Annu. Rev. Astron. Astrophys. 1996. 34: 155-206 Copyright © 1996 by Annual Reviews. All rights reserved

1. INTRODUCTION

The magnetic field of the Milky Way has been investigated for about 40 years, and those of external spiral galaxies for about 20 years. It now seems clear that spiral galaxies generally possess large-scale magnetic fields whose evolution and, possibly, their origins are controlled by induction effects in the partially ionized interstellar gas. Turbulent motions with scales below about 100 pc are present in this gas, and so the observed ubiquity of the large-scale galactic magnetic fields, coherent over scales of at least 1 kpc, requires special explanation. In fact, the theory of the galactic magnetic fields discussed in this review (known as mean-field magnetohydrodynamics) represents one of the earliest examples of synergetic theories describing how order can arise from chaos.

Our main emphasis is on magnetic fields whose scales exceed that of the interstellar turbulence. These are the fields - known as the mean, average, large-scale, global, or regular magnetic fields - that produce polarized radio emission in nearby spiral galaxies when observed at resolutions of 0.1-3 kpc. We also stress unresolved problems concerning the random (turbulent) magnetic fields in the interstellar medium (ISM), but we do not extend this discussion to the fields present in elliptical galaxies. Neither do we discuss phenomena connected with the central regions of the Milky Way.

The regular magnetic fields in the disks of spiral galaxies are usually considered to be the result of large-scale dynamo action, involving a collective inductive effect of turbulence (the -effect) and differential rotation. Even though alternatives to dynamo theory have been proposed, we believe that something resembling an -dynamo is the dominant mechanism, possibly sometimes modified by other hydromagnetic effects such as induction by streaming motions associated with spiral arms, other noncircular motions, and galactic fountains. The dynamo is the key ingredient of the theory: Other mechanisms by themselves are unable to maintain the observed large-scale galactic magnetic fields over galactic lifetimes.

The main rival of the dynamo theory is the primordial field theory. In this theory, one assumes that the observed magnetic patterns arise directly from a pregalactic magnetic field, distorted by the galactic differential rotation. We discuss why we believe that this theory, in spite of its appealing simplicity, cannot by itself give a detailed explanation of the range of field structures observed in spiral galaxies. A great conceptual advantage of the dynamo theory is that it can provide a universal explanation for the varied field configurations observed in spiral galaxies: axisymmetric and bisymmetric in azimuth; odd, even, and mixed parity vertically; etc. Of course, a primordial field may influence subsequent dynamo action, or it may be amplified by a dynamo.

The dynamo theory has its own difficulties. The linear version, which is valid when the magnetic field is too weak to significantly affect the velocity field, is relatively well developed and agrees favorably with observations wherever such a comparison is meaningful. However, the nonlinear saturation of the dynamo is not well understood and the conventional ideas were recently strongly criticized. They certainly need substantial improvement (Section 4). We argue, however, that the mathematical form of the mean-field dynamo equations is rather generic and robust, so that the available results are expected to be at least qualitatively correct, even though the details and the physical meaning of the coefficients of the dynamo equations may need to be revised.

The topics of this article have recently been reviewed by Wielebinski & Krause (1993) and Kronberg (1994). We have attempted to avoid unnecessary repetition of their material.