Stellar winds have been revealed in a large fraction of central stars of planetary nebulae from P Cygni profiles observed with the IUE satellite. The relevant lines are essentially the resonance lines NV λ 1240, Si IV λ 1397, CIV λ 1549 and the subordinate lines OIV∗ λ 1342, 0V∗ λ 1371, NIV∗ λ 1579. Edge velocities are of the order of 1000-3000 km s−1, similar to the case of population I O stars. Detailed determinations of the mass loss rate have been performed for NGC 6543, NGC 2371, IC 2149 and IC 3568 with values between 4.10−9 to 7. 10−7 Mo yr−1. The accuracy of these determinations is not well known. It is however clear from the variety of observed profiles in these and in several other objects that properties of the winds (ionization structure, etc.) varies considerably from object to object and that very likely the mass loss rate will span over a large interval. Some possible consequences of these winds are discussed.

For the case of optically thin lines, we show that the relation existing between the first order moment of a P Cygni line profile and the quantity n (level) (cf. Castor, Lutz and Seaton, 1981), where is the mass-loss rate of the central star and n (level) the fractional abundance of the relevant ion, is in fact independent of any Sobolev-type approximations used for the transfer of line radiation. Consequently, all results established in the context of “very rapidly” expanding atmospheres (see Surdej, 1982) and mainly referring to the non-dependence of W1 on various physical (underlying photospheric absorption line, limb darkening, collisions, multiplet line transitions, etc.) and geometrical (radial and rotational velocity fields, size of the atmosphere, etc) effects remain unchanged for arbitrary (e.g. non-Sobolev type) outward-accelerating velocity laws.

The high excitation planetaries NGC 1535 and 4361 were observed with IUE satellite in the short wavelength region at high dispersion. In NGC 1535 we found P Cygni profiles of NV 1238, 1242 and OV 1371 lines with a terminal wind velocity of about 2000 km/sec. In NGC 4361 these lines are narrow absorption lines (width 0.5 Å), probably of photospheric origin.

This review will be concentrated on the determination of the main atmospheric parameters (Teff, log g, helium abundance) of PN nuclei, and of other subluminous objects, by fitting the observed absorption line profiles with theoretical profiles obtained from non-LTE model atmosphere calculations.

Considerable progress has been made in our understanding of the evolution of the central stars of planetary nebulae (NPN) compared to the situation five years ago at the Ithaca Symposium where Shaviv (1978) and Paczynski (1978) reviewed the subject. Shaviv stressed the necessity to start theoretical calculations with realistic initial models but doubted - in view of the loops in the HR diagram made by flashing stars - if the Harman-Seaton sequence could be taken as a single evolutionary sequence. Paczynski pointed out how strongly the theoretical rate of evolution depends on the stellar mass - a result which had appeared in his earlier calculations (1971) - and expected the existence of more flashing NPN's of the FG Sagittae type among the luminous (L > 104 L⊙) central stars, for which the core mass luminosity relation (Mc > 0.7 M⊙) combined with the core mass interpulse time relation predicts fairly short (2.10 yrs) intervals between flashing events. Weidemann, however, at the Symposium and shortly thereafter (1977a) concluded in view of the lower effective temperature derived by Pottasch et al. (1978) and the observed narrow mass distribution of white dwarfs around a 0.6 Mo. combined with the theoretical predicted horizontal tracks from the red giant branch towards the NPN region at a luminosity given by the core mass luminosity relation that the high luminosity part (and also the “upturn”) of the Harman-Seaton sequence does not exist. He also proposed an increase in the distances by an average factor of 1.3 compared to the Seaton/Webster (Seaton, 1968) or Cahn/Kaler (1971) scale in order to bring the observed NPN on the 0.6 M⊙ track in the HR diagram and to lower the NPN birth rates to a value compatible with white dwarf birth rates.

