Velocity Dispersions in Spiral Galaxies.

Abstract

The recent development of Fourier techniques for the determination of stellar dispersions has provided a powerful new tool for the investigation of problems concerning the structure and dynamics of external galaxies. The majority of the studies using this new technique have concentrated on elliptical galaxies and have yielded some of the most important results in extragalactic astronomy. My goal in this thesis has been to extend the analysis to spiral galaxies; to determine how closely the spiral galaxies resemble elliptical galaxies, how much mass is present in the bulge, and how the bulge, disk, and perhaps a third dark halo component are related. Velocity dispersions have been determined using spectra obtained at the 1.3 m telescope of McGraw-Hill Observatory with a 2000 channel intensified Reticon scanner. A Fourier "quotient" method of estimating velocity dispersions was employed, with typical uncertainties of about 10%. We find that the spheroidal bulges of spiral galaxies obey the same L (TURN) (sigma)('4) relationship between luminosity and velocity dispersion that has been found in elliptical galaxies, but with dispersions smaller by 15% at a given absolute magnitude. Possible errors in the determination of the bulge luminosities, and uncertainties in our estimates of (sigma) due to contamination from the disk, dictate a cautious interpretation of this gap. Using a smaller sample of SO galaxies, we also find that the L (TURN) (sigma)('4) relationship holds for the spheroidal bulge component of these galaxies, with essentially no gap with the ellipticals. These results indicate that the spheroidal components of elliptical, SO, and spiral galaxies are dynamically similar, and have probably undergone similar evolutionary histories. The scatter in the L (TURN) (sigma)('4) relationship is roughly twice as large in the spiral bulges as in the elliptical galaxies. While part of this scatter is due to uncertainties in the observations, we also find a correlation between the residuals from the L (TURN) (sigma)('4) relationship and line strengths. Other parameters which may affect the scatter, or the gap between the relationships of the spiral bulges and ellipticals are: (1) differences in mass-to-light ratios, (2) differences in characteristic surface brightnesses, (3) differences in rotational support. A comparison of our velocity dispersions with rotational velocities indicates that the bulge and halo components of spiral galaxies are dynamically separate with the bulges having the smaller dispersions. A simple three-component model has been constructed to analyse these results. These models indicate that our observations may provide an independent test for the existence of massive halos in spiral galaxies. Our study of velocity dispersions in M31 indicates that the ratio of the intrinsic dispersion in the bulge to the dispersion in the nucleus is 0.83 (+OR-) 0.12. We find that mixing of light from the bulge in the spectrum of the nucleus, and rotational contamination, can severely affect the determination of this ratio. We have determined corrections for these contaminations by performing numerical experiments. Our results show that the mass-to-light ratio in the nucleus is probably four times larger than in the bulge. We find that the velocity dispersion in M32 fits the L (TURN) (sigma)('4) relationship for elliptical galaxies, even though M32 appears to be the tidal remnant of a once larger galaxy. This indicates that the relationship holds over at least eight magnitudes, and suggests that M32 has relaxed to an equilibrium state similar to normal elliptical galaxies.