The title which the organizers of the Symposium have suggested for this introductory report covers an enormous domain. Even if one limits oneself, as I intend to do, to the large-scale structure, it is impossible to summarize this adequately in one communication. I intend therefore to confine myself to give, firstly, a list of principal subjects in which advances have been made and of the main problems which these advances have raised, and, secondly, to comment somewhat more extensively on some points which at the moment appear to be especially important.

The recent emphasis on stellar evolution, especially as seen in star clusters, has given us — mainly through Sandage's synthesis of the available material — a good outline of the evolutionary process. Although work of a survey nature obviously must be continued, if only to turn up clusters with an age not represented among those now known, I believe the time has come to concentrate a major effort on detailed studies of the brighter known clusters of various ages. Our knowledge of even the Hyades and Pleiades clusters is far from complete. A full account cannot be given here but I would like to mention some examples of the details in the colour-magnitude arrays that may eventually lead to important tests, or revisions, of current evolutionary models.

The galactic star clusters are today the only objects whose individual distances and ages can be determined rather precisely. Therefore the space distribution as a function of time can be studied. In case they form a spiral structure, the lifetime of such a structure can be studied too.

NGC 6067 is a very rich cluster free of nebulosity in the Norma Cloud, whose colour-magnitude array has been published by Thackeray, Wesselink, and Harding (1962). The choice of stars for measurement was highly restricted to a square approximately 10′ × 10′ of arc near the centre, so that above B = 13·0 only 10% are expected to be field stars, and between B = 15·0 and 16·0 some 20–30%.

A study of the cluster NGC 4755 (κ Crucis) was begun as a result of the several interesting problems raised by an earlier study of NGC 3293 (Feast 1958). Determinations of radial velocity and spectral type have been made in NGC 4755 from 102 spectra of 35 stars. Full details are published elsewhere (Feast 1963). These results were combined with the three-colour photometry of Arp and van Sant (1958). The cluster has a nearly-vertical evolved main sequence at spectral type B1. The spectral types and B, V measures for giants and dwarfs yield a mean reddening of +0m48 and a distance of 2 · 36 kpc. These values are considerably greater than those deduced by Arp and van Sant but they have been confirmed from an analysis of the three-colour measurements alone.

For many generations astronomers have been fascinated by open clusters of stars, of which many are found close to the central plane of the Milky Way. Only in comparatively recent years, however, has it become possible to derive reasonably consistent distances for these stellar systems. The difficulty was that until techniques of photoelectric photometry in at least three colours were developed, there were no reliable estimates of the amount of interstellar absorption and reddening.

Information about the past history of the Galaxy may be obtained from a two-colour diagram, since the present distribution of stars in such a diagram depends on the past rate of star formation and the past metal abundances in the interstellar medium. As an illustration of this, I would like to discuss three diagrams that were recently prepared by Dr. M. E. Dixon from Cape photometric data (Dixon 1963a,b).

In the past 10 years numerous extremely rapid variables have been found in different stellar associations. The spectrophotometric characteristics of some of these explosive variables are so similar to those of the UV Ceti stars that we propose the term “flash stars” for this variety, in order to provisionally distinguish them from the classical flare stars.

A search was made for neutral hydrogen in the following galactic star clusters — the Pleiades, NGC 1502, the Trapezium-Orion cluster, NGC 2244, and NGC 6910 (Davies and Tovmassian 1963). Observations were made with a six-channel receiver having a 5 kc/s bandwidth; it was used with the 250-foot radio telescope which has a beamwidth of 12′ by 18′. The first stage of the receiver was a parametric amplifier with an excess temperature of 200°K. Simultaneous measurements with a 5 Mc/s bandwidth showed the existence of ionized hydrogen associated with the clusters.

Planetary nebulae are convenient objects for studying the large-scale structure of the Galaxy. Firstly, they are easily recognized up to considerable distances on plates taken through an objective prism, and secondly, methods have been devised by various authors to determine their distances from two observable quantities: angular diameter and surface brightness. The importance of the subsystem of planetary nebulae has been accentuated especially by the discoveries by Minkowski and Haro of large numbers of planetaries in the direction of the galactic centre. The distribution of planetaries on the sphere suggests that they are connected with the galactic nucleus, but no direct determination of their distances, which would either confirm or contradict this statement, is available. The most serious obstacle in studying the subsystem of planetaries is the lack of observing data. The aim of the reported paper (Perek 1963) is to give a tentative outline of the distribution of planetaries in space based on extensive new observing material.

The theory of radiative and collisional processes in gaseous nebulae makes possible precise predictions of line and continuum intensities as functions of density, electron temperature, and degree of ionization of the emitting gases. It predicts, for example, the relative intensities of successive members of the Balmer series (Balmer decrement), and of corresponding transitions in the Balmer and Paschen series. Hence accurate line intensity measurements are not only of astrophysical interest; they also serve to assess the amount of space absorption.

