1. INTRODUCTION

Since I want to describe some new results, I shall not dwell on the history of the subject as is customary on these occasions. Instead, I shall confine myself to a brief summary of the present state of affairs, without saying how we got there. Thus, I shall largely resist the temptation to dispense remarks of a vaguely philosophical character of the kind which M. J. Rees once described to me as "unspecific wisdom". Nevertheless, it is clear from the impressive number and variety of the talks and posters at this Workshop that the field of QSO absorption lines is just beginning a new phase, one in which the lines are actually being used to obtain information on the distant Universe. Accordingly, we shall hear reports on the properties and distribution of distant galaxies, on the intergalactic medium and the meta-galactic radiation field, all obtainable in no other way.

The QSO absorption lines are currently divided into three distinct categories:

1. the heavy-element redshifts;
2. the Ly- forest lines; and
3. the Broad Absorption Line (BAL) redshifts.

The heavy-element redshifts include two important sub-categories: the damped Ly- systems, which Wolfe (1988) will discuss, and the Lyman limit systems.

It is almost universally agreed that the Broad Absorption Line redshifts, whose properties will be described by Turnshek (1988), are caused by expanding gas which has been ejected from the QSO in question. I shall not say anything further about these systems. On the other hand, the origin of the first two categories has been more controversial. There are several reasons for believing that the majority of the sharp absorption lines are formed in cosmologically distributed, intervening objects. The strongest pieces of evidence are as follows.

1. The distribution of number of redshifts observed per object is close to Poissonian, as would be expected. (Later in this paper it will be shown that the distribution is not exactly Poissonian and that this is to be expected when the complex nature of the large-scale distribution of galaxies is taken into account.)

2. The presence of common absorption systems in the spectra of QSOs widely separated on the plane of the sky (Shaver and Robertson 1983a, b).

3. The energy problems which arise if the absorption systems are supposed to be produced in gas ejected from the QSO (Goldreich and Sargent 1976).

An exception to this generalization is provided by certain heavy-element absorption complexes which are found close to the emission redshift in certain categories of radio-QSOs (Foltz et al. 1986; Barthel 1987; Foltz et al. 1988). These complexes, which may be related to the broad absorption troughs, almost certainly arise in gas which has been ejected by the galaxy containing the radio source.