Absolute flux distributions in the ultraviolet continua of the central stars of three planetary nebulae of known distances have been derived from observations made with the International Ultraviolet Explorer. The observations confirm the existence of planetary nebulae nuclei with masses of ≈1M⊙ and indicate that the progenitors of the nebulae were carbon stars near the theoretical upper luminosity threshold of Mbol = −6.5. The derived masses, luminosities, and temperatures of the three stars indicate that they are currently on horizontal tracks in the HR diagram and probably have not yet attained their maximum luminosities. The present luminosities (≈ 4 × 104 L⊙) each are well above the Eddington luminosity for an 0.6 Mo star. The derived properties of the stars and associated nebulae (LMC P40, SMC N2, SMC N5) are consistent with a nebular ejection mechanism that involves radiation pressure on carbon grains.

Despite similar evolutionary histories and a common ultimate fate as white dwarfs, central stars of planetary nebulae have surprisingly diverse spectral properties. Their visual spectral types encompass all varieties known for hot stars, including Wolf-Rayet, O and Of, subdwarf O, white-dwarf, and continuous (Aller 1968, 1976), and O VI-emission types (Smith and Aller 1969, Heap 1982). Their spectroscopic temperatures range from less than 30,000°K (e.g. He 2-138, Mendez and Niemela 1979; the WC 11 stars, Houziaux and Heck 1982) to upwards to 150,000°K or more (e.g. NGC 246, Heap 1975; Abell 30, Greenstein 1981). Their atmospheres range from demonstrably helium- and carbon-rich (e.g. the WR stars, Barlow and Hummer 1982, Benvenutı et al. 1982) to apparently normal (e.g. the Of stars, Heap 1977a,b), to helium-poor (e.g. the nascent white dwarfs in Abell 7 and NGC 7293, where gravitational settling appears to have already taken effect, Mendez et al. 1981).

Determination of the distances to individual planetary nebulae are discussed. Especially those methods which are independent of assumed nebular properties (mass, absolute flux, etc.) are assembled and discussed. In this way, reasonable approximations to the distance can be obtained for about 50 planetary nebulae. The accuracy of the distances is tested by comparing nebular properties derived from these distances with the properties of nebulae at the galactic center or in the Magellanic clouds. A comparison is also made with the statistical distance determinations; the conclusion is that the assumption of constant mass often leads to an overestimate of the distance, while the assumption of constant Hβ flux leads to distances having individual uncertainties of up to a factor of 3.

The central star temperature determination is summarized. Individual central stars are placed on the HR diagram and compared with theoretical predictions. Deductions concerning the evolution which can be made from the observations are discussed.

The problem of determining the distance scale for planetary nebulae (PN) is approached through the kinematics of this subpopulation in the galaxy. To this end, we have compiled a catalogue of all known radial velocities for 457 galactic PN with standardized error statistics. External (calibration?) errors of the same magnitude as internal spectral line deviations are noted.

This review summarizes the techniques and results of a λ21cm and λ6cm search with the Westerbork telescope for planetary nebulae within 2° of the galactic center. After accounting for background sources and compact HII regions it appears that there are ~ 300 planetaries within 300 pc of Sgr A. This is consistent with a galactic population of ~ 21000 and agrees with the birthrate of white dwarfs. The surface density of galactic center planetaries falls off with galactic latitude approximately as b−1 and is best accounted for by a bulge with a mass of 1 X 1010 M⊙.

The Very Large Array (VLA) in Socorro, New Mexico, has been used to measure the 6 cm continuum flux densities and the angular sizes of 42 planetary nebulae (PN) in the direction of the galactic centre (GC). These were all optically confirmed PN for which the radial velocities (and positions on the sky) make it very likely that they are close to the GC. With a detection limit of about 1 mJy, 34 PN were detected. Their flux densities range from 2 to 100 mJy. Initially we used a configuration of the VLA with an instrumental resolution of 1″. About 80% of the detected PN could be clearly resolved with this resolution. The unresolved PN were observed again with a configuration of the VLA that has a resolution of 0″.4. For all but one of the 34 detected PN we could determine reliable angular sizes. The inferred total ionized masses range from < 0.01 to ~ 0.5 M⊙, assuming a distance to the GC of 9 kpc. The results argue strongly against the use of the Shklovsky method for distance determinations. Previous measurements of PN at the GC showed that their luminosities were substantially higher than those for nearby PN (Pottasch, 1980). The luminosity distribution of the PN in our sample is broader towards lower luminosities (up to the detection limit of the observations), but the luminosities are still high compared with nearby PN. We interpret this as a selection effect: by studying only optically confirmed PN, the intrinsically brightest PN are selected.