The Scorpio–Centaurus association extends from about lII=356°, bII=+20° towards lower longitudes over about 90° of the southern sky. Between lII=312 and 356° it is located between latitudes +10 and +25°, and appears well detached from the more distant B-star population at lower latitudes. At longitudes below 312° it approaches and crosses the galactic plane and is seen projected against the general low latitude B-star population from which it is more difficult to separate. These features are clearly brought out by Figures 5 and 6 of the author's thesis (1946) and had been recognized earlier by others (e.g. Kapteyn 1914).

Radial velocities of 570 B stars, mostly fainter than apparent magnitude 7 · 5 of spectral types between O8 and B5, have recently been determined at Victoria (Petrie and Pearce 1962). Spectroscopic absolute magnitudes based upon the measured strength of Hγ absorption have been determined for these stars. In addition to this totally new material, spectroscopic absolute magnitudes have been measured for an additional 180 stars. From the two lists 490 stars are available for a study of their distributions and radial velocities, since for this number magnitudes and colours have been measured with sufficient precision to allow reliable estimates of the interstellar extinction and the true distance moduli. The results given here are preliminary in the sense that approximately 200 additional stars will be added to the study when the program to determine photoelectric magnitudes and colours is completed. The radial velocities and absolute magnitudes of all the stars have already been measured.

At the Radcliffe Observatory, Dr. Feast and I have recently compiled a third list of radial velocities of O and B stars which was designed primarily to push our optical knowledge of the dynamics of the Galaxy to great distances from the Sun (Feast and Thackeray 1963). This program was preceded by the determination of distance moduli of 248 O and B stars in cooperation with Dr. Stoy of the Royal Observatory, Cape, and Dr. Wesselink at the Radcliffe Observatory (Feast et al. 1961). As a result of this work we now have radial velocities for a very considerable number of southern stars at distances greater than 2 and even 3 kpc. It is quite possible that many of these distances are overestimated and it is highly desirable that the distances should be improved by the Walraven multicolour method or by some other technique. In the meantime it can certainly be claimed that our knowledge of radial velocities of these distant stars is considerably better in the southern than in the northern hemisphere.

Recently a program of radial velocity determinations for 114 southern long-period variables has been completed. Full details of these results and an analysis of them are being published elsewhere (Feast 1963). Below are summarized some of the results which seem of particular interest for problems of galactic structure.

Internally consistent distances to classical cepheids can be derived with considerable accuracy because they satisfy period-luminosity and period-colour relations. Studies of five cepheids that are members of galactic clusters have led to a revised period-colour relation and a new zero point for the period-luminosity relation. Distances have been computed for all cepheids given in recently published photoelectric surveys. The material can be regarded as complete to a distance of about 1500 pc.

In this short paper, the calibration procedure for the absolute magnitudes derived from the Hβ indices will be discussed. These Hβ indices have been measured for the distant OB-type stars in our studies of galactic structure and are discussed in the paper by Bok which is found elsewhere in this volume (Paper 36).

The field of NGC 6522 is of especial importance because Baade considered that the cluster-type variable stars found in this field were located in a stellar concentration at the centre of the Galaxy. This particular field is a few degrees away from the actual centre, but is a region of comparatively small absorption, so that stars at considerable distances from the Sun can be seen and studied. Baade found a marked maximum in the frequency of these variables between 17m and 18m apparent photographic magnitude on plates which were exposed down to a limit of 20m. He supposed that this maximum was due to an actual concentration of cluster-type variables round the centre of the Galaxy. Assuming 0m0 for the absolute magnitude, and allowing 3 magnitudes for absorption, Baade found a value of 8·2 kpc for R0, the distance to the centre of the Galaxy. The absorption was computed from a rather high value for the reddening of the cluster NGC 6522 itself, which was given by Stebbins and Whitford.

The detection of the 21-cm line 12 years ago opened up a vast new field of galactic research. But even though a large amount of data was assembled in these 12 years, our knowledge of the large-scale structure of our Galaxy still remains fairly limited. Let us summarize a few of the major facts which seem well established.

A large program of observations is being undertaken with the 210-foot telescope at Parkes, to study the fine structure of the 21-cm line emission in the region of the galactic equator. With the aid of a multi-channel receiver, H-line profiles have been obtained every 6 min of arc (i.e. every half-beamwidth) along the equator from lII=300 to 60°, and also from bII = −3 to +3° along a number of constant-longitude lines. The bandwidth of each channel of the receiver is equivalent to 8 · 0 km/sec, with a spacing of 7 · 0 km/sec between the centres of adjacent channels. The data have been recorded on punched paper tape, and the main reduction is being carried out on the SILLIAC computer of the University of Sydney. A sample group of profiles, illustrating the great variability from point to point, has been published previously (Kerr 1962b).