In our Galaxy, it is possible to determine the absolute magnitude of a star by mean of the intensity of the hydrogen lines, if we know the spectral type of the star or its temperature (see for instance Petrie, 1965; Hutchings, 1966; Balona and Crampton, 1974; Crawford, 1973, 1978; …).

The bright stars of five LMC clusters were classified for deriving the distribution of various spectral types. The studied clusters are very young (NGC 2093) young (NGC 1818, NGC 2157) intermediate (NGC 1831) and old (NGC 1806) (Van den Bergh 1981). The spectral classification of the stars was carried out using film copies of the 1.2 m Schmidt telescope objective prism plates. Medium dispersion (830 Å at Hγ) unwidened YJ and widened UJ and low dispersion (2440 Å at Hγ) UJ were examined by means of a binocular microscope. Details of the criteria used for the classification are described by Kontizas et al (1985).

By comparison of U and B plates of M33 we extended the numerations of Humphreys and Sandage (1980) adding 54 new OB associations. The associations identified by HS are outlined with continuous line and ours are represented with dash on Figure 1. Many of the identified by us associations are near the nucleus of M33. Due to U plate used by us the background in the central region is fainter than that on the B plate and the associations are distinguished very well on U plate (Kunchev and Ivanov, 1984). Our associations of the outer region of M33 consist mainly of relatively fainter young B stars.

With classic models of stars, in which overshooting from convective cores is not taken into account, the transition mass, Mtr, separating stars which undergo core He-flash from stars which ignite He-burning nonviolently in a non degenerate core, is set around 2.2 M⊙ (Iben 1967) for Pop I chemical composition (X = 0.700, Z = 0.020). This value is known however to depend on the chemical abundances, being lower at increasing Y and decreasing Z (Wagner 1974; Sweigart and Gross 1978). However, in recent years many independent arguments have indicated that convective overshooting from the central cores may play an important role in stellar evoluion (Bertelli et al 1985). In order to assess the dependence of Mtr on overshooting, we have computed evolutionary sequences of 1.4 M⊙, 1.5 M⊙ and 1.6 M⊙ with chemical composition X = 0.700 and Z = 0.020 adopting the description of convective overshooting formulated by Bressan et al (1981) for their parameter λ = 1/Hp = 1. As shown in the Tc vs pc diagram of Fig 1, while the tracks of 1.4 M⊙ and 1.5 M⊙ stars deeply penetrate into the region of high degeneracy and likely undergo core He-flah, the 1.6 M⊙ star succeeds in igniting helium in non degenerate conditions. This means that the classic value of 2.2 M⊙ for Mtr is now lowered to the mass range 1.6 to 1.5 M⊙. The HR diagram of the 1.4 M⊙, 1.5 M⊙ and 1.6 M⊙ stars in presence of convective overshooting is shown in Fig 2. The region of stationary core He-burning for the 1.6 M⊙ and the extension of the red giant branches for the 1.4 M⊙ and 1.5 M⊙ stars are also indicated.

The cluster membership of Wolf-Rayet stars, supernova remnants and pulsars is used to study the late stages of stellar evolution of massive stars.

Observing luminosity functions and color-magnitude diagrams is a first step towards understanding the luminous stellar populations in galaxies. Dwarf irregular galaxies are of special interest for a variety of reasons. They are common at distances where they can be resolved, and generally have low background surface brightnesses which makes photometry of the resolved stars relatively easy. Dwarf irregulars exhibit a fairly wide spread in metal abundance allowing the study of massive star formation and evolution under a variety of conditions. Also, the dynamics of such systems is in general simple, without spiral density waves or other effects which might complicate understanding of the star formation process.

Observational constraints on the form of the high-mass stellar IMF are reviewed. The evidence includes star counts in the solar neighborhood, individual and composite star clusters, and nearby galaxies, and arguments based on integrated light and chemical evolution modeling. There is no convincing evidence for any systematic variations of the shape of the high-mass IMF. However, the various determinations are very uncertain, and do not allow any firm estimate of the logarithmic slope of the upper IMF; the appropriate value is somewhere between −1.3 and −2.3, with region-to-region variations smaller than about ±0.5. A number of lines of evidence suggest that the lower mass limit or mode mass of the IMF increases with increasing star formation rate, reaching perhaps 10–15 m⊙ in some starburst galaxies. It is also possible that the upper mass limit depends on metallicity, based on variations in excitation conditions of HII regions.

Some luminous stars undoubtedly explode as supernovae, but it is not yet certain that fate awaits them all. The connection between luminous stars and supernovae is reviewed in terms of the statistical rates of supernovae and pulsars, the constraints of nucleosynthesis, and the various classifications of supernovae by their spectra and light curves, including a newly confirmed class of peculiar Type I supernovae.

Lyman continuum (Lyc) photon production rates can be estimated from radio free-free emission and used to estimate the star formation rate (SFR) of 0 stars. If this SFR is linked to the total SFR through a constant IMF (m ≳0.1 m⊙) one derives for our Galaxy a present-day SFR of ∼10 m⊙ yr−1, which is close to the average SFR over the age of the galactic disk. This is difficult to reconcile with a formation law of the form SFR φ∝Mgas k with k>0 which yields SFRs which decrease with time. Even more severe is the fact that the mass distribution of the galactic disk cannot be reproduced by the present-day SFR with a constant IMF. Bimodal star formation, however, reduces the rate at which matter is permanently locked up in low mass and dead stars by nearly a factor of three, and gets reasonable agreement between the present-day distribution of stellar mass and lock-up rate. Bimodal star formation means that stars with m >0.1 m⊙ form in the interarm region while in spiral arms induced star formation produces only stars with m >mc ∼2–3 m⊙.