Recent observations of planetary nebulae have called into question the Shklovsky method of measuring distances. For those planetaries for which independent distance and electron density determinations are available, it is found that the ionized mass and the radius are linearly correlated (Maciel and Pottasch, 1980) and also that the ionized masses increase with decreasing electron density (Pottasch, 1981). These relations imply that the nebulae are optically thick in Ly continuum radiation and the distances based on the Shklovsky method are overestimates. Using an empirically determined mass-radius relationship Maciel and Pottasch have obtained new distances for the nebulae in the catalogue of Milne and Aller (1975). We have used the more complete catalogue of Cahn and Kaler (1971) to obtain distances corrected for possible variations in the ionized mass and have compiled a new list of local planetaries. We obtain a surface density of 15 ± 3 kpc−2 and a planar number density of 44 ± 4 kpc−3.

The identification and masses of Magellanic Cloud planetary nebulae are discussed. The masses are shown to be uncertain and should not be directly compared to values for galactic planetaries.

The kinematics suggest that the planetary nebulae belong to a younger rather than an older population. Abundance analyses show the Magellanic Cloud planetaries to be deficient in most elements, but the abundances of helium and carbon are comparable to values found for galactic planetaries.

Recent surveys for planetary nebulae have given the first identifications in Fornax, NGC 6822, M33, IC 10, Leo A, Sextans A, Pegasus, WLM, NGC 404, and M81, and extended the identifications in the SMC, the LMC, and M31. Observations of planetaries have established chemical compositions in old or intermediate age populations in 8 Local Group galaxies. The chemical compositions show that i) the helium abundance is higher in planetary nebulae than in H II regions in the same galaxy, and ii) nitrogen is overabundant relative to H II regions by factors of 4 to 100. Planetary nebulae are not a major source of helium in star-forming galaxies, and are a major source of nitrogen. The planetary in Fornax has a relatively high O abundance, and, together with Fornax's carbon stars, establishes the presence of at least 2 stellar populations. The abundance gradient derived from 3 planetaries in M31 is very shallow, and gives high abundances at ~ 20 kpc. By using planetary nebulae as standard candles, upper and lower distance limits have been set for 10 Local Group candidates, and a new distance estimated for M81.

We construct a parameterized function which describes the possible dependence of planetary nebulae formation upon metal abundance and stellar mass. Data on galaxies in the local group compared with predictions made from the parameterized function indicate that heavy element abundance is the principal agent influencing the formation of planetary nebulae; stars which are rich in heavy elements are the progenitors of planetary nebulae. Our analysis, when compared with the observations, argues for a modest degree of pre-enrichment in a few of the sample galaxies. The heavy element dependence of planetary nebulae formation also accounts for the deficit of planetary nebulae in the nuclei of NGC 221 and NGC 224, and in the bulge of our galaxy.

Tinsley (1978) has done an excellent review that illustrates the methods and concepts that can be developed to assess the effects of planetary nebulae (PN) on the long-term history of the galaxy. Tinsley concluded that research in PN could put constraints on the past rate of star formation and provide information on chemical enrichment by low mass stars.

Knowledge gained in the study of planetary nebulae has been, and can be further, transferred to understanding active galactic nuclei. Photoionization is the main energy-input mechanism in the narrow-line regions, and probably although by no means certainly in the broad-line regions as well. There are many detailed differences because of the much “harder” input spectrum in active galactic nuclei, compared with planetary nebulae. A tentative model of the structure of the gas distribution in a Seyfert-galaxy nucleus is presented.

What splendid laboratories have planetary nebulae been! We have heard of important observations in the ultraviolet, optical, infrared, and radio spectral ranges, all rich in astrophysical information. We are now certain that planetary nebulae represent a crucial phase of stellar evolution. We have also begun to study the fine details of the evolution of planetary nebulae and our endless aim is to understand these objects completely.

In order to detect faint nebulosities associated with planetary nebulae, long exposure plates are made on nine selected nebulae, using a large bandwidth (Δλ = 50 Å) interference filter coupled with an ITT image tube. Some peculiar features are observed, but they do not all account for “secondary structures” following Louise's terminology. We discuss the difficulties encountered by the photographic method.