The old populations of galaxies are particularly vital for understanding the formation and chemical evolution of galaxies. As in any field, ideas come and go, and today’s discussion has given us a valuable assessment of current knowledge and thinking on many topics.

The problems of star formation and the IMF are basic to many aspects of work on old populations. The discussion included some interesting ideas on star formation, and a useful warning that there are probably different modes of star formation which may give different IMFs: for example the mode of star formation that is widely studied now (in spiral arms) is probably different to the mode of star formation that was active in forming the stars of the galactic bulge.

It is not easy to measure distances to galaxies. That we even try is as much a measure of our audacity as it is an indication of our ingenuity. But we do try, and we feel that we must try, because so many of our grandest questions and most basic problems about the universe depend on our knowledge of its size and scale.

This Joint Discussion brings together some of the astronomers who have concerned themselves with this difficult task. They do not all agree among themselves. This is not a topic that has reached a state of selfcomplacent fixity, but rather is still a subject of controversy and doubt. The Joint Discussion was designed by its organizers to expose the problems rather than to solve them, to discover the questions rather than to agree on an answer to them.

In talking about the overall distance scale of the Universe and the Hubble Constant, the Magellanic Clouds are good places to start. They are stellar systems large enough to contain stars, clusters and nebulae of all types, covering a wide age range. With modern telescopes and detectors, we are able to observe stars from the very bright down to those fainter intrinsically than our own Sun. From comparative studies, we may thus establish our basic calibrations of bright objects before moving out to measure the Universe at large. At the same time, the fact that both Magellanic Clouds are independently evolving galaxies, enables us to separate the effects of stellar age and chemical evolution on the calibrations that we make.

The study of Cepheids in external galaxies has been a continuous probing of the near-by distance scale accompanied by a refining and redefining of the Period-Luminosity relation itself. For example, the nearest two late-type galaxies of the Local Group, the Magellanic Clouds have always been the testbed for calibrations, from the pioneering optical studies of Leavitt (1907) to the recent infrared calibration of McGonegal et al. (1982) three quarters of a century later.

With time the data have improved in quality and in quantity but the methods of analysis and interpretation have so proliferated that the field might appear to be less well defined than it was at the turn of the century. Discussions of the problems have been so open that many have been driven to the conclusion that it is perhaps better to gauge the distances to galaxies using a variety of indicators, of secondary quality, rather than risk systematic errors at the start, through a complete reliance on one primary distance indicator. That situation, I believe, now has the potential for reversing itself.

Until the discovery (Corey & Wilkinson, 1976) of the anisotropy of the cosmic microwave background, the Virgo cluster represented something like a Rosetta Stone for many observational cosmologists: in the absence of a significant peculiar velocity component for the Local Group in the direction of the Virgo cluster, its distance, accurately measured, might reveal the global expansion rate and the Hubble age. Although this simple picture has changed, the distance of the Virgo cluster remains important, partly for a sharper understanding of the properties of rich clusters and the galaxies they contain, but more importantly (for my purposes here) as an interesting distance over which we may test various constructions of the extragalactic distance scale.

In this paper a review will be made of how one can use nuclear physics to put rather stringent limits on the age of the universe and thus the cosmic distance scale. As the other papers in this session have demonstrated there is some disagreement on the distance scale and thus the limits on the age of the universe (if the cosmological constant Λ = 0.) However, the disagreement is only over the last factor of 2, the basic timescale seems to really be remarkably well agreed upon. The universe is billions of years old - not thousands, not quintillions but billions of years. That our universe has a finite age is philosophically intriguing. That we can estimate that age to a fair degree of accuracy is truly impressive.

No single measurement of the time since the Big Bang gives a specific, unambiguous age. Fortunately, we have at our disposal several methods that together fix the age with surprising precision.

Within the framework of a model cosmology, the ages of the globular clusters should provide upper limits to the value of Hubble’s constant. Should both the ages and H be known to high accuracy, we may ask if standard comsological models are adequate. This paper will summarize recent results obtained by many astronomers for the ages of the globular clusters, how large the uncertainties are thought to be, and will suggest some further work.

