Kinematics and Dynamics of Barred Galaxies

In this thesis I present several tools for the study of the kinematics and dynamics of barred galaxies. I investigate methods of interpreting observations of velocity fields. If streamlines in a bar have the same symmetry as the bar, the velocity field can be recovered. This inversion is singular when the line of nodes of projection corresponds to one of the axes of symmetry of the bar. Near these viewing geometries, the velocity field cannot be inverted; it is important to select target galaxies far from these singular points. I show that fitting to circular motion alone can lead to large systematic errors ( ~50 percent or more) in the rotation curve. I derive formulae for estimating forces from a non-circular velocity field. These methods were tested on synthesised observations of an N-body bar; the results were satisfactory. Williams' Fabry-Perot observations of the projected velocity field for NGC 1832 were analyzed. The bar is near a singular point of projection, yet there are large asymmetries in the projected field. It would require an unreasonably large amount of non-circular or vertical motion acting alone to produce these distortions. The velocity field can more reasonably be explained by a combination of barlike streaming motion (~upsilon _c/2) and a slight ( ~10-20^circ) warp. I present a method for computing analytical approximations to the potentials of nearly spherical galaxies. I have tested the accuracy and the effect on orbital structure. For the models tested, the approximations were accurate for axis ratios less than ~ 1.4, and surfaces of section were not significantly altered. A Mathematica program for automatic computation of these approximate formulae is presented. I present a self-consistent field method for the calculation of equilibrium models of rotating bars. The closed orbits in a potential are integrated, and each orbit's response density is calculated. After weighting these orbits with a function of action, a new density model is computed. This model is then used to generate a new potential, and the process is iterated until convergence. Several models are presented, and alternative formulations are discussed.