process under various conditions (which heating process dominates when and where?); the total energy balance; the role of gas turbulence: its generation, maintenance, and importance on the total energy budget and the kinematics of the gas; the role of large-scale magnetic fields and dynamos; i.e., are the energy densities of magnetic fields and cosmic rays in disk-halo interactions possibly larger than the thermal energy density? Or: does the magnetic field follow the ionized plasma or vice versa?, and quantitative measurements of the minimum energy input into the ISM to initiate outflows.
Spectroscopic studies of individual H II regions in face-on galaxies can possibly help resolve several of these open questions and expand our database of (nearby) galaxies with disk-halo interactions.
Additional information can be obtained from other reviews mentioned in this article and in particular from the proceedings of IAU Symposium 144 on ``The Interstellar Disk-halo Connection in Galaxies'', ed. H. Bloemen (1990).
It is a pleasure to thank M.-M. Mac Low and D. Breitschwerdt for fruitful discussions. Many thanks to S. Baggett, D. Breitschwerdt, T. Heckman, and M. Lehnert for very useful comments on this paper. I am also very grateful for the figures provided by S. Baggett, W. Pietsch, M. Lehnert, and E. Oliva (web posting of Fig. 5). This article has made extensive use of the Astrophysics Data System, and the NASA/IPAC Extragalactic Database, the contributions of which are hereby gratefully acknowledged.