The Tully-Fisher relation applied in the infrared appears to be the best global distance indicator presently available for determining the expansion rate and deviations from uniform Hubble flow. In this article recent results obtained using the IR/H I method are reviewed. A Virgo-directed Local Group velocity of about 300 km s–1 is indicated (implying a local value for the deceleration parameter qo ~ 0.05 – 0.1) along with a “best guess” value for the Hubble Constant of 85 km s-1 Mpc-1.

The 21-cm line width is a good indication of the maximum rotational velocity in a galaxy. However, the turbulent motions affect the observed line width and must be corrected. Not taking them into account introduces both accidental and systematic errors.

A first example of this effect is the dependence of the slope a of the Tully-Fisher relation:

on the level at which the line width ΔVobs. is measured. In their 1977 paper, Tully and Fisher find the following values:

A second example is the departure from linearity in the H-band Tully-Fisher relation obtained by Aaronson et al. (1982) (their figure 2): the narrow lines are more strongly affected by these turbulent motions.

Studies are now in progress on the nature of various possible systematic and random errors which may be influencing the values of the Hubble parameter Ho derived from application of the Tully-Fisher method to samples of cluster spirals outside of the Local Supercluster. Three effects seem to be of most importance. (1) Clusters yield slopes in the infrared Tully-Fisher diagram varying from 8 to 12, making it problematic as to how to derive a relative distance modulus from comparison with the local calibrators, (2) errors in measured 21-cm line widths (often measured at low signal-to-noise ratios) are the dominant source of error in derived relative distances, (3) errors in measured optical major and minor axes of a galaxy influence both its derived inclination and the H-magnitude as corrected to a standard isophote. Monte Carlo simulations of cluster samples, however, have shown that the tendency not to detect H I from edge-on and/or low-luminosity galaxies introduces no important biases. Overall, the relative distance of any of these clusters to the local calibrators appears to be good to ±20%.

Data from 60 field spiral galaxies shows that the Tully-Fisher (TF) relationships, MB vs. log(Vmax) and log(Dkpc) vs. log(Vmax), are both Hubble type dependent and that the slope in the MB–log(Vmax) correlation is steep, ~10.

Our study, conducted in collaboration with W.K. Ford, Jr. and D. Burstein, is principally aimed at elucidating the intrinsic range of properties of spiral galaxies. Hence, we picked a sample of 21 Sc’s, 23 Sb’s and 16 Sa’s with the widest range in luminosity and radius we could find at each Hubble type. Type classifications come principally from Sandage and Tammann (RSA); a few from de Vaucouleurs et al. (RC2) and Nilson (UGC). We stress that this is not a volume limited sample; the end members in the luminosity or radius range of the galaxies at each Hubble type are quite rare. Vmax comes from the maximum observed optical rotation velocity. We also have 21-cm profiles for all Sc’s, 2/3 of the Sb’s and 1/2 of the Sa’s; the result remains the same whether Vmax comes from 21-cm profiles or optical rotation curves. But note, ~1/3 of the Sa galaxies observed were undetectable at 21-cm. This must introduce a bias in the infrared 21-cm TF results. Magnitude corrections come from Heiles and Burstein (1978), but corrections from the RC2 give the same result. We currently have data for 8 galaxies in the Pegasus I and Cancer clusters, which are also consistent with the results shown below.

We have devised a procedure for evaluating the absolute magnitudes of galaxies from their optical rotation curves, as an extention of the conventional Fisher-Tully method. We describe here how this method can be employed to evaluate the Hubble constant. From observations of 23 Sb field galaxies with luminosities ranging from –19.5 to –23.0 (adopting H=50 km s–1 Mpc–1), we have produced synthetic rotation curves showing the systematic progression toward increasing velocity with increasing luminosity within a given Hubble type. (See Thonnard and Rubin, Carnegie Yrbk 80, p. 551 for details of producing such a set of curves). By matching even a small portion of a rotation curve with these curves, the absolute magnitude of an Sb galaxy can be estimated to about ±0.5 mag. This magnitude, of course, is based on an assumed value for H.