It is well known that some giant extragalactic star-forming regions contain WR stars. D'Odorico, Massey, Rosa and coworkers found many examples in nearby galaxies of giant HII regions whose spectra show that they contain WN, and occasionally, WC stars. The dwarf emission-line galaxies He 2–10 (Allen et al. 1976) and Tol 3 (Kunth & Sargent 1981) have a strong broad emission feature near HeII 4686Å; in the latter object ∼150 WN stars are required to explain the observed equivalent width.

Giant HII regions are important tracers of recent star formation in distant galaxies. For a selection of HII regions in our galaxy where the exciting stars can be identified, Rumstay (1985) finds that the measured Hα and radio continuum luminosities of an HII region correlate with the stellar ionizing flux derived from model atmospheres and the known exciting stars. Therefore, we use flux measurements of giant HII regions as an index of the distribution of O stars in M81.

A very small minority of all galaxies exhibit activity in their nuclear regions. The best known and most studied signature of nuclear activity is the presence of emission lines in the optical spectrum. In fact, active galaxies are classified as Seyferts, Liners or Starburst galaxies according to the strengths, widths and intensity ratios of their nuclear emission lines (figure 1).

If the integrated colours of a star cluster mainly depend on chemical composition and age, then theoretical calibrations of colours as function of age for different chemical compositions are very useful to obtain quantitative determinations of the age and composition of individual clusters, and thus to trace the chemical history of nearby galaxies. Several calibration curves exist in the literature which rest on the standard theory of stellar evolution. However, a growing amount of observational evidence seems to indicate that overshooting from convective cores may be an important phenomenon in stellar evolution. In fact models computed with overshooting are significantly different from the standard ones. The aim of this preliminary investigation is to study the effects of convective overshooting on the integrated colours of clusters whose turnoff mass is in that range in which convective overshooting is effective.

Deep CCD photometry of the star clusters NGC2162 and NGC2190 in LMC presented by Schommer et al. (1984) is used togheter with new evolutionary models computed by Bertelli et al. (1985a) which take into account overshooting from convective cores, to derive the clusters ages and the distance modulus of LMC. A preliminary analysis of the two clusters indicates that NGC 2162 and NGC 2190 belong to the same class of clusters discussed by Barbaro and Pigatto (1984). In fact, for the turn-off mass estimated by means of classical models (<2.2m⊙) these clusters should possess an extended red giant branch and a bimodal distribution of red stars (cifr. Fig.2). On the contrary they show a clump of red stars. This means that ages and other properties derived from classical models for this range of masses, may not correspond to reality. With the new models, stars of mass as low as 1.6 m⊙, ignite helium in non degenerate conditions, avoid the long lived RG phase, and burn helium as more massive stars. As consequence of it, a clump of red giants is expected. In Fig.1, we show new isochrones (Bertelli et al. 1985b) derived from models with overshooting, overlaid to the CM diagram of NGC 2162. Theoretical luminosities and Teff's are converted into Mv:(B-V)o plane by means of Teff:(B-V):BC scales based on models atmospheres collected from several authors (Chiosi, 1985). At any given age, the new isochrones run brighter than those of Ciardullo and Demarque (1977). By means of the luminosity function, a method more objective (Paczsynski, 1984) than the standard one of ZAMS and/or isochrone fitting, with a reddening of E(B-V)=0.06 and chemical composition X=0.700 and Z=0.02, we find ages of 1 109yr and a true distance modulus of (m-M)O=18.6 instead of 18.2±0.2 mag given by Schommer et al.(1984). Fig.2 shows the theoretical luminosity function at age 1 109yr, (age preliminarly assigned to the clusters by isochrone fitting) for main sequence and red giant stars obtained with Salpeter's IMF (top panel), compares it with the correspondent one of Ciardullo and Demarque(1977), and finally shows the observational LF we derive from stars counts(bottom panel) for NGC 2162. By imposing coincidence between theoretical and observational LF's at the side of main sequence fall-off and rising of the red giant clump, we derive the distance modulus (m-M)O=18.6. In conclusions, models with overshooting not only interpret the morphology of this class of clusters, but assigne LMC a distance modulus in agreement with other independent determinations (Walker, 1984; Visvanathan, 1985).

Giant H II regions are usually so distant that only the integrated properties of their stellar contents can be studied. However, the nearby giant H II regions 30 Doradus in the Large Magellanic Cloud and NGC 3603 in the Galaxy offer a rare opportunity for a spatially resolved study of their central stellar associations — NGC 2070 and NGC 3603. We have obtained 4m prime focus plates with interference filters centered on the He II 4686 line (Δλ = 65 A) and the adjacent blue continuum at 4765 A (Δλ = 70 A) to identify the Wolf-Rayet (WR) stars in their cores. We have also obtained CCD images in UBVRI for NGC 2070, and BVRI for NGC 3603 to determine their color-magnitude diagrams and luminosity functions. Since the WR stars in the core of 30 Dor have been reported by Chu, Cassinelli, and Wolfire (1984) and Moffat, Seggewiss, and Shara (1985), and the CCD photometry of NGC 2070 and 3603 is still preliminary, we will discuss in these proceedings only the WR stars resolved in the core of NGC 3603.

Highlights of the IAU Symposium 116 are reviewed. Some of the general themes running through the meeting are identified. These include:i) the fruitful interaction between observation, laboratory work and theory. ii) the need for understanding and, if possible, correcting for the effects of incompleteness and bias in observing lists. iii) the importance of the Magellanic Clouds, as the nearest independently evolving stellar systems, in the study of massive star formation and evolution in galaxies.