Conclusions

 It is widely recognized that M82 is the prototype of galactic wind systems. Most models have concentrated on explaining measurements taken along isolated position angles close to the galaxy's minor axis. However, our new study has shown that the filamentary system associated with the outflow is highly complex. Our simplest kinematic model requires at least two discrete structures for each side of the galaxy. The filaments are distributed in a network over these surfaces. More surprisingly, the axes of the outflows on either side of the disk are aligned neither with each other nor with the spin axis of the disk. The gas excitation suggests that stellar ionization dominates close to the disk with increased dominance of shock excitation further along the outflow. The clearest signature of the biconic outflow geometry is seen in the line ratio maps which suggest that radiation is escaping from the disk along a channel excavated by the hot rarefied wind. We find evidence for a smooth line-emitting halo which we associate with the linearly polarized halo seen in broadband studies. We have observed a warm ionized medium throughout the inclined spiral disk.

In our view, the most pressing issue is an explanation of the emission-line spectropolarimetry. We strongly urge that these observations are repeated to deeper levels, at least to the point where the diffuse line-emitting halo is detected. Observations which only detect the bright filaments cannot remove the contribution from the halo along the line of sight.

The disk-halo interaction in galaxies is a fundamental topic in its own right ([Bloemen 1990]). We suggest that M82 may provide the best observational constraints on fountain flows (e.g., [Shapiro & Benjamin 1991]) once the origin of the filaments is fully understood. What happens to the entrained material that is lifted into the galactic halo? What is the relationship between the relativistic electron halo, the wind filaments, and the diffuse line-emitting halo? We suspect that answers can only come from future observations of x-ray, radio and millimeter emission at a resolution and sensitivity comparable to the present study.

Acknowledgments:

Funding for this research was provided by the office of the Dean of Natural Sciences at Rice University, the Texas Space Grant Consortium, and the Sigma Xi Research Society. Additional funding was provided by AURA/STScI (grant GO-4382.01.92A), the National Science Foundation (grant AST 88-18900), the William F. Marlar Foundation, the National Optical Astronomy Observatories (NOAO), and Mr. and Mrs. William Gordon. This research was performed in partial fulfillment of the Ph.D. degree at Rice University. JBH acknowledges a Fullam Award from the Dudley Observatory.

We thank Drs. Reginald J. Dufour, Patrick M. Hartigan, Jon C. Weisheit, C. R. O'Dell, and Sylvain Veilleux for enlightening discussions of the material contained herein. Comments from our referee, Bill Keel, have led to notable improvements to the paper. Special thanks is due Drs. Joel Bregman and Brent Tully for providing the ROSAT and deep H$\alpha$ imagery, respectively, and Drs. Geoff Bicknell and Michael Dopita for their assistance with the stellar wind equations and high-velocity shock models, respectively.

This research has made use of the NASA/IPAC Extragalactic Database (NED) which is operated by the Jet Propulsion Laboratory, Caltech, under contract with the National Aeronautics and Space Administration. The Astrophysics Science Information and Abstract Service (ASIAS), administered by the Astrophysics Data System (ADS), was also used. Color versions of several figures in this paper are available from the authors.