The problem of the determination of distances in astronomy (the so-called problem of the distance scale) is a very old and important problem. About 280 BC Aristarchus of Samos, the famous greek astronomer of the Alexandrian school, already devised a method to find the relative distances to the Sun and Moon in terms of the size of the Earth. Later, Eratosthenes (about 200 BC), another greek astronomer, measured the Earth’s diameter; so the zero-point of this first distance scale was fixed.

Now we are interested by extragalactic distances but the same approach is made : (i) determination of relative distances (ii) determination of the zero-point to obtain absolute distances.

Some distance criteria can be used to determine the zero-point. These criteria cannot generally be used at a great distance. They permit a comparison between galactic objects, like Cepheids, Novae, Supergiants ..., globular clusters..., and the same counterparts recognized in external galaxies. Often application is limited to nearby galaxies. In a first section we will briefly present this kind of distance criteria. For more distant galaxies other criteria must be employed, the zero-point being here fixed with nearby galaxies whose distances are known from the preceding step. We will discuss these criteria in a second section.

A brief discussion of the systematic effects of selection bias on the extragalactic distance scale is given. Distance indicators with intrinsic scatter yield only upper limits to Ho, unless the true intrinsic scatter is either small or well determined. Several distance indicators (luminosity index, diameters, globular clusters, 21 cm line widths) are discussed. It is concluded that type I supernovae, calibrated through brightest M supergiants, yield presently the most reliable determination of the large-scale value of the Hubble constant, i.e. Ho(cosmic) = 50 ± 7 km s–1 Mpc–1.

The traditional approach to the extragalactic distance scale has been through several levels of primary, secondary and tertiary indicators of increasing range and decreasing precision. This piecemeal approach, fraught with the danger of cumulative errors, has led in recent years to two widely divergent scales, the “long” scale of Sandage and Tammann (1974–1982), leading to ≃ 50 km s-1 Mpc-1, and the “short” scale of the author (1978–79), leading to Ho ≃ 100 (de Vaucouleurs and Bollinger 1979; de Vaucouleurs and Peters 1981; de Vaucouleurs et al. 1981). The divergence, increasing from ~ 0.4 mag in the Local Group to ~ 1.2 mag at the distance of the Virgo Cluster, indicates that at least one of the scales is not linear (de Vaucouleurs 1981).

A brief summary is given of the current concepts of the icy conglomerate cometary nucleus and of the origin of comets. Evidence that the cores of comets may contain less than average volatile material, whether in formation or by radiative heating, raises the question of why at least two very faint short-period comets suddenly experienced violent outbursts (~4000 times in brightness). A preliminary study of close double comet nuclei as affected by differential nongravitational forces shows that a collision of a cometary satellite with its primary is a likely outcome. Thus double nuclei may possibly explain these rare but extreme outbursts. Statistics suggest, however that most comet splitting and comet outbursts represent intrinsic activities in extremely non-homogeneous nuclei.

Recent spectroscopic observations of comets, obtained especially in the vacuum ultraviolet spectral region from sounding rockets and satellites, have considerably increased our knowledge of the chemical composition of comets, and the physical and chemical processes occurring in their head and tail. From the UV spectra of comets observed so far, one can perhaps conclude that they have a very similar, homogeneous chemical composition, and possibly also a common origin.

Proton irradiation of a variety of ice mixtures were carried out to study the effect of galactic cosmic ray irradiation of comets in the Oort Cloud. Three significant effects were noted (1) production of new molecules; (2) production of a more energetic ice at low temperatures; (3) production of a non-volatile, complex organic residue. These phenomena suggest various effects on new comets approaching the sun including enhanced activity. Experiments on the condensation of silicate grains provide information on the condensation mechanism and properties of grains. Controlled annealing of the amorphous condensates shows how crystallization occurs. Infrared spectra of different stages of crystallization contain features that may identify composition, structure and history of refractory material.